WO2009007897A1

WO2009007897A1 - Method of operating an information retrieval system

Info

Publication number: WO2009007897A1
Application number: PCT/IB2008/052712
Authority: WO
Inventors: Olivier Schneider; Ronald L. C. Koymans
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2007-07-11
Filing date: 2008-07-07
Publication date: 2009-01-15
Also published as: EP2181406A1; CN101743543A

Abstract

A method of operating a system for providing information to a user includes causing user profiles of respective users of the system for providing information to be maintained. A search engine for searching content is caused to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system. An overview of information is caused to be made available to the first user, the information including data pertaining to content identified by the search engine in response to the search query. The information included in the overview is adapted based on a determination of relevance of information included in the user profiles of a plurality of further users to information derived at least in part from the search query.

Description

Method of operating an information retrieval system

FIELD OF THE INVENTION

The invention relates to a method of operating a system for providing information to a user, including: causing user profiles of respective users of the system for providing information to be maintained, causing a search engine for searching content to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system; and causing an overview of information to be made available to the first user, the information including data pertaining to content identified by the search engine in response to the search query.

The invention also relates to a system for providing information to a user, which system is configured to: maintain user profiles of respective users; cause a search engine for searching content to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system; and to make an overview of information available to the first user, the information including data pertaining to content identified by the search engine in response to the search query.

The invention also relates to a computer programme.

BACKGROUND OF THE INVENTION

US 6,691,106 discloses a system and method for building an instant web portal related to topics of interest based on a user profile. The user profile contains information about the user's interests, dislikes, current location, appointment status current tasks and any other information that may be stored on the user's PC or client device. The user profile also contains information on web sites visited, user actions and other details about a user's web activities for a particular category. The information from the user profile may be transmitted to a client-side filter, which contains the rules and instructions for proper transmission of the information to a server computer. A server- side filter receives the information. The server-side filter contains the rules and instructions for proper transmission of the information to the cache-crawler, which is used to retrieve and review the information from web pages and to determine their relevance to the user's query. The user profile may transmit a generic abstract to a server computer to form aggregated profiles from different clients of the same user or from different clients of different users having similar goals and/or behaviours. The information in the aggregated profiles may be transmitted to the cache- crawler to retrieve and review similarly the information from web pages and to determine their relevance to the user's query.

A problem of the known method is that the aggregated profiles may lead to inappropriate search queries and/or filter criteria. This is because, according to his or her profile, a given user may have many disparate interests or activities in common with many different other users. As a result, there is little extra functionality over a normal search engine in terms of personalisation of the portal.

SUMMARY OF THE INVENTION

It is an object of the invention to provide a method, system and computer programme of the types defined in the opening paragraphs that provide a personalised portal with extended functionality relevant to the requirements of the user.

This object is achieved by the method according to the invention, which is characterised by adapting the information included in the overview based on a determination of relevance of information included in the user profiles of a plurality of further users to information derived at least in part from the search query. Because user profiles of further users are matched to information derived at least in part from the search query, user profiles relevant to a particular context are identified. This adapts the method to the context of the interaction of the first user with the system. The functionality is potentially extended in that the information from the user profiles determined to be relevant forms an additional input for personalising the portal. For a given first user profile and search query, more information that is relevant is made available. This information is based on knowledge and/or experience accumulated by other relevant users. Because no further input from the user is required than the information obtained through interaction on which the search query is based, the system is operated in a way that makes it easy to use. In an embodiment, the relevance of information included in the user profiles of a plurality of further users is determined relative to information derived also from information included in the user profile of the first user.

An effect is to identify which of the further users are most like-minded to the first user, as well as being relevant to the search query. The overview of information can therefore be personalised more effectively to the first user, whilst still using the extra information that is available in the user profiles of further users.

In an embodiment, the information included in the overview includes an identification of at least one further user, adapted at least partly in accordance with the determination.

