WO2005109249A1 - Document information mining tool - Google Patents

Document information mining tool Download PDF

Info

Publication number
WO2005109249A1
WO2005109249A1 PCT/US2005/015012 US2005015012W WO2005109249A1 WO 2005109249 A1 WO2005109249 A1 WO 2005109249A1 US 2005015012 W US2005015012 W US 2005015012W WO 2005109249 A1 WO2005109249 A1 WO 2005109249A1
Authority
WO
WIPO (PCT)
Prior art keywords
key
document
indicator
computer
type
Prior art date
Application number
PCT/US2005/015012
Other languages
French (fr)
Inventor
Brent L. Hadley
Patrick J. Eames
Susan C. Chew
Original Assignee
The Boeing Company
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 The Boeing Company filed Critical The Boeing Company
Priority to JP2007511052A priority Critical patent/JP4808705B2/en
Priority to EP05742113.3A priority patent/EP1745396B1/en
Publication of WO2005109249A1 publication Critical patent/WO2005109249A1/en
Priority to HK07103281.5A priority patent/HK1095902A1/en

Links

Classifications

    • 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
    • G06F16/35Clustering; Classification
    • 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
    • G06F16/35Clustering; Classification
    • G06F16/355Class or cluster creation or modification
    • 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
    • G06F16/93Document management systems

