CA2504106A1 - Related term suggestion for multi-sense query - Google Patents

Related term suggestion for multi-sense query Download PDF

Info

Publication number
CA2504106A1
CA2504106A1 CA002504106A CA2504106A CA2504106A1 CA 2504106 A1 CA2504106 A1 CA 2504106A1 CA 002504106 A CA002504106 A CA 002504106A CA 2504106 A CA2504106 A CA 2504106A CA 2504106 A1 CA2504106 A1 CA 2504106A1
Authority
CA
Canada
Prior art keywords
term
phrase
recited
clusters
computer
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CA002504106A
Other languages
French (fr)
Other versions
CA2504106C (en
Inventor
Benyu Zhang
Hua-Jun Zeng
Li Li
Tarek Najm
Wei-Ying Ma
Ying Li
Zheng Chen
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Microsoft Corp
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Publication of CA2504106A1 publication Critical patent/CA2504106A1/en
Application granted granted Critical
Publication of CA2504106C publication Critical patent/CA2504106C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9532Query formulation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B42BOOKBINDING; ALBUMS; FILES; SPECIAL PRINTED MATTER
    • B42DBOOKS; BOOK COVERS; LOOSE LEAVES; PRINTED MATTER CHARACTERISED BY IDENTIFICATION OR SECURITY FEATURES; PRINTED MATTER OF SPECIAL FORMAT OR STYLE NOT OTHERWISE PROVIDED FOR; DEVICES FOR USE THEREWITH AND NOT OTHERWISE PROVIDED FOR; MOVABLE-STRIP WRITING OR READING APPARATUS
    • B42D25/00Information-bearing cards or sheet-like structures characterised by identification or security features; Manufacture thereof
    • B42D25/20Information-bearing cards or sheet-like structures characterised by identification or security features; Manufacture thereof characterised by a particular use or purpose
    • B42D25/29Securities; Bank notes
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B42BOOKBINDING; ALBUMS; FILES; SPECIAL PRINTED MATTER
    • B42DBOOKS; BOOK COVERS; LOOSE LEAVES; PRINTED MATTER CHARACTERISED BY IDENTIFICATION OR SECURITY FEATURES; PRINTED MATTER OF SPECIAL FORMAT OR STYLE NOT OTHERWISE PROVIDED FOR; DEVICES FOR USE THEREWITH AND NOT OTHERWISE PROVIDED FOR; MOVABLE-STRIP WRITING OR READING APPARATUS
    • B42D15/00Printed matter of special format or style not otherwise provided for
    • B42D15/0053Forms specially designed for commercial use, e.g. bills, receipts, offer or order sheets, coupons
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B42BOOKBINDING; ALBUMS; FILES; SPECIAL PRINTED MATTER
    • B42DBOOKS; BOOK COVERS; LOOSE LEAVES; PRINTED MATTER CHARACTERISED BY IDENTIFICATION OR SECURITY FEATURES; PRINTED MATTER OF SPECIAL FORMAT OR STYLE NOT OTHERWISE PROVIDED FOR; DEVICES FOR USE THEREWITH AND NOT OTHERWISE PROVIDED FOR; MOVABLE-STRIP WRITING OR READING APPARATUS
    • B42D15/00Printed matter of special format or style not otherwise provided for
    • B42D15/0073Printed matter of special format or style not otherwise provided for characterised by shape or material of the sheets
    • B42D15/0086Sheets combined with other articles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3322Query formulation using system suggestions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9538Presentation of query results
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • Y10S707/99934Query formulation, input preparation, or translation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • Y10S707/99935Query augmenting and refining, e.g. inexact access
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • Y10S707/99936Pattern matching access
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99937Sorting

Abstract

Systems and methods for related term suggestion are described. In one aspect, term clusters are generated as a function of calculated similarity of term vectors. Each term vector having been generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine. Responsive to receiving a term/phrase from an entity, the term/phrase is evaluated in view of terms/phrases in the term clusters to identify one or more related term suggestions.

Description

Related Term Suggestion for Multi-Sense Query RELATED APPLICATIONS
[0001] This patent application is related to the following patent applications, each of which are commonly assigned to assignee of this application, and hereby incorporated by reference:
~ U.S. Patent Application no. 10/427,548, titled "Object Clustering Using Inter-Layer Links", filed on 05/01/2003; and ~ U.S. Patent Application no. 10/826,159, titled "Reinforced Clustering of Multi-Type Data Objects for Search Term Suggestion", filed on 04/15/04.
TECIiNICAL FIELD
(0002] Systems and methods of the invention pertain to data mining.
BACKGROUND
[0003] A keyword or phrase is a word or set of terms submitted by a Web surfer to a search engine when searching for a related Web page/site on the World Wide Web (WWW). Search engines determine the relevancy of a Web site based on the keywords and keyword phrases that appear on the page/site. Since a significant percentage of Web site traffic results from use of search engines, Web site promoters know that proper keyword/phrase selection is vital to increasing site traffic to obtain desired site exposure. Techniques to identify keywords relevant to a Web site for search engine result optimization include, for example, evaluation by a human being of Web site content and purpose to identify relevant keyword(s). This evaluation may include the use of a keyword popularity tool.

Such tools determine how many people submitted a particular keyword or phrase including the keyword to a search engine. Keywords relevant to the Web site and determined to be used more often in generating search queries are generally selected for search engine result optimization with respect to the Web site.
[0004] After identifying a set of keywords for search engine result optimization of the Web site, a promoter may desire to advance a Web site to a higher position in the search engine's results (as compared to displayed positions of other Web site search engine results). To this end, the promoter bids on the keywords) to indicate how much the promoter will pay each time a Web surfer clicks on the promoter's listings associated with the keyword(s). In other words, keyword bids are pay-per-click bids. The larger the amount of the keyword bid as compared to other bids for the same keyword, the higher (more prominently with respect to significance) the search engine will display the associated Web site in search results based on the keyword.
[0005] In view of the above, systems and methods to better identify keywords relevant to Web site content would be welcomed by Web site promoters. This would allow the promoters to bid user preferred terms.
Ideally, these systems and methods would be independent of the need for a human being to evaluate Web site content to identify relevant keywords for search engine optimization and keyword bidding.
SUMMARY
[0006] Systems and methods for related term suggestion are described. In one aspect, term clusters are generated as a function of calculated similarity of term vectors. Each term vector having been generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine. Responsive to receiving a term/phrase from an entity, the term/phrase is evaluated in view of terms/phrases in the term clusters to identify one or more related term suggestions.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] In the figures, the left-most digit of a component reference number identifies the particular figure in which the component first appears.
[0008] Fig. 1 illustrates an exemplary system for related term suggestion for mufti-sense query.
[0009] Fig. 2 illustrates an exemplary procedure for related term suggestion for mufti-sense query.
[0010] Fig. 3 illustrates an exemplary procedure for related term suggestion for mufti-sense query. The operations of this procedure are based on the operations of Fig. 2.
[0011] Fig. 4 shows an exemplary suitable computing environment on which the subsequently described systems, apparatuses and methods for related term suggestion for mufti-sense query may be fully or partially implemented.
DETAILED DESCRIPTION
Overview [0012] It may appear that the simplest way to suggest related term/phrase is to use a substring matching approach, which judges two terms/phrases as related when one term/phrase includes some or all of the words of another term/phrase.
However, this technique is substantially limited. This method may neglect many semantically related terms, because related terms need not to contain common words. For example, consider that a footwear corporation wants to know related terms for "shoe". If the conventional matching approach is used, only "women's shoes", "discount shoes", etc. will be suggested. However, many other related terms exist such as "sneakers", "hiking boots", "Nike", etc.
[0013] The following systems and methods for related term suggestion for a multi-sense query address these limitations of conventional substring matching techniques. To this end, the systems and methods mine search engine results for terms/phrases that are semantically related to an end-user (e.g., Web site promoter, advertiser, etc.) submitted terms/phrases. The semantic relationship is constructed by mining context (e.g., text, andlor so on) from the search engine results, the context surrounding a term/phrase that can cast light on term/phrase meaning.
More particularly, a set of query terms is aggregated from a historical query log, with their frequency of occurrence (FOO) being counted. These query terms are submitted, one-by-one, to the search engine. In one implementation, the submitted the historical query log terms have a relatively high frequency of occurrence as compared to the frequency of occurrence of other historical query log terms.
[0014] Responsive to receiving respective ones of the submitted queries, the search engine returns a ranked list of search results, including URLs, result titles, and short descriptions of each result and/or context surrounding the submitted query. As search engine results are received, the systems and methods extract a set of features (keywords and the corresponding weights that are calculated using known TFIDF techniques) from select ones of the returned search results (e.g., one or more top-ranked results). After extracting features of corresponding search engine results from the submitted search queries, the extracted features are normalized. The normalized features are used to represent each submitted queries, and are used in a text clustering algorithm to group submitted query terms into clusters.
[0015] Responsive to receiving the term/phrase from the end-user, the term/phrase is compared to respective ones of the terms/phrases in the term clusters. Since the term clusters include terms that are contextually related to one another, when the term/phrase is compared to the terms within the clusters, the term phrase is evaluated in view of any multiple related contexts, or "senses." In one implementation, if a term/phrase matches a term from a cluster, the cluster is returned to the end-user in a suggested term list. The suggested term list includes terms/phrases determined to be semantically and/or contextually related to the term/phrase, respective term/phrase to term/phrase similarity measurements (confidence values), and respective term/phrase frequency of occurrence (FOO).
The returned list is ordered by a combination of FOO and confidence value. If the term/phrase matches terms in more than a single term cluster, multiple suggested term lists are generated. The lists are ordered by the cluster sizes; and the terms within each list are ordered by a combination of FOO and confidence value. If no matching clusters are identified, the query term is further matched against expanded clusters generated from query terms with low FOO.
[0016] In one implementation, query terms with low FOO are clustered by training a classifier (e.g., a K-nearest neighbor classifier) for the term clusters generated from the high frequency of occurrence historical query log terms.
Historical query terms determined to have low frequency of occurrence are submitted, one-by-one, to the search engine. Features are then extracted from select ones (e.g., a first top-ranked Web page, and/or so on) of the returned search results. The extracted features are normalized and used to represent the query terms with low FOO. The query terms are then classified into existing clusters to generate expanded clusters based on the trained classifier. The end-user submitted term/phrase is then evaluated in view of these expanded clusters to identify and return a suggested term list to the end-user.
(0017] These and other aspects of the systems and methods for related term/keyword suggestion for a mufti-sense query are now described in greater detail.
An Exemnlary System [0018] Turning to the drawings, wherein like reference numerals refer to like elements, the systems and methods for related term suggestion for mufti-sense query are described and shown as being implemented in a suitable computing environment. Although not required, the invention is described in the general context of computer-executable instructions (program modules) being executed by a personal computer. Program modules generally include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract data types. While the systems and methods are described in the foregoing context, acts and operations described hereinafter may also be implemented in hardware.
[0019] Fig. 1 shows an exemplary system 100 for related term suggestion for mufti-sense query. In this implementation, system 100 includes editorial verification server (EVS) 102 coupled across a network 104 to client computing device 106. Responsive to receiving a term/phrase 108, for example from client computing device 106 or another application (not shown) executing on EVS 102, EVS 102 generates and communicates suggested term list 110 to the client computing device 106 to allow an end-user to evaluate a set of terms semantically/contextually related to the term/phrase 108 prior to actually bidding on the term/phrase. Network 104 may include any combination of a local area network (LAN) and general wide area network (WAN) communication environments, such as those which are commonplace in offices, enterprise-wide computer networks, intranets, and the Internet. When system 100 includes a client computing device 106, the client computing device is any type of computing device such as a personal computer, a laptop, a server, a mobile computing device (e.g., a cellular phone, personal digital assistant, or handheld computer), etc.
[0020] A suggested term list 110 includes, for example, terms / phrases determined to be related to the term/phrase 108, respective term/phrase to term/phrase 108 similarity measurements (confidence values), and respective term/phrase frequency of occurrence (FOO) - frequency in the historical query log. Techniques for identifying related terms/phrases, generating similarity measurements, and generating FOO values are described in greater detail below in reference to sections titled keyword mining, feature extraction, and term clustering.
[0021] TABLE 1 shows an exemplary suggested term list 110 of terms determined to be related to a term/phrase 108 of "mail." Terms related to term/phrase 108 are shown in this example in column 1, titled "Suggested Term."