An effect is to extend the functionality of the portal provided by the system for providing information, in that relevant information is shown together with relevant contacts to communicate with. This embodiment increases the sense and simplicity of using the Internet for persons like senior citizens. It allows them to achieve a realisation of a good quality of life, in particular to reduce social isolation. It offers functionality similar to Google (http://www.google.com), Hyves (http://www.hyves.nl) or dating sites, but designed in a senior-centric way. This makes the advantages of these types of functionality easier to experience for seniors. Moreover, the invention offers an assisted way of browsing the Internet and it offers a personalised view on both Internet content and contacts. A striking and novel feature of the concept is the split of both content sources and contacts being displayed on one page, for example ranked according to relevance to the personal profile and the entered query of the user.

The Internet has proven to offer various beneficial tools for younger people to find information on self- fulfilment (e.g. Google, specific web sites), finding people with similar interests (e.g. Hyves (http://www.hyves.nl), MySpace (http://www.myspace.com)), meeting a new partner (via dating sites). However, most seniors are not so agile that they can find their way around on the Internet. Compared to existing senior portals (e.g. seniorweb.nl, seniornet.org, seniorplanet.fr, senior.com) this embodiment offers all of these functionalities, is easy to use, and is attractively designed with respect to lay-out and navigation. A variant of this embodiment includes providing a user interface to the system for providing information, which user interface enables any of the further users identified in the overview to be contacted by clicking on an icon representing that person in the overview.

An effect is to extend the functionality of the system for providing information in a user- friendly manner, by turning it into a portal allowing relevant other users to be identified and communicated with. Further navigation or the use of a separate application or device for interpersonal communication is not required.

A further variant includes providing a user interface enabling a user to switch between a status of available and not available as communicated to other users. An effect is to allow users to participate in an on-line community, and selectively preserve their privacy.

In an embodiment, wherein the information included in the overview includes identifications of a number of further users, the method of operating a system for providing information to a user further includes: ranking further users according to the determination of relevance of information included in the user profiles of the further users, and arranging the identifications of the number of further users in accordance with the ranking.

An effect is to provide a coupling between search/request query and resulting re-arrangement in the communication window with respect to the matches shown and their ranking according to relevance (indicating "experts" on the search/request query). An embodiment of the method includes ranking items of content identified by the search engine in response to the search query at least in part according to relevance to the user profile of the first user, and arranging the data pertaining to content in accordance with the ranking.

An effect is to provide a coupling between search/request query and resulting re-arrangement in an info window with respect to (sub-)categories shown (optionally) and ranking of the content according to relevance.

An embodiment of the method includes ranking items of content identified by the search engine in response to the search query at least in part according to relevance to information included in the user profiles of at least one further user identified in accordance with the determination.

An effect is to increase the likelihood that relevant information pertaining to content is brought to the attention of the first user. Use is made of information maintained in relation to further users who are relevant to the search query and/or similar to the first user, in order to rank items of content identified by a search engine in response to the search query.

In a variant, the ranking is further based on a weighting of the information included in the user profiles of a plurality of further users in accordance with the determination. An effect is further to increase the likelihood that relevant information pertaining to content is brought to the attention of the first user, in that the information from user profiles of the most relevant further users is used has a higher effect on the ranking than information from the user profiles of other further users. An embodiment of the method includes providing a user interface enabling the first user to view selected content by clicking on a link to a source of the content provided in the overview.

An effect is to increase the sense and simplicity of the interface provided by the system for providing information to a user. In an embodiment, a user profile of a user of the system for providing information is adapted continuously as the user interacts with the functionality of the system.

An effect is to provide a coupling between browsing and navigation behaviour and a resulting re-arrangement in the overview of information, for example a re-arrangement of a communication window with respect to the matches shown and their ranking according to relevance (indicating "experts" on the explored content items). This is because the browsing and navigation information can be used to update the user profiles.