Definitions

  • This invention relates generally to information extraction and, more specifically, to processing documents to enhance access to information stored in the documents.
  • hypertext linking has enabled referencing and cross-referencing of key terms simply by using a pointing device to select a point of interest in a document.
  • processes for information retrieval still can be improved. For example, creating a hyper-link to a source document involves human intervention to identify the term with which the link will be created and associating it with the link to the related information.
  • all the documents or at least those documents containing the links need to be computer-readable documents.
  • a non-computer-readable document may be scanned so that it is computer-viewable, but unless the document is computer-readable such as a text or graphics document, it is generally not possible to associate links with portions of the document.
  • a referenced document need not be computer-readable to be accessed from a link, if the reference target is not computer-readable, then a person linking to the document may be required to manually navigate through the target document to find the information of interest.
  • the task becomes even more complicated if one desires information in both documents and needs to switch back and forth between the documents.
  • Such inferential information might exist as an interrelationship between parts, or as an annotation regarding other parts that might be useful, and other similarly useful information. Extracting information expected to be relevant thus may obliterate other useful information.
  • SUMMARY OF THE INVENTION The various embodiments of the present invention are useful in mining documents to identify keys that facilitate the creation of references within the documents and that may also form a basis for links between documents.
  • the various embodiments exploit the fact that that documents of certain types adhere to formatting standards that may include titles, section headings, and other parts of the document that are useful for navigating to and within the document.
  • the various embodiments may also utilize pattern formats that may be present in documents of various types, such as character patterns that may include a number of charactershaving expected separators so that keysmay be identifiedwhen that pattern appears on the page.
  • the keys are stored in a structured format associated with the original document to facilitate linking the keys with related documents.
  • embodiments of the present invention are used to mine a .pdf type document to find textual and graphical keys in the document, using optical character recognition as needed to extract text from graphically-presented keys, and to store the keys in an extensible markup language (XML) document to facilitate creating links to the keys in the original .pdf document.
  • XML extensible markup language
  • embodiments of the present invention provide methods, computer- readable media, and systems for extracting referential keys from a document.
  • a document is parsed to identify at least one key, the key being identified from at least one contextual indication.
  • the key is classified according to a key type, the key type being identified from the contextual indication.
  • the key is extracted and then stored in a location in a structured shell with the location corresponding to the key type. As a result, the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
  • FIGURE 1 is a block representation of an embodiment of the present invention.
  • FIGURE 2 is a document to be mined according to an embodiment of the present invention.
  • FIGURE 3 is a flowchart of a routine for mining a document according to an embodiment of the present invention.
  • FIGURES 4-7 are examples of portions of documents mined to extract keys into a structured format.
  • FIGURE 8 is a structured format collecting a collection of keys mined from a document.
  • FIGURE 9 is a system according to an embodiment of the present invention.
  • the present invention relates generally to a system and method for linking related documents and, more specifically, to identifying, extracting, and collecting keys used in linking the related documents.
  • Many specific details of certain embodiments of the invention are set forth in the following description and in FIGURES 1-9 to provide a thorough understanding of such embodiments.
  • One skilled in the art will understand that the present invention may have additional embodiments, or that the present invention may be practiced without several of the details described in the following description. More particularly, embodiments of the present invention provide methods, computer- readable media, and systems for extracting referential keys from a document.
  • a document is parsed to identify at least one key, the key being identified from at least one contextual indication.
  • the key is classified according to a key type, the key type being identified from the contextual indication.
  • the key is extracted and then stored in a location in a structured shell with the location corresponding to the key type. As a result, the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
  • the document information mining tool reviews the document for elements an author of the document has included that can be used to navigate to and within the document.
  • the document mining tool identifies titles, headers, footers, page numbers, off-page references, annotations to graphics, and other authored elements that a human reader of the document might use to navigate for additional information.
  • the document mining tool identifies these navigational attributes automatically by dissecting the document.
  • the information mined is then stored using a suitable structured format, such as an XML document, allowing the mined information to be used as keys or tags to support linking to other documents.
  • the document mining tool is further suitably equipped with an optical character reading (OCR) capability to analyze text that appears in graphic form in order to mine it from the document.
  • OCR optical character reading
  • FIGURE 1 is a block diagram of a generalized system 100 for mining a source document 110 according to an embodiment of the invention.
  • the document 110 is subjected to examination according to mining rules 120 directed to identify and extract various navigational keys, as will be described in greater detail below.
  • the extracted keys are subsequently stored in a structured format 130.
  • the structured format 130 may include an XML document, although it will be appreciated that the extracted keys may also be stored in a substantive content-based document.
  • the keys are stored in a structured format used to facilitate navigation to and within the source document.
  • the mined information is used to navigate through the existing documents to facilitate access to the complete, underlying documents.
  • keys may be created by a user and transferred to the mining rules 120.
  • the source document 110 may then be parsed to locate at least one portion of the document 110 that includes a contextual indication corresponding to the key created by the user.
  • the user may then review the information located within the at least one portion identified during the parsing.
  • FIGURE 2 shows a sample document 200, which is a page selected from an aircraft maintenance manual (AMM).
  • the document 200 is formatted as a .pdf document and presented on a .pdf reader 210.
  • the document 200 could have been generated from a computer-readable document or scanned from a printed document.
  • a computer-readable document is one in which characters and figures are generated for display on a computer screen from character and figure representations.
  • non- computer-readable documents are documents received in image form because they were scanned into or otherwise stored in a computing system in an image format stored such as rasterized data.
  • the sample document 200 represents a document that will be mined to extract the keys in the document, which may include titles, task designations, and other information, that might be useful in navigating to, from, and within the document 200.
  • FIGURE 3 is a generalized flowchart of a method300 for mining a document such as the sample document 200 (FIGURE 2) according to an embodiment of the invention.
  • the routine 300 begins at a block 302.
  • the document to be mined is received.
  • the document may be a computer-readable document or non-computer-readable image-type document.
  • a next area of a document is parsed.
  • the exemplary routine 300 parses a document area by area. However, alternatively, the document could be parsed looking for particular features of the document as will be further described below.