AN EXEMPLARY SUGGESTED TERM LIST FOR
THE BID TERM "MAIL"
Su ested Term Similari re uenc <Context >

otmail 0.246142 93161 online e-mai ahoo 0.0719463 165722 elated ail.com .352664 1455 ahoo mail .0720606 39376 ww.mail.com 0.35367 711 mail.com .484197 25 .hot .186565 1579 ww.msn.com .189117 1069 ail. ahoo.com .0968248 481 ee email .130611 1189 ww.aolmail.com .150844 54 heck mail .221989 6 heck email .184565 59 sn ass ort 0.12222 55 ww.webmail.aol.com 0.0800538 108 ebmail. ahoo.com .08789 1 free email account .0836481 5 Su ested term Similarit re uenc ail 1 191 raditional mai s s .205141 316 elated s s.com .173754 779 nited arcel service.120837 41 ostal rates .250423 6 stam s .156702 02 tam collectin .143618 152 state abbreviations.104614 00 ostal .185255 6 osta a .180112 55 osta a rates .172722 51 s s zi codes .138821 8 s ostmaster .109844 58 [0022] Referring to TABLE 1, note that terms in the suggested term list are mapped to term similarity values (see, column 2, titled "Similarity") and frequency of occurrence scores (see, column 3, titled "Frequency"). Each term similarity value, calculated as described below in the section titled "Term Clustering", provides a similarity measure between a corresponding suggested term (column 1 ) and the term/phrase 108, which is "mail" in this example.
Each frequency value, or score, indicates the number of times that the suggested term occurs in the historical query log. The suggested term list is sorted as a function of term similarity, and/or frequency of occurence scores as a function of business goals.
[0023] Any given term/phrase 108 (e.g., mail, etc.) may have more than a single context within which the bid term may be used. To account for this, STS
model 112 provides an indication in suggested term list 110 of which suggested terms correspond to which of the multiple contexts of term/phrase 108. For example, referring to TABLE 1, the term/phrase 108 of "mail" has two (2) contexts: ( 1 ) traditional off line mail and (2) online e-mail. Note that a respective list of related terms is shown for each of these two bid term contexts.
[0024] Additionally, suggested terms for any term/phrase 108 may be more than synonyms of the bid term. For instance, referring to TABLE 1, the suggested term "usps" is an acronym for an organization that handles mail, not a synonym for the bid term "mail." However, "usps" is also a term very related to a "mail"
bid term, and thus, is shown in the suggested term list 110. In one implementation, STS model 112 determines the relationship between a related term R (e.g. "usps") and a target term T (e.g. "mail") as a function of the following association rule: itr(T) --> itr(R), wherein "itr" represents "interested in".
If a user (advertiser, Web site promoter, and/or the like) is interested in R, the user will also be interested in T.
[0025] EVS 102 includes a number of computer-program modules to generate suggested term list 110. The computer-program modules include, for example, search term suggestion (STS) module 112 and classification module 114.
STS module 112 retrieves a set of historical queries 116 from query log 118.
The historical queries include search query terms previously submitted to a search engine. STS module 112 evaluates historical queries 116 as a function of frequency of occurrence to identify high frequency of occurrence (FOO) search terms 120 and relatively lower frequency of occurrence search terms 122. In this implementation, a configurable threshold value is used to determine whether a historical query has a relatively higher or low frequency of occurrence. For example, search query terms in historical queries 116 that occur at least a threshold number of times are said to have a high frequency of occurrence.
Analogously, search query terms in historical queries 116 that occur less than the threshold number of time are said to have a low frequency of occurrence. For purposes of illustration, such a threshold value is shown as a respective portion of "other data" 124.
Keyword Mining and Feature Extraction (0026] STS module 112 mines semantic/contextual meaning high frequency of occurrence query terms 120 by submitting each query, one-by-one (search query 128), to search engine 126. Responsive to receiving search query 128, search engine 126, returns a ranked listing (whose number is configurable) in search result 130 to STS module 112. The ranked listing includes URLs, result titles, and short descriptions and/or contexts of query term related to the submitted search query 128. The ranked listing is stored in the search results 132. Such search result retrieval is done for each search query 128.
[0027] STS module 112 parses Web page Hypertext Markup Language (HTML) to extract the URLs, result titles and short descriptions and/or contexts of the query term for each query term 120 from each retrieved search result 132.
The URLs, result titles, short descriptions and/or contexts of the query term, and the search query 128 used to obtain the retrieved Search result 132 are stored by STS
module 112 in a respective record of extracted features 134.
[0028] After parsing search results 130 for the high frequency of occurrence query terms 120, STS module 112 performs text preprocessing operations on extracted features 134 to generate linguistic tokens (tokenize) from the extracted features into individual keywords. To reduce dimensionality of the tokens, STS
module 112 removes any stop-words (e.g., "the", "a", "is", etc.) and removes common suffixes to normalize the keywords, for example, using a known Porter stemming algorithm. STS module 112 arranges the resulting extracted features 134 into one or more term vectors 136.
[0029] Each term vector 136 has dimensions based on term frequency and inverted document frequency (TFIDF) scores. A weight for the i~' vector's j~' keyword is calculated as follows:
w;~ =TF,~xlog(NIDF~) wherein TF,~ represents term frequency (the number of occurrences of keyword j in the i~' record), N is the total number of query terms, and DF~ is the number of records that contain keyword j.