According to another aspect of the invention, there is provided a system for providing information to a user, which system is configured to: maintain user profiles of respective users; cause a search engine for searching content to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system; and to make an overview of information available to the first user, the information including data pertaining to content identified by the search engine in response to the search query, wherein the system is configured to adapt the information included in the overview based on a determination of relevance of information included in the user profiles of a plurality of further users to information derived at least in part from the search query.

In an embodiment, the system is configured to execute a method according to the invention. According to another aspect of the invention, there is provided a computer programme including a set of instructions capable, when incorporated in a machine-readable medium, of causing a system having information processing capabilities to perform a method according to the invention. BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be explained in further detail with reference to the accompanying drawings, in which:

Fig. 1 is a schematic diagram of a network including a system for implementing a portal for seniors;

Fig. 2 is a screen view of a home page of the senior portal;

Fig. 3 is a screen view of the home page with a request entered;

Fig. 4 is a screen view of the portal after a user has clicked on a link;

Fig. 5 is a screen view of the portal after a user has clicked on a link shown in the screen view of Fig. 4;

Fig. 6 is a screen view of the portal after a user has clicked on a link shown in the screen view of Fig. 5;

Fig. 7 is a schematic diagram of an embodiment of the system for providing information; Fig. 8 is a schematic flow chart of a method of adapting the page in response to receiving a search query; and

Fig. 9 is a schematic flow chart of a method of adapting the page in response to interaction of the user with the portal.

DETAILED DESCRIPTION

Referring to Fig. 1, a system for providing information to a registered user is hosted by a portal server 1 connected to the Internet 2. The Internet 2 is used as an example of a Wide Area Network, since the description given below could equally apply to a system for providing information to users of a corporate intranet, for example. For illustrative purposes, a single user's computer 3 is shown, although it is to be understood that many users will be connected to the Internet 2 at any one time. The computer 3 is connected to a screen 4, a pointer device 5 such as a mouse, joystick or trackball, and a keyboard 6. Other peripheral devices may also be connected (e.g. a sound system or a user input device of a further type). A browser is installed on the computer 3 for loading and displaying files including data representative of pages providing an overview of information. The data is encoded in a mark-up language such as HTML or xML. The pages of information include active elements having an action associated with them, which can be selected using the pointer device 5. In this manner, a user interface to the system for providing information is provided. Content data is hosted on a web server 7 or on any of a plurality of further web servers (not shown for purposes of clarity). A search engine is hosted on a search engine server 8.

In the following, a description is given of a system for providing information to a subscriber (i.e. a registered user) or group of subscribers to an Internet-based community and information service directed in particular to senior citizens. The same system could be adapted to any other community, e.g. a community of technical specialists, doctors, or other professionals. In the description given below of the system, the senior portal is referred to as WiseNet. A metaphor to communicate the WiseNet concept to seniors is that of a personalised newspaper-style view on the Internet ('Every day there is something or someone new to discover for you on the Internet. Now the Internet is as easy to use as reading a newspaper.'). In the illustration given below the user's first name is Ron. To convey the personalised characteristics of the portal it will show the personalised title 'Ron's WiseNet'.

Senior citizens (50+) are the fastest growing age group in particular in the USA and EU in the next decade. This group has a need to stay connected with their peers. It is a group with a large amount of free time and with above average income available to be spent on quality of life. Moreover, there are some common problems in this age group that complicate full realisation of good quality of life:

In particular when people retire from their professional life, a major issue is how to find alternative ways of belonging, esteem and self- fulfilment (Maslow's hierarchy of needs). Apart from shared interests, most partners also differ in the interests they pursue to fill their spare time. There is therefore a relatively high need to find other people that could help to fill this gap.

For the older group of senior citizens (70+) there is an increased chance of one's partner to die. As a consequence, people need extra social support to overcome the emotional shock, they have to find other people to socialise with on a daily basis and will have to find ways of coping without the skills (e.g. cooking, gardening, financial administration, driving a car) of their deceased partner.