  • a decision block 310 it is determined if a placement indicator is identified.
  • a placement indicator may indicate presence of a title, a header, footer, a page number, or another key having relevance as indicated by the position of the key on a page. Accordingly, the manner in which the key is detected may be based upon generalized knowledge of the formatting methods used in a document, or upon tribal knowledge of the formatting methods, or other specific document formatting rules.
  • Generalized knowledge includes knowledge that indicates that a series of numerals centered or in a corner of a header or footer of a series of pages represent page numbers.
  • General knowledge also may include, for example, that a term appearing at a top of a page generally includes a title.
  • tribal knowledge includes knowledge known to a group of persons familiar with a specific type of document.
  • One area of a document could include a placement-indicated key, a format-indicated key, and a font-indicated key, such as, for example, if footer includes both a page number and title or other identifier.
  • the position of a numeral alone, such as in a part of a header reserved for a particular identifier may alone be determinative. On the other hand, other aspects of the key may have to be considered. If it is determined at the decision block 312 that placement is determinative, the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied.
  • an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If a placement indicator is not identified or placement alone is not determinative, the routine proceeds to a decision block 316 where it is determined if a format indicator is identified.
  • a format-indicated key may include a string of digits identifiable because of their format. For example, a number including three numbers followed by a separator such as a dash, followed by three more numbers, another dash, and four more numbers might be expected to be a phone number.
  • a number including three numbers followed by a dash, followed by two more numbers followed by another dash, followed by four more numbers might be expected to be a Social Security number.
  • a string of characters followed by an "@" symbol followed by another string of characters including at least one period might be expected to be an e-mail address. Accordingly, a type of key can be determined based upon the format. Once a format indicator is identified at the decision block 316, the routine 300 proceeds to a decision block 318 where it is determined if the format is determinative of what type of key has been identified by itself or in combination with the placement indicator.
  • the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied. On the other hand, if the document is a non-computer-readable image-type document as previously described, an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If placement or format indicators are not identified, or they are not determinative alone, the routine proceeds to a decision block 320 where it is determined if a font indicator is identified.
  • a font-indicated key may include a key having a type signified by its font or typeface.
  • the routine 300 proceeds to a decision block 322 where it is determined if the font is determinative of what type of key has been identified by itself or in combination with other indicators. If it is determined at the decision block 322 that format alone or in combination with other indicators is determinative, the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied.
  • an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If indicators are not identified or are not determinative, the routine proceeds to a decision block 326 where it is determined if all areas of the document have been parsed. If not, the routine loops to the block 306 where the next area of the document will be parsed. On the other hand, if all areas of the document have been parsed, the routine 300 ends at a block 328.
  • FIGURE 3 shows a particular sequential ordering of the steps 310, 316 and 320, it will be appreciated that the order of the steps 310, 316 and 320 may be altered to suit a particular application.
  • FIGURES 4-7 illustrate examples of how keys of interest are mined from a document.
  • the keys are extracted and then stored in a structured document that, in these examples, are XML documents.
  • Example 400 (FIGURE 4) shows a key being extracted from a document segment 402.
  • the key includes a section title, "INTEGRATED DRIVE GENERATOR (IDG) - REMOVAL/INSTALLATION.”
  • the key is identified by rules 404 recognizing a combination of indicators using general, tribal, and/or other knowledge.
  • the potential key begins at a left side of a page and is a first line of a page body.
  • the potential key is all UPPERCASE.
  • the potential key is thus identified as a section title and identified as a key of interest.
  • the key is extracted and stored in a structured format 406, in this case an XML document, where the key is designated as a title so that if can be located for navigation purposes.
  • Example 500 shows a second key being extracted from a document segment 502.
  • the key includes a task identifier, an expected and sought-after key in an AMM.
  • the task identifier is determined in this case by format indicators.
  • the task identifier includes a string of two digits, two digits, two digits, three digits, and three digits all separated by dashes. Accordingly, a rule 504 identifies this format as a task identifier, which is thus identified as a key.
  • the key is extracted and stored in an XML document 506 where the key is designated as a task number so that it can be located for navigation purposes.
  • Example 600 (FIGURE 6) shows a third key being extracted from a document segment 602.
  • the key includes a chapter number (24-11-11), which may be another expected and sought- after key in an AMM.
  • the chapter number is determined in this case by placement and format indicators.
  • the chapter number is always on a right side of a page, and is always in the form of two digits, two digits, and two digits separated by dashes.
  • the present example also includes a page block (401) and a revision date (Jun 10/01) that follow the formatted chapter number.
  • a rule 604 identifies the placement and format indicators in the document segment 602 to identify the key.
  • the key is then extracted and stored in an XML document 606 where the key is designated as a chapter number so that it can be located for navigation purposes.
  • Example 700 shows a fourth key being extracted from a document segment 702.
  • the key includes a particular substantive section that, in this case, is "Effectivity" that is used to designate the applicability of a selected portion of the AMM to a particular aircraft.
  • the substantive section is determined in this case by placement and format indicators.
  • the substantive section of interest is always on a left side of a page (a placement indicator) and always follows the term "Effectivity.”
  • a rule 704 identifies the potential key as the substantive description "Effectivity” and identifies it as a key of interest.
  • the key is extracted and stored in an XML document 706 where the key is designated as "Effectivity" so that it can be located for navigation purposes.
  • FIGURE 8 shows an XML document 800 that collects the extracted keys.
  • the XML document 800 shows a page number 802, a chapter number 804 subdivided into its constituent parts, a title 806, a substantive section 808 (Effectivity), and a task identifier 810.
  • FIGURE 9 is a block diagram of an exemplary system 900 according to an embodiment of the present invention.
  • the system 900 is presented with a computer-readable or non- computer-readable document 910.
  • the system 900 uses a parser 920 to scan the document for potential keys as previously described.
  • the system also employs an identifier 930 that is programmed to use placement indicators, format indicators, font indicators, and other indicators to determine whether a potential key found by the parser is a key of interest. If it is a key of interest, a classifier 930 receives the key, classifies it according to indicators manifested by the key of interest, and stores the key in a structured format 950, such as in XML data. As previously described in connection with FIGURE 8, the XML data may then be used to navigate to, from, and within the document 910. While preferred and alternate embodiments of the invention have been illustrated and described, many changes can be made without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is not limited by the disclosure of the preferred and alternate embodiments. Instead, the invention should be determined entirely by reference to the claims that follow.