Term Clustering [0030] STS module 112 groups similar terms to generate term clusters 138 from term vectors 136. To this end, and in this implementation, given the vector representation of each term, a cosine function is used to measure the similarity between a pair of terms (recall that the vectors were normalized):
d sim(q~ ~ qk ~ - ~ w~ ~ wk i=1 Thus, the distance between the two terms (a similarity measurement) is defined as:
dist(q~, qk)=1-sim(q~, qk) Such similarity measurements are shown as a respective portion of "other data" 124. Exemplary such similarity values are shown in an exemplary suggested term list 110 of TABLE 1.
[0031] STS module 112 uses the calculated similarity measurements) to cluster/group terms represented by keyword vectors 134 into term clusters) 138.
More particularly, and in this implementation, STS module 112 uses a known density-based clustering algorithm (DBSCAN) to generate term clusters) 138.
DBSCAN uses two parameters: Eps and MinPts. Eps represents a maximum distance between points in a cluster 138. Here points are equivalent of vectors because each vector can be represented by the point of the vector's head when its tail is moved to the origin. MinPts represents a minimum number of points in a cluster 138. To generate a cluster 138, DBSCAN starts with an arbitrary point p and retrieves all points density-reachable from p with respect to Eps and MinPts. If p is a core point, this procedure yields a cluster 138 with respect to Eps and MinPts. If p is a border point, no points are density-reachable from p and DBSCAN visits the next point.
Term Matching [0032] Responsive to receiving the term/phrase 108 from an end-user (e.g., (e.g., an advertiser, Web site promoter, etc), STS module 112 compares the term/phrase 108 to respective ones of the terms/phrases in the term clusters 138.
Since term clusters 138 include terms that are contextually related to one another, the term/phrase 108 is evaluated in view of multiple related and historical contexts, or "senses." In one implementation, if STS module 112 determines that a term/phrase 108 matches a term/phrase from a cluster 138, search term suggestion module 112 generates suggested term list 110 from the cluster 138.
In this implementation, a match may be an exact match or a match with a small number of variations such as singular/plural forms, misspellings, punctuation marks, etc. The returned list is ordered by a combination of FOO and confidence value.
[0033] If STS module 112 determines that a term/phrase 108 matches terms in multiple term clusters 138, search term suggestion module 112 generates multiple suggested term lists 110 from terms in the multiple ones of term clusters 138. The lists are ordered by the cluster sizes; and the terms within each list are ordered by a combination of FOO and confidence value.
Classification of Low FOO Terms [0034] Classification module 114 generates suggested term list 110 when term clusters 138 generated from high frequency of occurrence (FOO) query terms 120 do not include same terms to end-user input term/phrase 108. To this end, classification module 114 generates trained classifier 140 from term clusters generated from high frequency of occurrence (FOO) query log terms 120. The terms in term clusters 138 already have corresponding keyword vectors in a vector space model suitable for classification operations. Additionally, stop-word removal and word stemming (suffix removal) reduced dimensionality of term vectors 136 (upon which clusters 138 are based). In one implementation, additional dimensionality reduction techniques, for example, feature selection or re-parameterization, may be employed.
[0035] In this implementation, to classify a class-unknown query term 120, classification module 114 uses the k-Nearest Neighbor classifier algorithm to find k most similar neighbors in all class-known query terms 120, relying on their corresponding feature vectors, and uses the a weighted majority of class labels of the neighbors to predict the class of the new query term. Here each query term already in term clusters 138 is assigned a label same to their corresponding clusters' label, while each cluster 138 is labeled by simple sequence numbers.
These neighbors are weighted using the similarity of each neighbor to X, where similarity is measured by Euclidean distance or the cosine value between two vectors. The cosine similarity is as follows:
xl'dl>
t; E(x(1D~ ) sim(X,DJ ) - IIX 12 ~ II D~ II2 where X is the test term, i.e. the query term to be classified, represented as a vector; D~ is the jth training term; t; is a word shared by X and D~; xi is the weight of keyword t; in X; d,~ is the weight of keyword t~ in D~; IIX~Iz - x; +x2 +x3 is the norm of X, and IID; IIZ is the norm of D~. Thus, the class label of the test term X is the weighted majority of all neighbors' class labels:
label(X) = argmax( ~ sim(X,DI)) 1; All D~ where lable(D; )=1, [0036] In another implementation, a different statistical classification and machine learning technique (e.g., including regression models, Bayesian classifiers, decision trees, neural networks, and support vector machines) other than a nearest-neighbor classification technique is used to generate trained classifier 140.
[0037] Classification module 114 submits low frequency of occurrence (FOO) query terms 122, one-by-one (via a respective search query 128), to search engine 126. Responsive to receiving search result 130 associated with a particular search query 128, and using techniques already described, classification module 114 extracts features (extracted features 134) from one or more retrieved search results 132 identified by the search result 130. In this implementation, features are extracted from a first top-ranked Search result 132. For each retrieved and parsed Search result 132, classification module 114 stores the following information in a respective record of extracted features 134: the URLs, result titles, short descriptions and/or contexts of the query term, and search query used to obtain the retrieved Search result 132. Next, classification module tokenizes, reduces dimensionality, and normalizes extracted features 134 derived from low FOO query terms 122 to generate term vectors 136. Then, classification 114 clusters the query terms into a respective set of clusters 138. This clustering operation is performed using trained classifier 140 (generated from high FOO
query terms 120).
[0038) Classification 114 evaluates end-user submitted term/phrase 108 in view of these expanded term clusters (generated based on low FOO query terms 122) to identify and return one or more suggested term lists 110 to the end-user.
An exemplary such procedure is described above in paragraphs [0032] and [0033], and in the following section.
An Exemnlary Procedure [0039) Fig. 2 illustrates an exemplary procedure 200 for related term suggestion for mufti-sense query. For purposes of discussion, operations of the procedure are discussed in relation to the components of Fig. 1. (All reference numbers begin with the number of the drawing in which the component is first introduced). At block 202, search term suggestion (STS) module 112 (Fig. 1) collects historic query terms 116 from query log 120. STS module 112 organizes the historical query is 116 as a function of frequency of occurrence. At block 204, STS module 112 sends high frequency of occurrence query terms 120 to search engine 132 and receives corresponding search results 130. At block 206, STS
module 112 extracts snippet descriptions from each search result 130 and merges the snippet descriptions (extracted features 134) together to form term vectors 136.
A respective term vector 136 is generated for each respective high frequency of occurrence query term 120.
[0040) At block 208, STS module 112 applies a clustering algorithm to group substantially similar terms based on term vectors 136 into term clusters 138.
At block 210, responsive to receiving a termlphrase 108 from an end-user, STS
module 112 generates a suggested term list 110 from any keywords/key phrases from the term clusters 138 determined to be substantially similar to the term/phrase 108. At block 212, STS module 112 determines whether any keywords/phrases from keyword clusters 138 were determined to be substantially similar to term/phrase 108. If so, the procedure continues at block 214, where in STS module 112 sends the suggested term list 110 to the end-user. Otherwise, the procedure continues at block 302 of Fig. 3 as shown by on-page reference "A".
[0041] Fig. 3 illustrates an exemplary procedure 300 for related term suggestion for mufti-sense query. The operations of procedure 300 are based on the operations of procedure 200 of Fig. 2. For purposes of discussion, operations of the procedure are discussed in relation to the components of Fig. 1. (All reference numbers begin with the number of the drawing in which the component is first introduced). At block 302, STS module 112 generates a classifier 140 from term clusters 138, which at this time are based on high frequency of occurrence prairie terms 120. At block 304, STS module 112 sends low frequency of occurrence query terms 122, one by one, to search engine 132 and receives corresponding search results 130. At block 306, STS module 112 extracts snippet descriptions (extracted features 134) from the search results 130, and generates term vectors 136 therefrom. At block 308, STS module 112 classifies term vectors 136 generated from low frequency of occurrence quey terms 122 in view of the trained classifier 140 to generate respective term clusters 138 based on the low frequency of occurrence query terms.
[0042] At block 310, STS module 112 generates a suggested term list 110 from the keywords/key phrases from term clusters 138 based on the low frequency of occurrence prairie terms 122 that are determined to be substantially similar to the term/phrase 108. At block 312, STS module 112 sends the suggested term list 110 to the end-user.
An Exemplary Ouerating Environment [0043] Fig. 4 illustrates an example of a suitable computing environment 400 on which the system 100 of Fig. 1 and the methodology of Figs.
2 and 3 for related term suggestion for mufti-sense query may be fully or partially implemented. Exemplary computing environment 400 is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of systems and methods the described herein.
Neither should computing environment 400 be interpreted as having any dependency or requirement relating to any one or combination of components illustrated in computing environment 400.
[0044] The methods and systems described herein are operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that may be suitable for use include, but are not limited to, personal computers, server computers, multiprocessor systems, microprocessor-based systems, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and so on. Compact or subset versions of the framework may also be implemented in clients of limited resources, such as handheld computers, or other computing devices. The invention is practiced in a distributed computing environment where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.
[0045] With reference to Fig. 4, an exemplary system for related term suggestion for mufti-sense query includes a general purpose computing device in the form of a computer 410. The following described aspects of computer 410 are exemplary implementations of client computing device PSS server 102 (Fig. 1) and/or client computing device 106. Components of computer 410 may include, but are not limited to, processing units) 420, a system memory 430, and a system bus 421 that couples various system components including the system memory to the processing unit 420. The system bus 421 may be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures. By way of example and not limitation, such architectures may include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus also known as Mezzanine bus.
[0046] A computer 410 typically includes a variety of computer-readable media. Computer-readable media can be any available media that can be accessed by computer 410 and includes both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer-readable media may comprise computer storage media and communication media.
Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by computer 410.
[0047] Communication media typically embodies computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and includes any information delivery media. The term "modulated data signal" means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example and not limitation, communication media includes wired media such as a wired network or a direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media.
Combinations of the any of the above should also be included within the scope of computer-readable media.
[0048] System memory 430 includes computer storage media in the form of volatile and/or nonvolatile memory such as read only memory (ROM) 431 and random access memory (RAM) 432. A basic input/output system 433 (BIOS), containing the basic routines that help to transfer information between elements within computer 410, such as during staxt-up, is typically stored in ROM 431.
RAM 432 typically contains data and/or program modules that are immediately accessible to and/or presently being operated on by processing unit 420. By way of example and not limitation, Fig. 4 illustrates operating system 434, application programs 435, other program modules 436, and program data 437. In one implementation, wherein computer 410 is a PSS server 102. In this scenario, application programs 435 comprise search term suggestion model 112, and classification model 114. In this same scenario, program data 437 comprises term/phrase 108, suggested term list 110, historical queries 116, search query 128, search result 130, search results 132, extracted features 134, term vectors 136, keyword clusters 138, trained classifier 140, and other data 124.
[0049] The computer 410 may also include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only, Fig. 4 illustrates a hard disk drive 441 that reads from or writes to non-removable, nonvolatile magnetic media, a magnetic disk drive 451 that reads from or writes to a removable, nonvolatile magnetic disk 452, and an optical disk drive 455 that reads from or writes to a removable, nonvolatile optical disk 456 such as a CD
ROM or other optical media. Other removable/non-removable, volatile/nonvolatile computer storage media that can be used in the exemplary operating environment include, but are not limited to, magnetic tape cassettes, flash memory cards, digital versatile disks, digital video tape, solid state RAM, solid state ROM, and the like. The hard disk drive 441 is typically connected to the system bus 421 through a non-removable memory interface such as interface 440, and magnetic disk drive 451 and optical disk drive 455 are typically connected to the system bus 421 by a removable memory interface, such as interface 450.
[0050] The drives and their associated computer storage media discussed above and illustrated in Fig. 4, provide storage of computer-readable instructions, data structures, program modules and other data for the computer 410. In Fig.
4, for example, hard disk drive 441 is illustrated as storing operating system 444, application programs 445, other program modules 446, and program data 447.
Note that these components can either be the same as or different from operating system 434, application programs 435, other program modules 436, and program data 437. Operating system 444, application programs 445, other program modules 446, and program data 447 are given different numbers here to illustrate that they are at least different copies.
(0051] A user may enter commands and information into the computer 410 through input devices such as a keyboard 462 and pointing device 461, commonly referred to as a mouse, trackball or touch pad. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner, or the like.
These and other input devices are often connected to the processing unit 420 through a user input interface 460 that is coupled to the system bus 421, but may be connected by other interface and bus structures, such as a parallel port, game port or a universal serial bus (USB).
[0052] A monitor 491 or other type of display device is also connected to the system bus 421 via an interface, such as a video interface 490. In addition to the monitor, computers may also include other peripheral output devices such as speakers 497 and printer 496, which may be connected through an output peripheral interface 495.
[0053] The computer 410 operates in a networked environment using logical connections to one or more remote computers, such as a remote computer 480. The remote computer 480 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and as a function of its particular implementation, may include many or all of the elements described above relative to the computer 410, although only a memory storage device 481 has been illustrated in Fig. 4. The logical connections depicted in Fig. 4 include a local area network (LAN) 471 and a wide area network (WAN) 473, but may also include other networks. Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets and the Internet.
[0054] When used in a LAN networking environment, the computer 410 is connected to the LAN 471 through a network interface or adapter 470. When used in a WAN networking environment, the computer 410 typically includes a modem 472 or other means for establishing communications over the WAN 473, such as the Internet. The modem 472, which may be internal or external, may be connected to the system bus 421 via the user input interface 460, or other appropriate mechanism. In a networked environment, program modules depicted relative to the computer 410, or portions thereof, may be stored in the remote memory storage device. By way of example and not limitation, Fig. 4 illustrates remote application programs 485 as residing on memory device 481. The network connections shown are exemplary and other means of establishing a communications link between the computers may be used.