People's decreasing physical skills will make them less mobile, which leads to a higher risk of social isolation.

To use the system for providing information, a user registers his/her identity as a member and completes a questionnaire, preferably at first time use. Optionally the questionnaire may be completed at a later time or may be completed for a second or further time to reflect a change in one's areas of interest. The questionnaire is the basis for the creation of an initial (or optionally an updated) user profile. This user profile is not necessarily visible to the end-user, and may consist of a set of keywords to characterise the user's preferences in terms of content sources, topics and kind of social contacts. This profile will be adapted continuously as the user interacts with the functionality of the portal e.g. through browsing, entering queries, and engaging in chat sessions with other members of the portal.

Referring to Figs. 2-6 a home page represents the contents of an overview 9 that is initially shown by the Wisenet portal. An upper half 10 of the WiseNet portal shows relevant information to the users. A lower part 11 of the portal shows relevant contacts to communicate with. In between these two there is a 'MyRequest' bar 12 where the user can type his/her request to have support on (see Fig. 3). Once the request is entered, the contents of both the upper and lower half 10,11 of the screen will be ranked in order of relevance to the request and the personal profile of the user. In case the MyRequest bar 12 is left blank, the content is only ranked based on the personal profile of the user.

The information in the upper part 10 is structured into high-level categories (see Fig. 3). The categories that contain relevant sources are coloured green or otherwise visually distinguished (see Fig. 3). By clicking on one of these categories, the next level of information structuring appears as sub-categories (see Fig. 4) and below these a list of ranked information sources. When one of these sources is clicked on, one gets at the lowest level of information sources structuring as a ranked list (see Fig. 5). Clicking on one of these listed items, finally, shows the content of the information source (see Fig. 6).

MyContacts is a list 13 of WiseNet members created by the user to have more easy access to their favourite contacts. MyNewContacts is a list 14 of contacts that are ranked according to relevance to the user's profile and the entered request, and have not been added to the MyContacts list 13 yet. Contacts highlighted in green are on-line. The contacts coloured blue are currently off-line. In other embodiments, a different technique is used to display the status of the contacts identified in the lower half 11 of the overview 9. A MyConversations section 15 is activated when a contact from one of the previously mentioned lists 13,14 is clicked on. In case of an on-line contact a live two-way chat-like communication will be shown in the MyConversations window 15. In case of contacting a person who is off-line, only the user's message to be sent to that contact will be shown, or received messages of that contact person will be shown.

Another user is obviously able to contact the user (Ron) to whom the overview 9 is provided. In an embodiment, an overview is provided showing incoming communication requests as blinking icons (not shown) in the list 13 of contacts. One implementation of WiseNet provides a ranked list of matching contacts with a top-five of best matches. In case an incoming requester is not in that top-five, the corresponding icon will appear at the sixth position in rank. In another embodiment of the portal, an "Availability Button" (not shown) comprises a toggle button enabling the user to switch between available and not available status of the user as communicated to other members of the portal. Another embodiment of the portal provides a mechanism to present incoming requests in a convenient way to the receiving user, for instance: filtering: on the basis of a user-determined threshold (e.g. based on content of request, quality of the match with the profile of the requester, desired number of requests per time unit), a limited number of requests are put through to the communication window; or feedback: via a user interaction mechanism (e.g. slider, Lickert scale, thumbs up/down) the user can influence the threshold level of filtering (cf. TiVo).

In yet another embodiment, features of at least two of these variants are combined. Summarising, the user may view the selected content by clicking (in the usual way) on a link in the top-five shown in Fig. 5. One can contact a new person by clicking on the icon representing that person in the top-five 'MyNewContacts' list 14 shown.

Besides using the content already available in the portal, the user may add new content to the portal. Various embodiments of the system for providing information provide at least one of the following three ways of doing this: submitting a hyperlink to existing Internet content; submitting a piece of content in a certain file format (e.g. in Adobe's Portable Document Format (PDF), Microsoft's MS Word format (.doc), or the Joint Photographic Experts Group's picture format (.jpeg) for graphic content); - adding a comment to content (Internet or file content) that is accessible within the portal.