Abstract

Embodiments of the present invention provide methods, computer-readable media, and systems for extracting referential keys from a document. A document is parsed to identify at least one key, the key being identified from at least one contextual indication. The key is classified according to a key type, the key type being identified from the contextual indication. The key is extracted and then stored in a location in a structured shell with the location corresponding to the key type. As a result, the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.

Description

DOCUMENT INFORMATION MINING TOOL INVENTORS Brent L. Hadley Patrick J. Eames Susan C. Chew
FIELD OF THE INVENTION This invention relates generally to information extraction and, more specifically, to processing documents to enhance access to information stored in the documents.
BACKGROUND OF THE INVENTION People often need to access information that is recorded in documents. Such documents may range in length from a single page to many volumes. Certainly, the longer the documents are, the more difficult it is to access the specific information desired. As a result, long or complex documents include multiple tables and indices to facilitate hierarchical and key word searches for the information of interest. These tables and indices are time-consuming to create. Furthermore, although the tables and indices are helpful in locating information, using them is time-consuming since each page referred to in the table or index must be individually examined to find the entry of interest. The foregoing may be repeated numerous times before the substantive entry of interest is found. The proliferation of computers has revolutionized how information is accessed. Accordingly, computer-readable documents may now be searched using various software routines for terms of interest in the document.In particular, hypertext linking has enabled referencing and cross-referencing of key terms simply by using a pointing device to select a point of interest in a document. Despite the great advantages that computers permit in accessing and retrieving information, processes for information retrieval still can be improved. For example, creating a hyper-link to a source document involves human intervention to identify the term with which the link will be created and associating it with the link to the related information. Further, to create hyper-linked information, all the documents or at least those documents containing the links need to be computer-readable documents.As a result, a non-computer-readable document may be scanned so that it is computer-viewable, but unless the document is computer-readable such as a text or graphics document, it is generally not possible to associate links with portions of the document. Similarly, even though a referenced document need not be computer-readable to be accessed from a link, if the reference target is not computer-readable, then a person linking to the document may be required to manually navigate through the target document to find the information of interest. Certainly, the task becomes even more complicated if one desires information in both documents and needs to switch back and forth between the documents. In such cases, to avoid the difficulty of navigating back and forth, a user desiring portions of such documents will typically print the needed documents or parts of the documents. When users print such documents instead of accessing them on the computer clearly undermines one of the objectives of making such documents accessible by computer. To avoid the complexities of moving back and forth between documents, one possible is to extract information from documents that is expected to be relevant. Unfortunately, removing only the content from the documents may present other problems. For example, some regulatory agencies require that extracted content be verified as accurately including the content of the original document before it can be used. This verification is a time-consuming and costly process. In addition, extraction of content also may obliterate inferential information a user might otherwise obtain from the document. Such inferential information might exist as an interrelationship between parts, or as an annotation regarding other parts that might be useful, and other similarly useful information. Extracting information expected to be relevant thus may obliterate other useful information. SUMMARY OF THE INVENTION The various embodiments of the present invention are useful in mining documents to identify keys that facilitate the creation of references within the documents and that may also form a basis for links between documents. The various embodiments exploit the fact that that documents of certain types adhere to formatting standards that may include titles, section headings, and other parts of the document that are useful for navigating to and within the document. The various embodiments may also utilize pattern formats that may be present in documents of various types, such as character patterns that may include a number of charactershaving expected separators so that keysmay be identifiedwhen that pattern appears on the page. Once these parts of the document representing potentially useful navigational keys have been identified, the keys are stored in a structured format associated with the original document to facilitate linking the keys with related documents. In one non-limiting example, embodiments of the present invention are used to mine a .pdf type document to find textual and graphical keys in the document, using optical character recognition as needed to extract text from graphically-presented keys, and to store the keys in an extensible markup language (XML) document to facilitate creating links to the keys in the original .pdf document. More particularly, embodiments of the present invention provide methods, computer- readable media, and systems for extracting referential keys from a document. A document is parsed to identify at least one key, the key being identified from at least one contextual indication. The key is classified according to a key type, the key type being identified from the contextual indication. The key is extracted and then stored in a location in a structured shell with the location corresponding to the key type. As a result, the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
BRIEF DESCRIPTION OF THE DRAWINGS The preferred and alternative embodiments of the present invention are described in detail below with reference to the following drawings. FIGURE 1 is a block representation of an embodiment of the present invention. FIGURE 2 is a document to be mined according to an embodiment of the present invention. FIGURE 3 is a flowchart of a routine for mining a document according to an embodiment of the present invention. FIGURES 4-7 are examples of portions of documents mined to extract keys into a structured format. FIGURE 8 is a structured format collecting a collection of keys mined from a document. FIGURE 9 is a system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION The present invention relates generally to a system and method for linking related documents and, more specifically, to identifying, extracting, and collecting keys used in linking the related documents. Many specific details of certain embodiments of the invention are set forth in the following description and in FIGURES 1-9 to provide a thorough understanding of such embodiments. One skilled in the art, however, will understand that the present invention may have additional embodiments, or that the present invention may be practiced without several of the details described in the following description. More particularly, embodiments of the present invention provide methods, computer- readable media, and systems for extracting referential keys from a document. A document is parsed to identify at least one key, the key being identified from at least one contextual indication. The key is classified according to a key type, the key type being identified from the contextual indication. The key is extracted and then stored in a location in a structured shell with the location corresponding to the key type. As a result, the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted. Using prior knowledge of the formats employed by authors in creating documents, for a given document format (e.g., .pdf files) and document type (e.g., aircraft maintenance manuals), the document information mining tool reviews the document for elements an author of the document has included that can be used to navigate to and within the document. For example, the document mining tool identifies titles, headers, footers, page numbers, off-page references, annotations to graphics, and other authored elements that a human reader of the document might use to navigate for additional information. The document mining tool identifies these navigational attributes automatically by dissecting the document. The information mined is then stored using a suitable structured format, such as an XML document, allowing the mined information to be used as keys or tags to support linking to other documents. The document mining tool is further suitably equipped with an optical character reading (OCR) capability to analyze text that appears in graphic form in order to mine it from the document. The document information mining tool thus provides a low-level, literal extraction process by parsing a document, identifying sections of the document that can be used as navigational points to navigate to, from, and throughout the document, and storing the mined information in a useful structure. FIGURE 1 is a block diagram of a generalized system 100 for mining a source document 110 according to an embodiment of the invention. The document 110 is subjected to examination according to mining rules 120 directed to identify and extract various navigational keys, as will be described in greater detail below. The extracted keys are subsequently stored in a structured format 130. The structured format 130 may include an XML document, although it will be appreciated that the extracted keys may also be stored in a substantive content-based document. In one particular embodiment, the keys are stored in a structured format used to facilitate navigation to and within the source document. In still another particular embodiment, The mined information is used to navigate through the existing documents to facilitate access to the complete, underlying documents. Referring still to FIGURE 1, in another particular embodiment, keys may be created by a user and transferred to the mining rules 120. The source document 110 may then be parsed to locate at least one portion of the document 110 that includes a contextual indication corresponding to the key created by the user. The user may then review the information located within the at least one portion identified during the parsing. FIGURE 2 shows a sample document 200, which is a page selected from an aircraft maintenance manual (AMM). The document 200 is formatted as a .pdf document and presented on a .pdf reader 210. It will be appreciated that the document 200 could have been generated from a computer-readable document or scanned from a printed document. For purposes of this description, a computer-readable document is one in which characters and figures are generated for display on a computer screen from character and figure representations. In contrast, non- computer-readable documents are documents received in image form because they were scanned into or otherwise stored in a computing system in an image format stored such as rasterized data. The sample document 200 represents a document that will be mined to extract the keys in the document, which may include titles, task designations, and other information, that might be useful in navigating to, from, and within the document 200. FIGURE 3 is a generalized flowchart of a method300 for mining a document such as the sample document 200 (FIGURE 2) according to an embodiment of the invention. The routine 300 begins at a block 