Conclusion [0055] Although the systems and methods for related term suggestion for multi-sense query have been described in language specific to structural features and/or methodological operations or actions, it is understood that the implementations defined in the appended claims are not necessarily limited to the specific features or actions described. Accordingly, the specific features and actions are disclosed as exemplary forms of implementing the claimed subject matter.

Claims (50)

1. A method for related term suggestion, the method comprising:
generating term clusters as a function of calculated similarity of term vectors, each term vector being generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine; and responsive to receiving a term/phrase from an entity, evaluating the term/phrase in view of terms/phrases in the term clusters to identify one or more related term suggestions.
2. A method as recited in claim 1, and wherein a multi-sense query comprises the term/phrase.
3. A method as recited in claim 1, and wherein the entity is a computer-program application and/or an end-user.
4. A method as recited in claim 1, further comprising determining the calculated similarity as follows:

wherein weight w for the i th vector's j th term is calculated as follows:
w ij = TF ij x log(N / DF j) ; and wherein TF ij represents term frequency, N is a total number of query terms, and DF j is a number of extracted feature records that contain term j.
5. A method as recited in claim 1, further comprising:
collecting historic query terms from a query log; and determining ones of the historic query terms with a high FOO.
6. A method as recited in claim 1, further comprising before creating the term clusters:
reducing dimensionality of the term vectors; and normalizing the term vectors.
7. A method as recited in claim 1, wherein evaluating further comprises:
identifying a match between the term/phrase and term(s)/phrase(s) from one or more term clusters; and responsive to identifying, generating related term suggestion(s) comprising the term(s)/phrase(s).
8. A method as recited in claim 7, wherein the related term suggestion(s) further comprise for each term/phrase of the term(s)/phrase(s), a frequency of occurrence value indicating a number of times the term/phrase occurs in a set of mined historical queries.
9. A method as recited in claim 1, wherein generating the term clusters further comprises:
sending respective ones of the high FOO historical queries to the search engine to obtain the search results;
extracting features from at least a subset of search results corresponding to the respective ones; and producing term vectors from the features as a function of term and inverted term frequencies.
10. A method as recited in claim 9, and wherein the features comprise a title, description, and/or context for the respective ones of the high FOO historical query terms.
11. A method as recited in claim 9, and wherein the respective ones comprise top ranked ones of the search results.
12. A method as recited in claim 1, wherein the term clusters are a first set of term clusters, and wherein the method further comprises:
determining that there is no match between the term/phrase and the terms/phrases; and responsive to the determining:
making a second set of term clusters from calculated similarity of term vectors, each term vector being generated from search results associated with a set of low FOO historical queries previously submitted to the search engine;
and evaluating the term/phrase in view of terms/phrases of the second set of term clusters to identify one or more related term suggestions.
13. A method as recited in claim 12, wherein making further comprises:
identifying the low FOO historical queries from historical queries mined from a query log;
sending respective ones of at least a subset of the low FOO historical queries to the search engine to obtain search results;
extracting features from at least a subset of search results; and producing the term vectors from the features as a function of term and inverted term frequencies.
14. A method as recited in claim 13, and further comprising after clustering:
determining that there is no match between the term/phrase and term(s)/phrase(s) from the first set of term clusters, the first set being based on high FOO historical queries; and responsive to the determining, identifying a match between the term/phrase and term(s)/phrase(s) from one or more of the second set of term clusters, the second set being based on low FOO historical queries; and responsive to identifying, generating related term suggestion(s) comprising the term(s)/phrase(s).
15. A computer-readable medium comprising computer-executable instructions for:
generating term clusters as a function of calculated similarity of term vectors, each term vector being generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine; and responsive to receiving a term/phrase from an entity, evaluating the term/phrase in view of terms/phrases in the term clusters to identify one or more related term suggestions.
16. A computer-readable medium as recited in claim 15, and wherein a multi-sense query comprises the term/phrase.
17. A computer-readable medium as recited in claim 15, and wherein the entity is a computer-program application and/or an end-user.
18. A computer-readable medium as recited in claim 15, further comprising computer-executable instructions for determining the calculated similarity as follows:

wherein weight w for the i th vector's j th term is calculated as follows:
w ij = TF ij x log(N / DF j); and wherein TF ij represents term frequency, N is a total number of query terms, and DF j is a number of extracted feature records that contain term j.
19. A computer-readable medium as recited in claim 15, further comprising computer-executable instructions for:
collecting historic query terms from a query log; and determining ones of the historic query terms with a high FOO.
20. A computer-readable medium as recited in claim 15, before creating the term clusters, further comprising computer-executable instructions for:
reducing dimensionality of the term vectors; and normalizing the term vectors.
21. A computer-readable medium as recited in claim 15, wherein evaluating further comprises computer-executable instructions for:
identifying a match between the term/phrase and term(s)/phrase(s) from one or more term clusters; and responsive to identifying, generating related term suggestion(s) comprising the term(s)/phrase(s).
22. A computer-readable medium as recited in claim 21, wherein the related term suggestions) further comprise for each term/phrase of the term(s)/phrase(s), a frequency of occurrence value indicating a number of times the term/phrase occurs in a set of mined historical queries.
23. A computer-readable medium as recited in claim 15, wherein generating the term clusters further comprises computer-executable instructions for:
sending respective ones of the high FOO historical queries to the search engine to obtain the search results;
extracting features from at least a subset of search results corresponding to the respective ones; and producing term vectors from the features as a function of term and inverted term frequencies.
24. A computer-readable medium as recited in claim 23, and wherein the features comprise a title, description, and/or context for the respective ones of the high FOO historical query terms.
25. A computer-readable medium as recited in claim 23, and wherein the respective ones comprise top ranked ones of the search results.
26. A computer-readable medium as recited in claim 15, wherein the term clusters are a first set of term clusters, and wherein the computer-executable instructions further comprise instructions for:
determining that there is no match between the term/phrase and the terms/phrases; and responsive to the determining:
making a second set of term clusters from calculated similarity of term vectors, each term vector being generated from search results associated with a set of low FOO historical queries previously submitted to the search engine;
and evaluating the term/phrase in view of terms/phrases of the second set of term clusters to identify one or more related term suggestions.
27. A computer-readable medium as recited in claim 26, wherein making further comprises computer-executable instructions for:
identifying the low FOO historical queries from historical queries mined from a query log;
sending respective ones of at least a subset of the low FOO historical queries to the search engine to obtain search results;
extracting features from at least a subset of search results; and producing the term vectors from the features as a function of term and inverted term frequencies
28. A computer-readable medium as recited in claim 27, and further comprising computer-executable instructions, after clustering, for:
determining that there is no match between the term/phrase and term(s)/phrase(s) from the first set of term clusters, the first set being based on high FOO historical queries; and responsive to the determining, identifying a match between the term/phrase and term(s)/phrase(s) from one or more of the second set of term clusters, the second set being based on low FOO historical queries; and responsive to identifying, generating related term suggestion(s) comprising the term(s)/phrase(s).
29. A computing device comprising:
a processor; and a memory coupled to the processor, the memory comprising computer-program instructions executable by the processor for:
generating term clusters as a function of calculated similarity of term vectors, each term vector being generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine; and responsive to receiving a term/phrase from an entity, evaluating the term/phrase in view of terms/phrases in the term clusters to identify one or more related term suggestions.
30. A computing device as recited in claim 29, and wherein a minti-sense query comprises the term/phrase.
31. A computing device as recited in claim 29, and wherein the entity is a computer-program application and/or an end-user.
32. A computing device as recited in claim 29, further comprising computer-executable instructions for determining the calculated similarity as follows:

wherein weight w for the i th vector's j th term is calculated as follows:

w ij= TF ij x log(N / DF j) ; and wherein TF ij represents term frequency, N is a total number of query terms, and DF j is a number of extracted feature records that contain term j.
33. A computing device as recited in claim 29, further comprising computer-executable instructions for:
collecting historic query terms from a query log; and determining ones of the historic query terms with a high FOO.
34. A computing device as recited in claim 29, before creating the term clusters, further comprising computer-executable instructions for:
reducing dimensionality of the term vectors; and normalizing the term vectors.
35 35. A computing device as recited in claim 29, wherein evaluating further comprises computer-executable instructions for:
identifying a match between the term/phrase and term(s)/phrase(s) from one or more term clusters; and responsive to identifying, generating related term suggestions) comprising the term(s)/phrase(s).
36. A computing device as recited in claim 35, wherein the related term suggestion(s) further comprise for each term/phrase of the term(s)/phrase(s), a frequency of occurrence value indicating a number of times the term/phrase occurs in a set of mined historical queries.
37. A computing device as recited in claim 29, wherein generating the term clusters further comprises computer-executable instructions for:
sending respective ones of the high FOO historical queries to the search engine to obtain the search results;
extracting features from at least a subset of search results corresponding to the respective ones; and producing term vectors from the features as a function of term and inverted term frequencies.
38. A computing device as recited in claim 37, and wherein the features comprise a title, description, and/or context for the respective ones of the high FOO historical query terms.
39. A computing device as recited in claim 37, and wherein the respective ones comprise top ranked ones of the search results.
40. A computing device as recited in claim 29, wherein the term clusters are a first set of term clusters, and wherein the computer-executable instructions further comprise instructions for:
determining that there is no match between the term/phrase and the terms/phrases; and responsive to the determining:
making a second set of term clusters from calculated similarity of term vectors, each term vector being generated from search results associated with a set of low FOO historical queries previously submitted to the search engine;
and evaluating the term/phrase in view of terms/phrases of the second set of term clusters to identify one or more related term suggestions.
41. A computing device as recited in claim 40, wherein making further comprises computer-executable instructions for:
identifying the low FOO historical queries from historical queries mined from a query log;
sending respective ones of at least a subset of the low FOO historical queries to the search engine to obtain search results;
extracting features from at least a subset of search results; and producing the term vectors from the features as a function of term and inverted term frequencies.
42. A computing device as recited in claim 41, and further comprising computer-executable instructions, after clustering, for:
determining that there is no match between the term/phrase and term(s)/phrase(s) from the first set of term clusters, the first set being based on high FOO historical queries; and responsive to the determining, identifying a match between the term/phrase and term(s)/phrase(s) from one or more of the second set of term clusters, the second set being based on low FOO historical queries; and responsive to identifying, generating related term suggestions) comprising the term(s)/phrase(s).
43. A computing device comprising:
generating means to generate term clusters as a function of calculated similarity of term vectors, each term vector being generated from search results associated with a set of high frequency of occurrence (FOO) historical queries previously submitted to a search engine; and responsive to receiving a term/phrase from an entity, evaluating means to evaluate the term/phrase in view of terms/phrases in the term clusters to identify one or more related term suggestions.
44. A computing device as recited in claim 43, and wherein a multi-sense query comprises the term/phrase.
45. A computing device as recited in claim 43, and wherein the entity is a computer-program application and/or an end-user.
46. A computing device as recited in claim 43, and further comprising:
collecting means to collect historic query terms from a query log; and determining means to determine ones of the historic query terms with a high FOO.
47. A computing device as recited in claim 43, wherein the evaluating means further comprise:
identifying means to identify a match between the term/phrase and term(s)/phrase(s) from one or more term clusters; and responsive to identifying, generating means to generate related term suggestion(s) comprising the term(s)/phrase(s).
48. A computing device as recited in claim 43, wherein the generating means ti generate the term clusters further comprise:
sending means to send respective ones of the high FOO historical queries to the search engine to obtain the search results;
extracting means to extract features from at least a subset of search results corresponding to the respective ones; and producing means to produce term vectors from the features.
49. A computing device as recited in claim 43, wherein the term clusters are a first set of term clusters, and wherein the computing device further comprises:
determining means to determine that there is no match between the term/phrase and the terms/phrases; and responsive to the determining:
making means to make a second set of term clusters from calculated similarity of term vectors, each term vector being generated from search results associated with a set of low FOO historical queries previously submitted to the search engine; and evaluating means to evaluate the term/phrase in view of terms/phrases of the second set of term clusters to identify one or more related term suggestions.
50. A computing device as recited in claim 49, and further comprising:
calculating means to calculate that there is no match between the term/phrase and term(s)/phrase(s) from the first set of term clusters, the first set being based on high FOO historical queries; and responsive to the calculating, identifying means to identify a match between the term/phrase and term(s)/phrase(s) from one or more of the second set of term clusters, the second set being based on low FOO historical queries;
and responsive to identifying, generating means to generate related term suggestion(s) comprising the term(s)/phrase(s).
CA2504106A 2004-04-15 2005-04-14 Related term suggestion for multi-sense query Expired - Fee Related CA2504106C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US10/825,894 US7428529B2 (en) 2004-04-15 2004-04-15 Term suggestion for multi-sense query
US10/825,894 2004-04-15

Publications (2)

Publication Number Publication Date
CA2504106A1 true CA2504106A1 (en) 2005-10-15
CA2504106C CA2504106C (en) 2014-06-17

Family

ID=34939289

Family Applications (1)

Application Number Title Priority Date Filing Date
CA2504106A Expired - Fee Related CA2504106C (en) 2004-04-15 2005-04-14 Related term suggestion for multi-sense query

Country Status (10)

Country Link
US (1) US7428529B2 (en)
EP (1) EP1587011A1 (en)
JP (1) JP4726528B2 (en)
KR (1) KR101109236B1 (en)
CN (1) CN100517311C (en)
AU (1) AU2005201682A1 (en)
BR (1) BRPI0501320A (en)
CA (1) CA2504106C (en)
MX (1) MXPA05004099A (en)
RU (1) RU2393533C2 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116628201A (en) * 2023-05-18 2023-08-22 浙江数洋科技有限公司 Intelligent grouping and pushing method for text database