Yet another embodiment of the system for providing information to a subscriber provides for annotation of existing web sites at specific locations with comments similar to Fleck.com (http://fleck.com), but instead of displaying the text of the comment at that specific position only a number is shown. The comment itself is displayed in a separate numbered comments section (not shown).

Yet another embodiment of the system provides an option that allows one generally to comment on an existing web site, by adding a comment to a comments section (not shown) below the viewed web site. The comments section could also be located next to a portion of the screen reserved for showing the viewed web site.

Yet another embodiment of the system for providing information provides that if the user navigates to an external web site, this web site is displayed in a frame (not shown) within the portal. In an embodiment, comments made by other users in relation to that web site are shown below or next to the web site frame ranked in order of relevance to a submitted search/request query and the user profile.

Figs. 2-7 illustrate sequential steps in the use case of a person who wants to visit the Rijksmuseum in Amsterdam together with another - initially unknown - member of the WiseNet community. In this use case the following steps are taken

1. User 'A' wants to visit the Rijksmuseum and types "who likes to visit the Rijksmuseum with me" in the 'Request bar' 12.

2. A 'key words and key phrases extraction' process 16 distils the words/phrases 'visit', 'Rijksmuseum' and 'with me' from the phrase above. 3. These three words/phrases are submitted (process 17) to a generic search engine (e.g. Yahoo! or Google). The result is a list 18 of query results, ranked by the search engine's own standard.

4. Parallel to step 3, the extracted key words and phrases from step 2, are used to obtain profiles 19 of like-minded people in the WiseNet portal community (process 20). This means like-mindedness based on both the current user profile of the end-user and the user profiles of people that are very close/similar. This similarity means: firstly, matching with the submitted query (step 2) and secondly, matching the user profile of the first user. So, in this specific use case, people within the WiseNet community with expert knowledge on the Rijksmuseum qualify for matching as well as those that have visited the Rijksmuseum before. In particular, the best match would be people with expertise, who visited the Rijksmuseum before and are willing to join for a new visit with someone. All contacts in the 'Communication' user interface are ranked according to their level of similarity to the request. Moreover, via 'Smart content ranking & matching' algorithms 21, the content items of an 'Info' user interface of which the output corresponds to the upper half 10 of the overview 9 are ranked in order of relevance to the user request submitted in step 1.

Step 1 as described above may be implemented in the following ways: Ia. The user simply clicks on one of the categories in WiseNet. For example, in the case of info on the Rijksmuseum, the user clicks on 'Culture & Recreations'. This click will cause both a new screen with subcategories and ranked content items per selected subcategory (according to the relevance of the user) to be displayed. Moreover, the contacts in the 'Communication' user interface will also be ranked differently, again according to the relevance to the selected category, sub-category and content item. Ib. The user can simply click on one of the contacts in the 'Communication' interface e.g. in the 'My Contacts' and 'MyNewContacts' lists 13,14. This will cause a change of info ranking as in step Ia. Moreover, whenever a selected contact answers a communication request (caused by clicking on one of the contacts), a chat session can be opened. The contents of the chat session may serve as input to refinement/extension of the user profile of user 'A'. In addition or alternatively it may also serve to improve the query- results process similar to the description in step 4 above.

Fig. 7 will now be explained in slightly more detail. In the shown embodiment, words and key phrases are extracted from a user's action such as typing phrases in a request bar or navigating through categories shown to the user. The behaviour of the system depends on a user interface component 22.

Information 26 extracted from the user interface 22 preferably comprises information extracted from an Info user interface 23 (of which the upper half is the visible output component), information extracted from a request bar user interface 24 (of which the request bar 12 is the visible component), or information extracted from the communication user interface 25 (corresponding to the lower part 11), or a combination of two or more of these.