302. At a block 304, the document to be mined is received. As previously described, the document may be a computer-readable document or non-computer-readable image-type document. At a block 306, a next area of a document is parsed. The exemplary routine 300 parses a document area by area. However, alternatively, the document could be parsed looking for particular features of the document as will be further described below. At a decision block 310 it is determined if a placement indicator is identified. A placement indicator may indicate presence of a title, a header, footer, a page number, or another key having relevance as indicated by the position of the key on a page. Accordingly, the manner in which the key is detected may be based upon generalized knowledge of the formatting methods used in a document, or upon tribal knowledge of the formatting methods, or other specific document formatting rules. Generalized knowledge, for example, includes knowledge that indicates that a series of numerals centered or in a corner of a header or footer of a series of pages represent page numbers. General knowledge also may include, for example, that a term appearing at a top of a page generally includes a title. By contrast, tribal knowledge includes knowledge known to a group of persons familiar with a specific type of document. For example, in certain industries or organizations, it may be common practice to place a subject identifier in a header or footer of each page, even though the inclusion of that key may not be indicated by a specific rule. Finally, specific document formatting rules are rules literally dictated by document conventions. For example, it may be an established practice in an industry or organization to position system identifiers on a page of a document in a generally similar location. For example, the format used in legal briefs submitted to courts include titles, page numbers, and other elements placed in generally similar locations according to rules specified by the courts. Still referring to FIGURE 3, once a placement indicator is identified at the decision block 310, the routine 300 proceeds to a decision block 312 where it is determined if the placement is determinative of what type of key has been identified. One area of a document could include a placement-indicated key, a format-indicated key, and a font-indicated key, such as, for example, if footer includes both a page number and title or other identifier. The position of a numeral alone, such as in a part of a header reserved for a particular identifier may alone be determinative. On the other hand, other aspects of the key may have to be considered. If it is determined at the decision block 312 that placement is determinative, the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied. On the other hand, if the document is a non-computer-readable image-type document as previously described, an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If a placement indicator is not identified or placement alone is not determinative, the routine proceeds to a decision block 316 where it is determined if a format indicator is identified. A format-indicated key may include a string of digits identifiable because of their format. For example, a number including three numbers followed by a separator such as a dash, followed by three more numbers, another dash, and four more numbers might be expected to be a phone number. A number including three numbers followed by a dash, followed by two more numbers followed by another dash, followed by four more numbers might be expected to be a Social Security number. In a further example, a string of characters followed by an "@" symbol followed by another string of characters including at least one period might be expected to be an e-mail address. Accordingly, a type of key can be determined based upon the format. Once a format indicator is identified at the decision block 316, the routine 300 proceeds to a decision block 318 where it is determined if the format is determinative of what type of key has been identified by itself or in combination with the placement indicator. If it is determined at the decision block 318 that format alone or with the placement is determinative, the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied. On the other hand, if the document is a non-computer-readable image-type document as previously described, an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If placement or format indicators are not identified, or they are not determinative alone, the routine proceeds to a decision block 320 where it is determined if a font indicator is identified. A font-indicated key may include a key having a type signified by its font or typeface. For example, according to commonly accepted document conventions, information that is underlined, or that appears in bold type, or is presented in an enlarged font, or any combination of the foregoing may represent a title, a part name, a document identifier, or similar information. Once a font indicator is identified at the decision block 320, the routine 300 proceeds to a decision block 322 where it is determined if the font is determinative of what type of key has been identified by itself or in combination with other indicators. If it is determined at the decision block 322 that format alone or in combination with other indicators is determinative, the routine 300 proceeds to a block 314 where the key is extracted and stored in a structured format. If the document is a computer-readable document, characters representing the key are copied. On the other hand, if the document is a non-computer-readable image-type document as previously described, an optical character recognition (OCR) routine is invoked to scan the identified key to extract the characters from the key. If indicators are not identified or are not determinative, the routine proceeds to a decision block 326 where it is determined if all areas of the document have been parsed. If not, the routine loops to the block 306 where the next area of the document will be parsed. On the other hand, if all areas of the document have been parsed, the routine 300 ends at a block 328. Although FIGURE 3 shows a particular sequential ordering of the steps 310, 316 and 320, it will be appreciated that the order of the steps 310, 316 and 320 may be altered to suit a particular application. Moreover, the steps 310, 316 and 320 may also be executed in parallel, so that keys based on placement, format and font may be identified simultaneously. FIGURES 4-7 illustrate examples of how keys of interest are mined from a document. The examples 400, 500, 600, and 700 from FIGURES 4, 5, 6, and 7, respectively show examples of keys being extracted from a page of an aircraft maintenance manual (AMM). The keys are extracted and then stored in a structured document that, in these examples, are XML documents. Example 400 (FIGURE 4) shows a key being extracted from a document segment 402. In particular, the key includes a section title, "INTEGRATED DRIVE GENERATOR (IDG) - REMOVAL/INSTALLATION." The key is identified by rules 404 recognizing a combination of indicators using general, tribal, and/or other knowledge. In terms of placement, the potential key begins at a left side of a page and is a first line of a page body. In terms of font, the potential key is all UPPERCASE. The potential key is thus identified as a section title and identified as a key of interest. The key is extracted and stored in a structured format 406, in this case an XML document, where the key is designated as a title so that if can be located for navigation purposes. Example 500 (FIGURE 5) shows a second key being extracted from a document segment 502. The key includes a task identifier, an expected and sought-after key in an AMM. The task identifier is determined in this case by format indicators. The task identifier includes a string of two digits, two digits, two digits, three digits, and three digits all separated by dashes. Accordingly, a rule 504 identifies this format as a task identifier, which is thus identified as a key. The key is extracted and stored in an XML document 506 where the key is designated as a task number so that it can be located for navigation purposes. Example 600 (FIGURE 6) shows a third key being extracted from a document segment 602. The key includes a chapter number (24-11-11), which may be another expected and sought- after key in an AMM. The chapter number is determined in this case by placement and format indicators. In the present example, the chapter number is always on a right side of a page, and is always in the form of two digits, two digits, and two digits separated by dashes. The present example also includes a page block (401) and a revision date (Jun 10/01) that follow the formatted chapter number. Again, a rule 604 identifies the placement and format indicators in the document segment 602 to identify the key. The key is then extracted and stored in an XML document 606 where the key is designated as a chapter number so that it can be located for navigation purposes. It will be appreciated that upon finding a formatted key such as the chapter number, the chapter number maybe further subdivided and stored in its constituent parts, including chapter (24), section (11), subject (11), and page block (401). Example 700 (FIGURE 7) shows a fourth key being extracted from a document segment 702. The key includes a particular substantive section that, in this case, is "Effectivity" that is used to designate the applicability of a selected portion of the AMM to a particular aircraft. The substantive section is determined in this case by placement and format indicators. In the present example, the substantive section of interest is always on a left side of a page (a placement indicator) and always follows the term "Effectivity." As in the previous example, a rule 704 identifies the potential key as the substantive description "Effectivity" and identifies it as a key of interest. The key is extracted and stored in an XML document 706 where the key is designated as "Effectivity" so that it can be located for navigation purposes. FIGURE 8 shows an XML document 800 that collects the extracted keys. The XML document 800 shows a page number 802, a chapter number 804 subdivided into its constituent parts, a title 806, a substantive section 808 (Effectivity), and a task identifier 810. The keys 802, 804, 806, 808, and 810 all are in a computer-readable format and are associated with the original document from which the keys 802, 804, 806, 808, and 810 wereextracted. Thus, the XML document 800 can be searched or otherwise accessed to find the keys 802, 804, 806, 808, and 810 of interest. The keys 802, 804, 806, 808, and 810 are in turn linked to the document 800 from which they were drawn. Thus, the mining tool facilitates access to the document of interest and the portion of interest. FIGURE 9 is a block diagram of an exemplary system 900 according to an embodiment of the present invention. The system 900 is presented with a computer-readable or non- computer-readable document 910. The system 900 uses a parser 920 to scan the document for potential keys as previously described. The system also employs an identifier 930 that is programmed to use placement indicators, format indicators, font indicators, and other indicators to determine whether a potential key found by the parser is a key of interest. If it is a key of interest, a classifier 930 receives the key, classifies it according to indicators manifested by the key of interest, and stores the key in a structured format 950, such as in XML data. As previously described in connection with FIGURE 8, the XML data may then be used to navigate to, from, and within the document 910. While preferred and alternate embodiments of the invention have been illustrated and described, many changes can be made without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is not limited by the disclosure of the preferred and alternate embodiments. Instead, the invention should be determined entirely by reference to the claims that follow.