Families Citing this family (158)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7194483B1 (en) 2001-05-07 2007-03-20 Intelligenxia, Inc. Method, system, and computer program product for concept-based multi-dimensional analysis of unstructured information
USRE46973E1 (en) 2001-05-07 2018-07-31 Ureveal, Inc. Method, system, and computer program product for concept-based multi-dimensional analysis of unstructured information
US8589413B1 (en) 2002-03-01 2013-11-19 Ixreveal, Inc. Concept-based method and system for dynamically analyzing results from search engines
US10032452B1 (en) 2016-12-30 2018-07-24 Google Llc Multimodal transmission of packetized data
US7836076B2 (en) * 2004-08-20 2010-11-16 Hewlett-Packard Development Company, L.P. Distributing content indices
US7801899B1 (en) * 2004-10-01 2010-09-21 Google Inc. Mixing items, such as ad targeting keyword suggestions, from heterogeneous sources
KR100669971B1 (en) * 2004-11-24 2007-01-16 엔에이치엔(주) Method for providing a search service and system for executing the method
US7620628B2 (en) * 2004-12-06 2009-11-17 Yahoo! Inc. Search processing with automatic categorization of queries
US7428533B2 (en) * 2004-12-06 2008-09-23 Yahoo! Inc. Automatic generation of taxonomies for categorizing queries and search query processing using taxonomies
US8375017B1 (en) * 2005-01-28 2013-02-12 Manta Media, Inc. Automated keyword analysis system and method
US7668808B2 (en) * 2005-03-10 2010-02-23 Yahoo! Inc. System for modifying queries before presentation to a sponsored search generator or other matching system where modifications improve coverage without a corresponding reduction in relevance
US8438142B2 (en) 2005-05-04 2013-05-07 Google Inc. Suggesting and refining user input based on original user input
US7725485B1 (en) 2005-08-01 2010-05-25 Google Inc. Generating query suggestions using contextual information
CN101351795B (en) 2005-10-11 2012-07-18 Ix锐示公司 System, method and device for concept based searching and analysis
US8015065B2 (en) * 2005-10-28 2011-09-06 Yahoo! Inc. Systems and methods for assigning monetary values to search terms
EP1949273A1 (en) * 2005-11-16 2008-07-30 Evri Inc. Extending keyword searching to syntactically and semantically annotated data
US8010523B2 (en) * 2005-12-30 2011-08-30 Google Inc. Dynamic search box for web browser
US7676485B2 (en) 2006-01-20 2010-03-09 Ixreveal, Inc. Method and computer program product for converting ontologies into concept semantic networks
US7747083B2 (en) * 2006-03-27 2010-06-29 Yahoo! Inc. System and method for good nearest neighbor clustering of text
US20070244862A1 (en) * 2006-04-13 2007-10-18 Randy Adams Systems and methods for ranking vertical domains
US9330170B2 (en) * 2006-05-16 2016-05-03 Sony Corporation Relating objects in different mediums
US20070271255A1 (en) * 2006-05-17 2007-11-22 Nicky Pappo Reverse search-engine
US7937397B2 (en) * 2006-08-22 2011-05-03 Fuji Xerox Co., Ltd. Apparatus and method for term context modeling for information retrieval
US7685201B2 (en) * 2006-09-08 2010-03-23 Microsoft Corporation Person disambiguation using name entity extraction-based clustering
US7689548B2 (en) * 2006-09-22 2010-03-30 Microsoft Corporation Recommending keywords based on bidding patterns
US7577643B2 (en) * 2006-09-29 2009-08-18 Microsoft Corporation Key phrase extraction from query logs
US20080113801A1 (en) * 2006-11-13 2008-05-15 Roland Moreno Game based on combinations of words and implemented by computer means
US7603348B2 (en) * 2007-01-26 2009-10-13 Yahoo! Inc. System for classifying a search query
US7685084B2 (en) * 2007-02-09 2010-03-23 Yahoo! Inc. Term expansion using associative matching of labeled term pairs
WO2008097194A1 (en) * 2007-02-09 2008-08-14 Agency For Science, Technology And Research Keyword classification and determination in language modelling
CA2717462C (en) * 2007-03-14 2016-09-27 Evri Inc. Query templates and labeled search tip system, methods, and techniques
CN101276361B (en) * 2007-03-28 2010-09-15 阿里巴巴集团控股有限公司 Method and system for displaying related key words
US7702614B1 (en) 2007-03-30 2010-04-20 Google Inc. Index updating using segment swapping
US8166021B1 (en) 2007-03-30 2012-04-24 Google Inc. Query phrasification
US8086594B1 (en) 2007-03-30 2011-12-27 Google Inc. Bifurcated document relevance scoring
US8166045B1 (en) 2007-03-30 2012-04-24 Google Inc. Phrase extraction using subphrase scoring
US7925655B1 (en) 2007-03-30 2011-04-12 Google Inc. Query scheduling using hierarchical tiers of index servers
US7693813B1 (en) 2007-03-30 2010-04-06 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US7698344B2 (en) 2007-04-02 2010-04-13 Microsoft Corporation Search macro suggestions relevant to search queries
KR101254362B1 (en) 2007-05-18 2013-04-12 엔에이치엔(주) Method and system for providing keyword ranking using common affix
US7822774B2 (en) * 2007-06-11 2010-10-26 Microsoft Corporation Using link structure for suggesting related queries
JP2009031931A (en) * 2007-07-25 2009-02-12 Univ Waseda Search word clustering device, method, program and recording medium
US8583670B2 (en) * 2007-10-04 2013-11-12 Microsoft Corporation Query suggestions for no result web searches
US9195743B2 (en) * 2007-10-04 2015-11-24 Yahoo! Inc. System and method for creating and applying predictive user click models to predict a target page associated with a search query
WO2009052308A1 (en) 2007-10-17 2009-04-23 Roseman Neil S Nlp-based content recommender
US8594996B2 (en) 2007-10-17 2013-11-26 Evri Inc. NLP-based entity recognition and disambiguation
JP4724701B2 (en) * 2007-10-30 2011-07-13 日本電信電話株式会社 Text search server computer, text search method, text search program, and recording medium recording the program
US8019748B1 (en) 2007-11-14 2011-09-13 Google Inc. Web search refinement
US8458179B2 (en) * 2007-11-29 2013-06-04 Palo Alto Research Center Incorporated Augmenting privacy policies with inference detection
US20090171929A1 (en) * 2007-12-26 2009-07-02 Microsoft Corporation Toward optimized query suggeston: user interfaces and algorithms
US7984004B2 (en) * 2008-01-17 2011-07-19 Microsoft Corporation Query suggestion generation
US10269024B2 (en) * 2008-02-08 2019-04-23 Outbrain Inc. Systems and methods for identifying and measuring trends in consumer content demand within vertically associated websites and related content
US20120053990A1 (en) * 2008-05-07 2012-03-01 Nice Systems Ltd. System and method for predicting customer churn
US8386476B2 (en) * 2008-05-20 2013-02-26 Gary Stephen Shuster Computer-implemented search using result matching
US7890516B2 (en) * 2008-05-30 2011-02-15 Microsoft Corporation Recommending queries when searching against keywords
US20090327223A1 (en) * 2008-06-26 2009-12-31 Microsoft Corporation Query-driven web portals
US20100070339A1 (en) * 2008-09-15 2010-03-18 Google Inc. Associating an Entity with a Category
AU2009304581A1 (en) * 2008-10-14 2010-04-22 Remarqueble Pty Ltd Search, analysis and categorization
US8171043B2 (en) * 2008-10-24 2012-05-01 Yahoo! Inc. Methods for improving the diversity of image search results
US8463808B2 (en) * 2008-11-07 2013-06-11 Raytheon Company Expanding concept types in conceptual graphs
US8386489B2 (en) * 2008-11-07 2013-02-26 Raytheon Company Applying formal concept analysis to validate expanded concept types
US9305051B2 (en) * 2008-12-10 2016-04-05 Yahoo! Inc. Mining broad hidden query aspects from user search sessions
US9158838B2 (en) * 2008-12-15 2015-10-13 Raytheon Company Determining query return referents for concept types in conceptual graphs
US8577924B2 (en) * 2008-12-15 2013-11-05 Raytheon Company Determining base attributes for terms
US9087293B2 (en) * 2008-12-23 2015-07-21 Raytheon Company Categorizing concept types of a conceptual graph
CN101464897A (en) * 2009-01-12 2009-06-24 阿里巴巴集团控股有限公司 Word matching and information query method and device
US9245243B2 (en) 2009-04-14 2016-01-26 Ureveal, Inc. Concept-based analysis of structured and unstructured data using concept inheritance
CA2796408A1 (en) * 2009-04-16 2010-10-21 Evri Inc. Enhanced advertisement targeting
CN102625936B (en) * 2009-08-04 2016-11-23 谷歌公司 Query suggestion from document
US20110040604A1 (en) * 2009-08-13 2011-02-17 Vertical Acuity, Inc. Systems and Methods for Providing Targeted Content
US20110040774A1 (en) * 2009-08-14 2011-02-17 Raytheon Company Searching Spoken Media According to Phonemes Derived From Expanded Concepts Expressed As Text
CN102056335B (en) * 2009-11-06 2014-02-19 华为技术有限公司 Mobile search method, device and system
US20110145226A1 (en) * 2009-12-10 2011-06-16 Microsoft Corporation Product similarity measure
US20110161091A1 (en) * 2009-12-24 2011-06-30 Vertical Acuity, Inc. Systems and Methods for Connecting Entities Through Content
US8631004B2 (en) * 2009-12-28 2014-01-14 Yahoo! Inc. Search suggestion clustering and presentation
US8150841B2 (en) * 2010-01-20 2012-04-03 Microsoft Corporation Detecting spiking queries
US8645125B2 (en) 2010-03-30 2014-02-04 Evri, Inc. NLP-based systems and methods for providing quotations
JP5165719B2 (en) * 2010-03-30 2013-03-21 ヤフー株式会社 Information processing apparatus, data extraction method, and program
US20120130816A1 (en) * 2010-03-31 2012-05-24 Rakuten, Inc. Search system, search method, search program and recording medium
US8161073B2 (en) 2010-05-05 2012-04-17 Holovisions, LLC Context-driven search
US20110314001A1 (en) * 2010-06-18 2011-12-22 Microsoft Corporation Performing query expansion based upon statistical analysis of structured data
US8725739B2 (en) 2010-11-01 2014-05-13 Evri, Inc. Category-based content recommendation
EP2635965A4 (en) * 2010-11-05 2016-08-10 Rakuten Inc Systems and methods regarding keyword extraction
US20120124072A1 (en) 2010-11-16 2012-05-17 Microsoft Corporation System level search user interface
US10346479B2 (en) 2010-11-16 2019-07-09 Microsoft Technology Licensing, Llc Facilitating interaction with system level search user interface
US10073927B2 (en) 2010-11-16 2018-09-11 Microsoft Technology Licensing, Llc Registration for system level search user interface
US8515984B2 (en) 2010-11-16 2013-08-20 Microsoft Corporation Extensible search term suggestion engine
RU2604677C2 (en) 2010-12-22 2016-12-10 Конинклейке Филипс Электроникс Н.В. Creating access control policy based on consumer privacy preferences
CN102567408B (en) 2010-12-31 2014-06-04 阿里巴巴集团控股有限公司 Method and device for recommending search keyword
CN102646103B (en) * 2011-02-18 2016-03-16 腾讯科技(深圳)有限公司 The clustering method of term and device
US8732151B2 (en) 2011-04-01 2014-05-20 Microsoft Corporation Enhanced query rewriting through statistical machine translation
US20120310690A1 (en) * 2011-06-06 2012-12-06 Winshuttle, Llc Erp transaction recording to tables system and method
US20120323627A1 (en) * 2011-06-14 2012-12-20 Microsoft Corporation Real-time Monitoring of Public Sentiment
US8930393B1 (en) * 2011-10-05 2015-01-06 Google Inc. Referent based search suggestions
US10013152B2 (en) 2011-10-05 2018-07-03 Google Llc Content selection disambiguation
US8724904B2 (en) * 2011-10-25 2014-05-13 International Business Machines Corporation Anomaly detection in images and videos
US9235565B2 (en) 2012-02-14 2016-01-12 Facebook, Inc. Blending customized user dictionaries
US9330083B2 (en) 2012-02-14 2016-05-03 Facebook, Inc. Creating customized user dictionary
US9330082B2 (en) 2012-02-14 2016-05-03 Facebook, Inc. User experience with customized user dictionary
US20130211915A1 (en) * 2012-02-14 2013-08-15 Erick Tseng Advertising Based on Customized User Dictionary
CN103365839B (en) * 2012-03-26 2017-12-12 深圳市世纪光速信息技术有限公司 The recommendation searching method and device of a kind of search engine
US9922334B1 (en) 2012-04-06 2018-03-20 Google Llc Providing an advertisement based on a minimum number of exposures
US10776830B2 (en) 2012-05-23 2020-09-15 Google Llc Methods and systems for identifying new computers and providing matching services
US10152723B2 (en) 2012-05-23 2018-12-11 Google Llc Methods and systems for identifying new computers and providing matching services
US9465875B2 (en) * 2012-05-31 2016-10-11 Hewlett Packard Enterprise Development Lp Searching based on an identifier of a searcher
US9442909B2 (en) 2012-10-11 2016-09-13 International Business Machines Corporation Real time term suggestion using text analytics
US20140188894A1 (en) * 2012-12-27 2014-07-03 Google Inc. Touch to search
US10735552B2 (en) 2013-01-31 2020-08-04 Google Llc Secondary transmissions of packetized data
US10650066B2 (en) 2013-01-31 2020-05-12 Google Llc Enhancing sitelinks with creative content
US9501506B1 (en) 2013-03-15 2016-11-22 Google Inc. Indexing system
US10152538B2 (en) * 2013-05-06 2018-12-11 Dropbox, Inc. Suggested search based on a content item
US9483568B1 (en) 2013-06-05 2016-11-01 Google Inc. Indexing system
US20150026153A1 (en) 2013-07-17 2015-01-22 Thoughtspot, Inc. Search engine for information retrieval system
US9405794B2 (en) 2013-07-17 2016-08-02 Thoughtspot, Inc. Information retrieval system
WO2015161515A1 (en) * 2014-04-25 2015-10-29 Yahoo! Inc. Systems and methods for commercial query suggestion
CN105183733A (en) * 2014-06-05 2015-12-23 阿里巴巴集团控股有限公司 Methods for matching text information and pushing business object, and devices for matching text information and pushing business object
CN105446984A (en) * 2014-06-30 2016-03-30 阿里巴巴集团控股有限公司 Expansion word pair screening method and device
JP6325502B2 (en) * 2015-10-08 2018-05-16 Necパーソナルコンピュータ株式会社 Information processing apparatus, information processing system, and information processing method
US10699183B2 (en) * 2016-03-31 2020-06-30 ZenDesk, Inc. Automated system for proposing help center articles to be written to facilitate resolving customer-service requests
US11436610B2 (en) * 2016-03-31 2022-09-06 ZenDesk, Inc. Automatically clustering customer-support requests to form customer-support topics
JP6555174B2 (en) * 2016-04-01 2019-08-07 京セラドキュメントソリューションズ株式会社 Information processing apparatus and program
CN107784029B (en) * 2016-08-31 2022-02-08 阿里巴巴集团控股有限公司 Method, server and client for generating prompt keywords and establishing index relationship
US10242113B2 (en) * 2016-10-13 2019-03-26 International Business Machines Corporation Modifying messages to be more discoverable on a social network
US11205103B2 (en) 2016-12-09 2021-12-21 The Research Foundation for the State University Semisupervised autoencoder for sentiment analysis
US10298837B2 (en) 2016-12-28 2019-05-21 Facebook, Inc. Systems and methods for presenting content based on unstructured visual data
US20180181844A1 (en) * 2016-12-28 2018-06-28 Facebook, Inc. Systems and methods for presenting content based on unstructured visual data
US10593329B2 (en) 2016-12-30 2020-03-17 Google Llc Multimodal transmission of packetized data
US10708313B2 (en) 2016-12-30 2020-07-07 Google Llc Multimodal transmission of packetized data
JP6790942B2 (en) * 2017-03-16 2020-11-25 富士通株式会社 Search support program, search support device and search support method
EP3673383A4 (en) 2017-10-10 2021-01-20 Thoughtspot Inc. Automatic database analysis
US10769141B2 (en) 2017-10-27 2020-09-08 Microsoft Technology Licensing, Llc Expanding search queries
US10747793B2 (en) * 2017-10-27 2020-08-18 Microsoft Technology Licensing, Llc Expanding search queries
US11276089B1 (en) * 2017-12-20 2022-03-15 Groupon, Inc. Method, apparatus, and computer program product for adaptive tail digital content object bid value generation
US11593844B1 (en) 2017-12-20 2023-02-28 Groupon, Inc. Method, apparatus, and computer program product for predictive initial electronic bid value generation for new digital content objects
US11157564B2 (en) 2018-03-02 2021-10-26 Thoughtspot, Inc. Natural language question answering systems
US11042896B1 (en) * 2018-03-12 2021-06-22 Inmar Clearing, Inc. Content influencer scoring system and related methods
EP3550444B1 (en) 2018-04-02 2023-12-27 Thoughtspot Inc. Query generation based on a logical data model
US10872303B2 (en) 2018-05-31 2020-12-22 Microsoft Technology Licensing, Llc Generating semantic representations for organizations and computing peer organization groups
US11409910B2 (en) 2018-05-31 2022-08-09 Microsoft Technology Licensing, Llc Predicting confidential data value insights at organization level using peer organization group
EP3588303B1 (en) * 2018-06-12 2021-02-17 Syntel, Inc. System and method for identifying test cases for software development
US11023486B2 (en) 2018-11-13 2021-06-01 Thoughtspot, Inc. Low-latency predictive database analysis
US11544239B2 (en) 2018-11-13 2023-01-03 Thoughtspot, Inc. Low-latency database analysis using external data sources
US11580147B2 (en) 2018-11-13 2023-02-14 Thoughtspot, Inc. Conversational database analysis
US11416477B2 (en) 2018-11-14 2022-08-16 Thoughtspot, Inc. Systems and methods for database analysis
US11334548B2 (en) 2019-01-31 2022-05-17 Thoughtspot, Inc. Index sharding
CN109947924B (en) * 2019-03-21 2021-08-31 百度在线网络技术(北京)有限公司 Dialogue system training data construction method and device, electronic equipment and storage medium
US11288453B1 (en) * 2019-03-22 2022-03-29 Pinterest, Inc. Key-word identification
US11928114B2 (en) 2019-04-23 2024-03-12 Thoughtspot, Inc. Query generation based on a logical data model with one-to-one joins
US11442932B2 (en) 2019-07-16 2022-09-13 Thoughtspot, Inc. Mapping natural language to queries using a query grammar
US10970319B2 (en) 2019-07-29 2021-04-06 Thoughtspot, Inc. Phrase indexing
US11354326B2 (en) 2019-07-29 2022-06-07 Thoughtspot, Inc. Object indexing
US11200227B1 (en) 2019-07-31 2021-12-14 Thoughtspot, Inc. Lossless switching between search grammars
US11409744B2 (en) 2019-08-01 2022-08-09 Thoughtspot, Inc. Query generation based on merger of subqueries
US11100170B2 (en) * 2019-11-15 2021-08-24 Microsoft Technology Licensing, Llc Domain-agnostic structured search query exploration
JP2021149613A (en) * 2020-03-19 2021-09-27 株式会社野村総合研究所 Natural language processing apparatus and program
US11544272B2 (en) 2020-04-09 2023-01-03 Thoughtspot, Inc. Phrase translation for a low-latency database analysis system
US11580111B2 (en) 2021-04-06 2023-02-14 Thoughtspot, Inc. Distributed pseudo-random subset generation
CN113505281B (en) * 2021-06-15 2022-11-08 北京三快在线科技有限公司 Search suggestion word generation method and device, electronic equipment and readable storage medium