With regard to information extracted from the Info user interface 23, information is extracted from or parsed from the links (Uniform Resource Locators) clicked upon or activated in another way by the user or from the corresponding web content subsequently shown to the user, or both (process 16). Meaningful key words and/or phrases are extracted by techniques such as or similar to NLP (Natural Language Parsing). Other relevant user interaction parameters, such as for instance the length of stay of a user on a web site or the ranking of the presented search result that is selected by the user, are also comprised in the information extracted from the Info user interface 23.

With regard to the information extracted from the request bar user interface 24, relevant key words and/or phrases are extracted from the text typed in by the user in the request bar, in a similar way as with the Info user interface 23.

With regard to the information extracted from the communication user interface, meaningful key words and/or phrases are extracted from all or part of the textual content captured by the communication user interface 25 (e.g. typed text, speech) by techniques similar to the ones used in the Info user interface 23 or Request bar user interface 24.

The search engine query is derived (process 17) from the extracted key words and/or phrases 26. The query is preferable submitted to a known internet search engine such as Google (http://www.google.com) or Yahoo! (http://www.yahoo.com), or a search engine incorporated in the system comprising the portal server 1 (Fig. 1), or a combination of two or more of such search engines. The resulting list of query results found with this search engine or combination of search engines is used in a next step, in which the search results are ranked according to their relevance with respect to user interaction parameters, a user profile 27 of the current user and user profiles 19 of like-minded users (other users having a profile that matches or partially matches with the profile of the user).

In a process of adapting the user profile 27, the key words and/or phrases 26 extracted from the user input are added to the user profile 27. This is preferably done in such a way that the user profile 27 reflects the importance of these key words and/or phrases to specific users. E.g. infrequently used key words and/or phrases will be categorized as less interesting than frequently used ones. In this adaptation, key words and/or phrases that have not been used for a long time may also be deleted from the user profile. In another embodiment, key words and/or phrases are associated with weights, the values of the weights changing to decrease the importance of the key word and/or phrase each time it is not used. The decrease could be linear or exponential.

In an embodiment, user interaction parameters (e.g. the length of stay of a user on a web site, the ranking of the presented search result selected by the user) are stored in the user profile 27.

The process 20 of profile matching entails that, within profiles of a plurality or all of the users (or members of the community), those users are selected that have the best match with the user profile 27 in the context represented by the extracted key words and/or phrases 26.

The smart content ranking and matching algorithms 21 involve re-ranking and, if applicable, merging query results for the used search engine or engines in dependence of one or more of the following:

-user interaction parameters;

-user profile 27; and

-profiles of like minded (as defined above). This merging and re-ranking results in a list 28 of peer (or like minded) ranked search results. Turning to Fig. 8, a variant of the method described above is illustrated in more detail in the form of a flow chart. This chart shows the response of the portal server 1 to a search query entered in the appropriate field in the request bar 12.

In a first step 29, the portal server 1 receives the search query, e.g. in the form of an http (hypertext transfer protocol) message sent by the computer 3 of a first user over the Internet 2. Concurrently, or in advance, the identity of the first user is made known to the portal server 1.

The portal server 1 causes a user profile 30 of the first user to be maintained as a file on the portal server 1 and/or the computer 3. In an embodiment, the user profile 30 consists of two parts: a part that is derived from information entered by the user at a first time of use of the portal (e.g. age, gender, hobbies, willingness to do things together etc.); and another part that is continuously built up through all user behaviour in the portal (searches, requests, chat sessions, etc.). In an embodiment, the profile comprises index words or key words that are derived from either part or from both parts. An embodiment of the invention provides a generic ontology for all users that is used by the portal and continuously refined.

In the illustrated embodiment, a file 31 representing the current context of the session of use of the Wisenet portal by the first user is also maintained. The current context is determined, amongst others, by the category of information being viewed in the upper half 10 of the overview 9, and is based on user interaction parameters.