Claims

What is claimed is:
1. A method for extracting referential keys from a document, the method comprising: parsing the document to identify at least one key, the key being identified from at least one indicator; classifying the key according to a key type, the key type being identified from the at least one indicator; extracting the key; and storing the key in a location in a structured shell, the location corresponding to the key type such that the key can be found by a search seeking at least one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
2. The method of Claim 1, wherein the document includes one of -a computer-readable content and a non-computer-readable content.
3. The method of Claim 1 , wherein the document includes a .pdf document.
4. The method of Claim 1, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
5. The method of Claim 1, wherein the indicator includes at least one of a placement indicator, a format indicator, and a font indicator.
6. The method of Claim 5, wherein the indicator is derived from knowledge including at least one of a general knowledge, a tribal knowledge, and a specific document formatting rules.
7. The method of Claim 5, wherein the placement indicator includes a page position.
8. The method of Claim 5, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
9. The method of Claim 5, wherein the font indicator includes at least one of a typeface, a boldface, an underlined text portion, and a font size.
10. The method of Claim 1 , wherein extracting the key includes copying the key.
11. The method of Claim 1, further comprising using optical character recognition to extract the key when the document includes a non-computer-readable content.
12. The method of Claim 1, wherein the structured shell includes an extensible markup language (XML) document.
13. A method for extracting referential keys from a document, the method comprising: parsing a .pdf document to identify at least one key, the key being identified from at least one indicator, the indicator including at least one of a placement indicator, a format indicator, and a font indicator; classifying the key according to a key type, the key type being identified from the at least one indicator; extracting the key; and storing the key in a location in an extensible markup language (XML) document, the location corresponding to the key type such that the key can be found by a search seeking at least one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
14. The method of Claim 13, wherein the document includes one of a computer-readable content and a non-computer-readable content.
15. The method of Claim 13, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
16. The method of Claim 13, wherein the indicator is derived from knowledgeincluding at least one of a general knowledge, a tribal knowledge, and a specific document formatting rule.
17. The method of Claim 13, wherein the placement indicator includes a page position.
18. The method of Claim 13, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
19. The method of Claim 13, wherein the font indicator includes at least one of a typeface, a boldface, a underlined text portion, and a font size.
20. The method of Claim 13, wherein extracting the key includes copying the key.
21. The method of Claim 13, further comprising using optical character recognition to extract the key when the document includes a non-computer-readable content.
22. A computer-readable medium having stored thereon instructions for extracting referential keys from a document, the computer-readable medium comprising: a first computer program portion adapted to parse the document to identify at least one key, the key being identified from at least one indicator; a second computer program portion adapted to classify the key according to a key type, the key type being identified from the at least one indicator; a third computer program portion adapted to extract the key; and a fourth computer program portion adapted to store the key in a location in a structured shell, the location corresponding to the key type such that the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
23. The computer-readable medium of Claim 22, wherein the document includes one of a computer-readable content and a non-computer-readable content.
24. The computer-readable medium of Claim 22, wherein the document includes a .pdf document.
25. The computer-readable medium of Claim 22, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
26. The computer-readable medium of Claim 22, wherein the indicator includes at least one of a placement indicator, a format indicator, and a font indicator.
27. The computer-readable medium of Claim 26, wherein the indicator is derived from knowledge including at least one of a general knowledge, a tribal knowledge, and a specific document formatting rule.
28. The computer-readable medium of Claim 26, wherein the placement indicator includes a page position.
29. The computer-readable medium of Claim 26, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
30. The computer-readable medium of Claim 26, wherein the font indicator includes at least one of a typeface, a boldface, a underlined text portion, and a font size.
31. The computer-readable medium of Claim 22, wherein extracting the key includes copying the key.
32. The computer-readable medium of Claim 23, further comprising a fifth computer program portion configured to use optical character recognition to extract the key when the document includes a non-computer-readable content.
33. The computer-readable medium of Claim 22, wherein the structured shell includes an extensible markup language (XML) document.
34. A computer-readable medium for extracting referential keys from a document, the computer-readable medium comprising: a first computer program portion adapted to parse a .pdf document to identify at least one key, the key being identified from at least one indicator, the indicator including at least one of a placement indicator, a format indicator, and a font indicator; a second computer program portion adapted to classify the key according to a key type, the key type being identified from the at least one indicator; a third computer program portion adapted to extract the key; and a fourth computer program portion adapted to store the key in a location in an extensible markup language (XML) document, the location corresponding to the key type such that the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
35. The computer-readable medium of Claim 34, wherein the document includes one of a computer-readable content and a non-computer-readable content.
36. The computer-readable medium of Claim 34, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
37. The computer-readable medium of Claim 34, wherein the indicator is derived from knowledge including at least one of a general knowledge, a tribal knowledge, and a specific document formatting rule.
38. The computer-readable medium of Claim 34, wherein the placement indicator includes a page position.
39. The computer-readable medium of Claim 34, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
40. The computer-readable medium of Claim 34, wherein the font indicator includes at least one of a typeface, a boldface, an underlined text portion, and a font size.
41. The computer-readable medium of Claim 34, wherein extracting the key includes copying the key.
42. The computer-readable medium of Claim 34, further comprising using optical character recognition to extract the key when the document includes a non-computer-readable content.
43. A system for extracting referential keys from a document, the system comprising: a parser configured to parse the document to identify at least one key, the key being identified from at least one indicator; an identifier configured to identify the key according to a key type, the key type being identified from the at least one indicator; and a classifier configured to extract the key and store the key in a location in a structured shell, the location corresponding to the key type such that the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
44. The system of Claim 43, wherein the document includes one of a computerrreadable content and a non-computer-readable content.
45. The system of Claim 43, wherein the document includes a .pdf document.
46. The system of Claim 43, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
47. The system of Claim 43, wherein the indicator includes at least one of a placement indicator, a format indicator, and a font indicator.
48. The system of Claim 47, wherein the indicator is derived from knowledge including at least one of a general knowledge, a tribal knowledge, and a specific document formatting rule.
49. The system of Claim 47, wherein the placement indicator includes a page position.
50. The system of Claim 47, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
51. The system of Claim 47, wherein the font indicator includes at least one of a typeface, boldface, an underlined text portion, and a font size.
52. The system of Claim 43, wherein extracting the key includes copying the key.
53. The system of Claim 43, further comprising an optical character recognizer configured to use optical character recognition to extract the key when the document includes non- computer-readable content.
54. The system of Claim 43, wherein the structured shell includes an extensible markup language (XML) document.
55. A system for extracting referential keys from a document, the system comprising: a parser configured to parse a .pdf document to identify at least one key, the key being identified from at least one indicator, the indicator including at least one of a placement indicator, a format indicator, and a font indicator; an identifier configured to identify the key according to a key type, the key type being identified from the at least one indicator; a classifier configured to extract the key and store the key in a location in an extensible markup language (XML) document, the location corresponding to the key type such that the key can be found by a search seeking one of the key and the key-type allowing a searcher to identify the document from which the key was extracted.
56. The system of Claim 55, wherein the document includes one of a computer-readable content and a non-computer-readable content.
57. The system of Claim 55, wherein the key type includes at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier, and a page number.
58. The system of Claim 55, wherein the indicator is derived from knowledge including at least one of a general knowledge, a tribal knowledge, and a specific document formatting rule.
59. The system of Claim 55, wherein the placement indicator includes a page position.
60. The system of Claim 55, wherein the format indicator includes a character pattern including at least one of a pattern of digits and a pattern of separators.
61. The system of Claim 55, wherein the font indicator includes at least one of a typeface, a boldface, an underlined text portion, and a font size.
62. The system of Claim 55, wherein extracting the key includes copying the key.
63. The system of Claim 55, further comprising using optical character recognition to extract the key when the document includes a non-computer-readable content.
64. A method of information searching in a document, comprising: creating a reference key corresponding to at least one contextual indicator present in the document; parsing successive portions of the document; identifying at least one portion of the document that includes the reference key; and reviewing the information included in the at least one portion.
65. The method of claim 64, wherein creating a reference key further comprises creating the key according to a predetermined key type.
66. The method of claim 65, wherein the key type comprises at least one of a title, a header, a footer, a document type, a document identifier, a subject identifier, a section identifier, a chapter identifier and a page number.
67. The method of claim 64, wherein the indicator comprises at least one of a placement indicator, a format indicator and a font indicator.
68. The method of claim 67, wherein the format indicator comprises a character pattern that includes at least one of a pattern of digits and a pattern of separators.
69. The method of claim 67, wherein the font indicator comprises at least one of a typeface, a boldface, an underlined text portion and a font size.
PCT/US2005/015012 2004-04-30 2005-04-29 Document information mining tool WO2005109249A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2007511052A JP4808705B2 (en) 2004-04-30 2005-04-29 Document information mining tool
EP05742113.3A EP1745396B1 (en) 2004-04-30 2005-04-29 Document information mining tool
HK07103281.5A HK1095902A1 (en) 2004-04-30 2007-03-27 Document information mining tool

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US10/835,903 2004-04-30
US10/835,903 US7756869B2 (en) 2004-04-30 2004-04-30 Methods and apparatus for extracting referential keys from a document

Publications (1)

Publication Number Publication Date
WO2005109249A1 true WO2005109249A1 (en) 2005-11-17

Family

ID=34967798

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2005/015012 WO2005109249A1 (en) 2004-04-30 2005-04-29 Document information mining tool

Country Status (5)

Country Link
US (2) US7756869B2 (en)
EP (1) EP1745396B1 (en)
JP (1) JP4808705B2 (en)
HK (1) HK1095902A1 (en)
WO (1) WO2005109249A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10579716B2 (en) 2017-11-06 2020-03-03 Microsoft Technology Licensing, Llc Electronic document content augmentation
US11341180B2 (en) * 2006-09-29 2022-05-24 Google Llc Displaying search results on a one or two dimensional graph
US11640499B2 (en) 2017-12-26 2023-05-02 RELX Inc. Systems, methods and computer program products for mining text documents to identify seminal issues and cases

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7756869B2 (en) * 2004-04-30 2010-07-13 The Boeing Company Methods and apparatus for extracting referential keys from a document
JP2006301835A (en) * 2005-04-19 2006-11-02 Fuji Xerox Co Ltd Transaction document management method and system
JP4802672B2 (en) * 2005-11-14 2011-10-26 株式会社日立製作所 Electronic document management apparatus, electronic document management program, electronic document management system
US7827177B2 (en) * 2006-10-16 2010-11-02 Oracle International Corporation Managing compound XML documents in a repository
US7853595B2 (en) 2007-01-30 2010-12-14 The Boeing Company Method and apparatus for creating a tool for generating an index for a document
US20090251338A1 (en) * 2008-04-03 2009-10-08 Livescribe, Inc. Ink Tags In A Smart Pen Computing System
KR100905434B1 (en) * 2008-08-08 2009-07-02 (주)이스트소프트 File uploading method with function of abstracting index-information in real-time and web-storage system using the same
US8713053B2 (en) * 2010-03-09 2014-04-29 Cisco Technology, Inc Active tags
US9582483B2 (en) 2012-07-13 2017-02-28 Xerox Corporation Automatically tagging variable data documents
JP5753828B2 (en) * 2012-09-27 2015-07-22 京セラドキュメントソリューションズ株式会社 Image processing apparatus and image processing program
US9208449B2 (en) 2013-03-15 2015-12-08 International Business Machines Corporation Process model generated using biased process mining
US10936895B2 (en) * 2017-07-26 2021-03-02 Vmware, Inc. Managing camera actions
CN114661904B (en) * 2022-03-10 2023-04-07 北京百度网讯科技有限公司 Method, apparatus, device, storage medium, and program for training document processing model