Family Cites Families (46)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03122770A (en) * 1989-10-05 1991-05-24 Ricoh Co Ltd Method for retrieving keyword associative document
US5265065A (en) * 1991-10-08 1993-11-23 West Publishing Company Method and apparatus for information retrieval from a database by replacing domain specific stemmed phases in a natural language to create a search query
US5488725A (en) * 1991-10-08 1996-01-30 West Publishing Company System of document representation retrieval by successive iterated probability sampling
US5442778A (en) 1991-11-12 1995-08-15 Xerox Corporation Scatter-gather: a cluster-based method and apparatus for browsing large document collections
US7251637B1 (en) * 1993-09-20 2007-07-31 Fair Isaac Corporation Context vector generation and retrieval
US5576954A (en) 1993-11-05 1996-11-19 University Of Central Florida Process for determination of text relevancy
US5812134A (en) 1996-03-28 1998-09-22 Critical Thought, Inc. User interface navigational system & method for interactive representation of information contained within a database
US6188776B1 (en) 1996-05-21 2001-02-13 Interval Research Corporation Principle component analysis of images for the automatic location of control points
JP3108015B2 (en) 1996-05-22 2000-11-13 松下電器産業株式会社 Hypertext search device
US6457004B1 (en) * 1997-07-03 2002-09-24 Hitachi, Ltd. Document retrieval assisting method, system and service using closely displayed areas for titles and topics
US5987460A (en) * 1996-07-05 1999-11-16 Hitachi, Ltd. Document retrieval-assisting method and system for the same and document retrieval service using the same with document frequency and term frequency
GB9701866D0 (en) 1997-01-30 1997-03-19 British Telecomm Information retrieval
US5819258A (en) * 1997-03-07 1998-10-06 Digital Equipment Corporation Method and apparatus for automatically generating hierarchical categories from large document collections
US6298351B1 (en) 1997-04-11 2001-10-02 International Business Machines Corporation Modifying an unreliable training set for supervised classification
US6470307B1 (en) * 1997-06-23 2002-10-22 National Research Council Of Canada Method and apparatus for automatically identifying keywords within a document
US5927042A (en) * 1997-07-24 1999-07-27 Last; Harry J. Composite beam enclosure structure
US5845278A (en) 1997-09-12 1998-12-01 Inioseek Corporation Method for automatically selecting collections to search in full text searches
US6003027A (en) 1997-11-21 1999-12-14 International Business Machines Corporation System and method for determining confidence levels for the results of a categorization system
US6006225A (en) 1998-06-15 1999-12-21 Amazon.Com Refining search queries by the suggestion of correlated terms from prior searches
US6742003B2 (en) 2001-04-30 2004-05-25 Microsoft Corporation Apparatus and accompanying methods for visualizing clusters of data and hierarchical cluster classifications
US6598054B2 (en) 1999-01-26 2003-07-22 Xerox Corporation System and method for clustering data objects in a collection
US6226408B1 (en) 1999-01-29 2001-05-01 Hnc Software, Inc. Unsupervised identification of nonlinear data cluster in multidimensional data
US7225182B2 (en) 1999-05-28 2007-05-29 Overture Services, Inc. Recommending search terms using collaborative filtering and web spidering
US7835943B2 (en) * 1999-05-28 2010-11-16 Yahoo! Inc. System and method for providing place and price protection in a search result list generated by a computer network search engine
US6711585B1 (en) 1999-06-15 2004-03-23 Kanisa Inc. System and method for implementing a knowledge management system
US7725307B2 (en) * 1999-11-12 2010-05-25 Phoenix Solutions, Inc. Query engine for processing voice based queries including semantic decoding
US7392185B2 (en) 1999-11-12 2008-06-24 Phoenix Solutions, Inc. Speech based learning/training system using semantic decoding
US6556983B1 (en) 2000-01-12 2003-04-29 Microsoft Corporation Methods and apparatus for finding semantic information, such as usage logs, similar to a query using a pattern lattice data space
US6615209B1 (en) 2000-02-22 2003-09-02 Google, Inc. Detecting query-specific duplicate documents
EP1275042A2 (en) 2000-03-06 2003-01-15 Kanisa Inc. A system and method for providing an intelligent multi-step dialog with a user
US6697998B1 (en) 2000-06-12 2004-02-24 International Business Machines Corporation Automatic labeling of unlabeled text data
US6578032B1 (en) * 2000-06-28 2003-06-10 Microsoft Corporation Method and system for performing phrase/word clustering and cluster merging
US7490092B2 (en) 2000-07-06 2009-02-10 Streamsage, Inc. Method and system for indexing and searching timed media information based upon relevance intervals
US7587428B2 (en) 2000-10-13 2009-09-08 Microsoft Corporation Maintaining a relationship between two different items of data
US6772120B1 (en) 2000-11-21 2004-08-03 Hewlett-Packard Development Company, L.P. Computer method and apparatus for segmenting text streams
US7356530B2 (en) * 2001-01-10 2008-04-08 Looksmart, Ltd. Systems and methods of retrieving relevant information
WO2002071243A1 (en) 2001-03-01 2002-09-12 Biowulf Technologies, Llc Spectral kernels for learning machines
US6892193B2 (en) 2001-05-10 2005-05-10 International Business Machines Corporation Method and apparatus for inducing classifiers for multimedia based on unified representation of features reflecting disparate modalities
US20030065632A1 (en) 2001-05-30 2003-04-03 Haci-Murat Hubey Scalable, parallelizable, fuzzy logic, boolean algebra, and multiplicative neural network based classifier, datamining, association rule finder and visualization software tool
US20030046389A1 (en) 2001-09-04 2003-03-06 Thieme Laura M. Method for monitoring a web site's keyword visibility in search engines and directories and resulting traffic from such keyword visibility
JP3918664B2 (en) 2002-07-10 2007-05-23 ヤマハ株式会社 Acoustic signal processing device
US7136876B1 (en) * 2003-03-03 2006-11-14 Hewlett-Packard Development Company, L.P. Method and system for building an abbreviation dictionary
US20040249808A1 (en) * 2003-06-06 2004-12-09 Microsoft Corporation Query expansion using query logs
US7225184B2 (en) * 2003-07-18 2007-05-29 Overture Services, Inc. Disambiguation of search phrases using interpretation clusters
US7165119B2 (en) 2003-10-14 2007-01-16 America Online, Inc. Search enhancement system and method having rankings, explicitly specified by the user, based upon applicability and validity of search parameters in regard to a subject matter
US7188762B2 (en) * 2004-05-07 2007-03-13 Advanced Card Technologies Llc Secure card package for transaction cards and method of activating the same