User profiles of a plurality of further users of the portal are stored in a database 37. Information included in these user profiles is searched to identify a set 38 of user profiles including information similar to information derived in part from the search query and in part from the information included in the user profile 30 of the first user (step 39).

Preferably, information more relevant for a particular user (determined by his profile) is presented in a more prominent way, as will be explained. This is also determined by user profiles ("folksonomy", also cf. Amazon.com-like recommendations) with the best match in the current context. For example, two users Olivier and Ron may share employer and music interest & advanced play. So, on a general profile level there is a good match. When going to the music category, they still match well, but as soon as Olivier goes to jazz, Ron will be less prominent on his list, being replaced by someone with more interest in jazz (and not necessarily employed by the same employer). Subsequent to receiving the search query, the portal server 1 causes the file 31 (step 32) and the user profile 30 (step 33) to be updated. Key words and phrases are added to the user profile 30. This is done in such a way that the user profile 30 reflects the importance of these key words and phrases to the user. Simultaneously, the portal server 1 formulates a search query associated with the first user and submits it to at least one search engine (step 34) in a manner similar to that outlined above with regard to Fig. 7. In response, it receives (step 35) search results 36 including data pertaining to content identified as relevant by the search engine in response to the search query. In a next step 40, the further users whose user profiles were included in the set 38 are ranked according to the determination of how relevant the information included in these user profiles is to the information based on the current search query, the user profile 30 and the current context. A sequence 41 of further user profiles results from this step 40.

In the same step 40, the items of content identified in the search results 36 are ranked according to relevance to the user profile 30, the current context and the information included in the user profiles in the sequence 41. Information included in the user profiles in the sequence 41 is weighted according to the positions of the user profiles in the sequence 41. Thus, for example, if the most relevant user has visited a particular web site five times, and another user has visited another web site five times as well, then the particular web site is ranked higher than the other web site. The result is a ranked set 42 of information including data pertaining to content identified by the search engine, e.g. ranked links to files stored on the web server 7.

Finally (step 43), the overview 9 is adapted by (re-)arranging the lists 13,14 of further users to reflect the sequence 41, and by (re-)arranging at least one of the (sub-)- category windows and the lists of links provided in the upper half 10 of the overview to reflect the set 42 of ranked information. Thus, the portal provides a coupling between search/request query and resulting re-arrangement in the info window comprised in the upper half 10 of the overview 9, with respect to (sub-)categories shown and ranking of the content according to relevance. The portal also provides a coupling between search/request query and resulting re-arrangement in the communication window (comprised in the lower part 11 of the overview 9) with respect to the matches shown and their ranking according to relevance (indicating "experts" on the search/request query). All information - (sub- categories, the content, and contacts shown in the portal - is ranked according to the relevance to information based at least in part on the user profile. In an example implementation, the portal will, after each submitted request, show the top-five best matching content items and top-five contacts.

The system for providing information to a user also provides a coupling between browsing and navigation behaviour and resulting re-arrangement in the communication window with respect to the matches shown and their ranking according to relevance (indicating "experts" on the explored content items), as is illustrated in Fig. 9.

In a first step 44, information is obtained as a function of user interaction with the user interface provided by the system. For example, information can be extracted from or parsed from the links (Uniform Resource Locators) clicked upon or activated in another way by the user or from the corresponding web content subsequently shown to the user, or both. Other relevant user interaction parameters include, for instance, the length of stay of a user on a web site or the ranking of the presented search result that is selected by the user. In yet another embodiment, the user may express his or her (dis-)satisfaction with returned search results or newly suggested contacts via a 'thumbs up '/'thumbs down' feedback button. This will influence the personal user profile 30, as well as the ranking of the content or person under consideration for other members of the portal with similar user profiles and submitting a similar request. In an embodiment, the user interface provides pre-defined suggestions for search queries. For example, for certain (sub-)categories that can be shown in the upper half 10 of the overview 9, the same commonly asked questions exist. For the category 'Culture & recreations', these include: 'How to get there?'. For such a set of cases, the system offers suggestions e.g. a link to public transportation sites, map sites, route planners. How the user interacts with these menu options constitutes information obtained through interaction of the user with the functionality of the system, which information is collected in the first step 44 shown in Fig. 9. The portal server 1 adapts (step 45) the user profile 30 and/or the current context 31 according to the information obtained in the preceding step 44. Next (step 46), it obtains a new set 38 of user profiles of further users, that it determines include information relevant to information based at least in part on the updated user profile 30 and the current context 31. If information is still present in the request bar 12, then relevance is determined with respect to information based also on the contents of the request bar 12.