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020065857A1 (en) 2000-10-04 2002-05-30 Zbigniew Michalewicz System and method for analysis and clustering of documents for search engine
US6424982B1 (en) 1999-04-09 2002-07-23 Semio Corporation System and method for parsing a document using one or more break characters

Family Cites Families (60)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US589470A (en) * 1897-09-07 Matrix
US5444840A (en) * 1990-06-12 1995-08-22 Froessl; Horst Multiple image font processing
US5926565A (en) * 1991-10-28 1999-07-20 Froessl; Horst Computer method for processing records with images and multiple fonts
US5875263A (en) * 1991-10-28 1999-02-23 Froessl; Horst Non-edit multiple image font processing of records
JPH06290251A (en) * 1993-03-31 1994-10-18 Hitachi Ltd Document image processing method of electronic file device
US5666490A (en) * 1994-05-16 1997-09-09 Gillings; Dennis Computer network system and method for managing documents
DE69520177T2 (en) * 1994-08-05 2001-08-02 Canon Kk Method and apparatus for processing documents
DE69525401T2 (en) * 1994-09-12 2002-11-21 Adobe Systems Inc Method and device for identifying words described in a portable electronic document
US5708825A (en) * 1995-05-26 1998-01-13 Iconovex Corporation Automatic summary page creation and hyperlink generation
US5781914A (en) * 1995-06-30 1998-07-14 Ricoh Company, Ltd. Converting documents, with links to other electronic information, between hardcopy and electronic formats
US5752022A (en) * 1995-08-07 1998-05-12 International Business Machines Corp. Method for creating a hypertext language for a distributed computer network
US5835712A (en) * 1996-05-03 1998-11-10 Webmate Technologies, Inc. Client-server system using embedded hypertext tags for application and database development
US6516321B1 (en) * 1996-07-30 2003-02-04 Carlos De La Huerga Method for database address specification
KR100219117B1 (en) * 1996-08-24 1999-09-01 구자홍 Thin-film transistor liquid crystal display device and its manufacturing method
US6658624B1 (en) * 1996-09-24 2003-12-02 Ricoh Company, Ltd. Method and system for processing documents controlled by active documents with embedded instructions
US5787425A (en) * 1996-10-01 1998-07-28 International Business Machines Corporation Object-oriented data mining framework mechanism
US6070161A (en) * 1997-03-19 2000-05-30 Minolta Co., Ltd. Method of attaching keyword or object-to-key relevance ratio and automatic attaching device therefor
US5895470A (en) * 1997-04-09 1999-04-20 Xerox Corporation System for categorizing documents in a linked collection of documents
US6262729B1 (en) * 1997-04-14 2001-07-17 Apple Computer, Inc. Method and apparatus for binding user interface objects to application objects
US6910184B1 (en) * 1997-07-25 2005-06-21 Ricoh Company, Ltd. Document information management system
JP3478725B2 (en) * 1997-07-25 2003-12-15 株式会社リコー Document information management system
US6658623B1 (en) * 1997-09-15 2003-12-02 Fuji Xerox Co., Ltd. Displaying in a first document a selectable link to a second document based on a passive query
US6256631B1 (en) * 1997-09-30 2001-07-03 International Business Machines Corporation Automatic creation of hyperlinks
US6134552A (en) * 1997-10-07 2000-10-17 Sap Aktiengesellschaft Knowledge provider with logical hyperlinks
US6574644B2 (en) * 1997-11-26 2003-06-03 Siemens Corporate Research, Inc Automatic capturing of hyperlink specifications for multimedia documents
US6138129A (en) * 1997-12-16 2000-10-24 World One Telecom, Ltd. Method and apparatus for providing automated searching and linking of electronic documents
US6185549B1 (en) * 1998-04-29 2001-02-06 Lucent Technologies Inc. Method for mining association rules in data
US6324533B1 (en) * 1998-05-29 2001-11-27 International Business Machines Corporation Integrated database and data-mining system
US6718329B1 (en) * 1998-07-16 2004-04-06 Verizon New York Inc. Method and apparatus for generating typed nodes and links in a hypertext database from formation documents
US6370521B1 (en) * 1998-08-25 2002-04-09 Bell & Howell Mail Messaging Technologies Company Tracking system, method and computer program product for document processing
US6356898B2 (en) * 1998-08-31 2002-03-12 International Business Machines Corporation Method and system for summarizing topics of documents browsed by a user
US6496208B1 (en) * 1998-09-10 2002-12-17 Microsoft Corporation Method and apparatus for visualizing and exploring large hierarchical structures
US6253631B1 (en) 1999-01-14 2001-07-03 Bac-Lamb Wire Line Services Pumping unit lock down apparatus
US6665669B2 (en) * 2000-01-03 2003-12-16 Db Miner Technology Inc. Methods and system for mining frequent patterns
US6771816B1 (en) * 2000-01-19 2004-08-03 Adobe Systems Incorporated Generating a text mask for representing text pixels
US6584469B1 (en) * 2000-03-16 2003-06-24 International Business Machines Corporation Automatically initiating a knowledge portal query from within a displayed document
US20020049705A1 (en) * 2000-04-19 2002-04-25 E-Base Ltd. Method for creating content oriented databases and content files
US6745181B1 (en) * 2000-05-02 2004-06-01 Iphrase.Com, Inc. Information access method
US6654737B1 (en) * 2000-05-23 2003-11-25 Centor Software Corp. Hypertext-based database architecture
US6691120B1 (en) * 2000-06-30 2004-02-10 Ncr Corporation System, method and computer program product for data mining in a normalized relational database
JP3650929B2 (en) * 2000-08-09 2005-05-25 文孝 村山 Character content providing method and character content providing system
US6957384B2 (en) * 2000-12-27 2005-10-18 Tractmanager, Llc Document management system
US6684208B2 (en) * 2001-03-07 2004-01-27 Rockwell Technologies, Llc System and method for quantifying an extent to which a data mining algorithm captures useful information in input data
US6718327B1 (en) * 2001-08-31 2004-04-06 Openwave Systems Inc. Fault-tolerant queue with autonomous client operation
US7007231B2 (en) * 2002-01-07 2006-02-28 Chi Hung Dang Document management system employing multi-zone parsing process
JP2003308314A (en) * 2002-02-15 2003-10-31 Ricoh Co Ltd Device for supporting document preparation
US7003522B1 (en) * 2002-06-24 2006-02-21 Microsoft Corporation System and method for incorporating smart tags in online content
US7818308B2 (en) * 2003-10-01 2010-10-19 Nuance Communications, Inc. System and method for document section segmentation
US20050097080A1 (en) * 2003-10-30 2005-05-05 Kethireddy Amarender R. System and method for automatically locating searched text in an image file
US7707039B2 (en) * 2004-02-15 2010-04-27 Exbiblio B.V. Automatic modification of web pages
US7376642B2 (en) * 2004-03-30 2008-05-20 Microsoft Corporation Integrated full text search system and method
US20070300142A1 (en) * 2005-04-01 2007-12-27 King Martin T Contextual dynamic advertising based upon captured rendered text
US7299972B2 (en) * 2004-04-08 2007-11-27 Adobe Systems Incorporated Creating and using documents with machine-readable codes
US7756869B2 (en) 2004-04-30 2010-07-13 The Boeing Company Methods and apparatus for extracting referential keys from a document
JP4510535B2 (en) * 2004-06-24 2010-07-28 キヤノン株式会社 Image processing apparatus, control method therefor, and program
US20050289447A1 (en) 2004-06-29 2005-12-29 The Boeing Company Systems and methods for generating and storing referential links in a database
US20050289159A1 (en) 2004-06-29 2005-12-29 The Boeing Company Web-enabled real-time link selection apparatus and method
US20050289185A1 (en) 2004-06-29 2005-12-29 The Boeing Company Apparatus and methods for accessing information in database trees
US20050289161A1 (en) 2004-06-29 2005-12-29 The Boeing Company Integrated document directory generator apparatus and methods
US7610274B2 (en) * 2004-07-02 2009-10-27 Canon Kabushiki Kaisha Method, apparatus, and program for retrieving data