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116628201A (en) * 2023-05-18 2023-08-22 浙江数洋科技有限公司 Intelligent grouping and pushing method for text database
CN116628201B (en) * 2023-05-18 2023-10-20 浙江数洋科技有限公司 Intelligent grouping and pushing method for text database

Also Published As

Publication number Publication date
BRPI0501320A (en) 2005-12-06
JP4726528B2 (en) 2011-07-20
US20050234879A1 (en) 2005-10-20
MXPA05004099A (en) 2005-10-19
EP1587011A1 (en) 2005-10-19
RU2393533C2 (en) 2010-06-27
KR20060045770A (en) 2006-05-17
CN1684072A (en) 2005-10-19
RU2005111000A (en) 2006-10-20
CA2504106C (en) 2014-06-17
US7428529B2 (en) 2008-09-23
KR101109236B1 (en) 2012-01-30
JP2005302042A (en) 2005-10-27
CN100517311C (en) 2009-07-22
AU2005201682A1 (en) 2005-10-27

Similar Documents

Publication Publication Date Title
CA2504106C (en) Related term suggestion for multi-sense query
US7689585B2 (en) Reinforced clustering of multi-type data objects for search term suggestion
US7260568B2 (en) Verifying relevance between keywords and web site contents
Broder et al. Search advertising using web relevance feedback
US7509313B2 (en) System and method for processing a query
JP4908214B2 (en) Systems and methods for providing search query refinement.
US8468156B2 (en) Determining a geographic location relevant to a web page
US7809715B2 (en) Abbreviation handling in web search
US8156097B2 (en) Two stage search
US20120240020A1 (en) System and method for document collection, grouping and summarization
US20070136251A1 (en) System and Method for Processing a Query
US7822752B2 (en) Efficient retrieval algorithm by query term discrimination
Baeza-Yates et al. Query clustering for boosting web page ranking
Duan et al. Dependency relation based detection of lexicalized user goals
KEI JEHN MING Document clustering on target entities using persons and organizations

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed

Effective date: 20220301

MKLA Lapsed

Effective date: 20200831