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6424982B1 (en) 1999-04-09 2002-07-23 Semio Corporation System and method for parsing a document using one or more break characters
US20020065857A1 (en) 2000-10-04 2002-05-30 Zbigniew Michalewicz System and method for analysis and clustering of documents for search engine

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JENKINS C ET AL: "Adaptive automatic classification on the web", DATABASE AND EXPERT SYSTEMS APPLICATIONS, 2000. PROCEEDINGS. 11TH INTERNATIONAL WORKSHOP ON 4-8 SEPTEMBER 2000, PISCATAWAY, NJ, USA,IEEE, 4 September 2000 (2000-09-04), pages 504 - 511, XP010515542, ISBN: 0-7695-0680-1 *
JENKINS C. ET AL.: "Adaptive Automatic Classification on the Web", DATABASE AND EXPERT SYSTEMS APPLICATIONS, 2000, 4 September 2000 (2000-09-04), pages 504 - 511
JUNG-WON LEE ET AL.: "Preparations for Semantics-Based XML Mining", DATA MINING, 2001. ICDM 2001, 29 November 2001 (2001-11-29), pages 345 - 352
JUNG-WON LEE ET AL: "Preparations for semantics-based XML mining", DATA MINING, 2001. ICDM 2001, PROCEEDINGS IEEE INTERNATIONAL CONFERENCE ON SAN JOSE, CA, USA 29 NOV.-2 DEC. 2001, LOS ALAMITOS, CA, USA,IEEE COMPUT. SOC, US, 29 November 2001 (2001-11-29), pages 345 - 352, XP010583296, ISBN: 0-7695-1119-8 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11341180B2 (en) * 2006-09-29 2022-05-24 Google Llc Displaying search results on a one or two dimensional graph
US10579716B2 (en) 2017-11-06 2020-03-03 Microsoft Technology Licensing, Llc Electronic document content augmentation
US10699065B2 (en) 2017-11-06 2020-06-30 Microsoft Technology Licensing, Llc Electronic document content classification and document type determination
US10909309B2 (en) 2017-11-06 2021-02-02 Microsoft Technology Licensing, Llc Electronic document content extraction and document type determination
US10915695B2 (en) 2017-11-06 2021-02-09 Microsoft Technology Licensing, Llc Electronic document content augmentation
US10984180B2 (en) 2017-11-06 2021-04-20 Microsoft Technology Licensing, Llc Electronic document supplementation with online social networking information
US11640499B2 (en) 2017-12-26 2023-05-02 RELX Inc. Systems, methods and computer program products for mining text documents to identify seminal issues and cases

Also Published As

Publication number Publication date
HK1095902A1 (en) 2007-05-18
EP1745396A1 (en) 2007-01-24
US8060511B2 (en) 2011-11-15
JP2007535771A (en) 2007-12-06
JP4808705B2 (en) 2011-11-02
US7756869B2 (en) 2010-07-13
US20050246351A1 (en) 2005-11-03
EP1745396B1 (en) 2014-11-05
US20100316301A1 (en) 2010-12-16

Similar Documents

Publication Publication Date Title
US8060511B2 (en) Method for extracting referential keys from a document
Constantin et al. PDFX: fully-automated PDF-to-XML conversion of scientific literature
US7958444B2 (en) Visualizing document annotations in the context of the source document
EP1679625B1 (en) Method and apparatus for structuring documents based on layout, content and collection
JP3289968B2 (en) Apparatus and method for electronic document processing
EP0544432B1 (en) Method and apparatus for document processing
US9135249B2 (en) Number sequences detection systems and methods
US7991709B2 (en) Method and apparatus for structuring documents utilizing recognition of an ordered sequence of identifiers
Déjean et al. A system for converting PDF documents into structured XML format
US20080065671A1 (en) Methods and apparatuses for detecting and labeling organizational tables in a document
US20090138466A1 (en) System and Method for Search
US20080114757A1 (en) Versatile page number detector
US20080263023A1 (en) Indexing and search query processing
US20090019015A1 (en) Mathematical expression structured language object search system and search method
JP2005526314A (en) Document structure identifier
JPH11110416A (en) Method and device for retrieving document from data base
US20090046918A1 (en) Systems and methods for notes detection
EP0155284A1 (en) Indexing subject-locating method.
Yurtsever et al. Figure search by text in large scale digital document collections
JP3711636B2 (en) Information retrieval apparatus and method
Myka et al. Automatic hypertext conversion of paper document collections
Souza et al. ARCTIC: metadata extraction from scientific papers in pdf using two-layer CRF
Taghva et al. Evaluation of an automatic markup system
JPH1011443A (en) Document code check system
Shere et al. Identifying and Extracting Hierarchical Information from Business PDF Documents

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2007511052

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

WWW Wipo information: withdrawn in national office

Country of ref document: DE

WWE Wipo information: entry into national phase

Ref document number: 2005742113

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2005742113

Country of ref document: EP