The portal server 1 also identifies the current search results 36 (step 47). In one embodiment, a new search query is formulated and submitted to one or more search engines, based on information included in the user profile 30. If a search query is present in the request bar 12, then that search query is used instead or additionally. In another embodiment, a stored copy of previously obtained search results 36 is retrieved.

In a next step 48, the further users whose user profiles are included in the set 38 are ranked according to the determination of relevance of information included in the user profiles of the further users. Thus, the sequence 41 of user profiles is re-arranged. The items of content identified in the search results 36 are also ranked according to relevance to the user profile 30 of the first user and according to relevance to information included in the user profiles of the re-arranged ranked sequence 41, the ranking of information in the set 42 being based on a weighting of the information included in the user profiles of the ranked sequence 41.

Finally, the overview 9 is re-arranged to reflect the new composition of the set 42 of ranked information pertaining to content identified by the search engine and the new composition of the ranked sequence 41 of further user profiles.

It should be noted that the embodiments described above illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. Use of the verb "comprise" and its conjugations does not exclude the presence of elements or steps other than those stated in a claim. The article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

For example, the file 31 representing the current context can be combined with the user profile 30 in one data structure.

Claims

CLAIMS:

1. Method of operating a system for providing information to a user, including: causing user profiles of respective users of the system for providing information to be maintained, causing a search engine for searching content to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system; and causing an overview of information to be made available to the first user, the information including data pertaining to content identified by the search engine in response to the search query, characterised by adapting the information included in the overview based on a determination of relevance of information included in the user profiles of a plurality of further users to information derived at least in part from the search query.

2. Method according to any one of the preceding claims, wherein the relevance of information included in the user profiles of a plurality of further users is determined relative to information derived also from information included in the user profile of the first user.

3. Method according to claim 1 or 2, wherein the information included in the overview includes an identification of at least one further user, adapted at least partly in accordance with the determination.

4. Method according to claim 3, including providing a user interface to the system for providing information, which user interface enables any of the further users identified in the overview to be contacted by clicking on an icon representing that person in the overview.

5. Method according to claim 4, including providing a user interface enabling a user to switch between a status of available and not available as communicated to other users.

6. Method according to any one of claims 3-5, wherein the information included in the overview includes identifications of a number of further users, further including: ranking further users according to the determination of relevance of information included in the user profiles of the further users, and arranging the identifications of the number of further users in accordance with the ranking.

7. Method according to any one of the preceding claims, including ranking items of content identified by the search engine in response to the search query at least in part according to relevance to the user profile of the first user, and arranging the data pertaining to content in accordance with the ranking.

8. Method according to any one of the preceding claims, including ranking items of content identified by the search engine in response to the search query at least in part according to relevance to information included in the user profiles of at least one further user identified in accordance with the determination.

9. Method according to claim 8, wherein the ranking is further based on a weighting of the information included in the user profiles of a plurality of further users in accordance with the determination.

10. Method according to any one of the preceding claims, including providing a user interface enabling the first user to view selected content by clicking on a link to a source of the content provided in the overview.

11. Method according to any one of the preceding claims, wherein a user profile of a user of the system for providing information is adapted continuously as the user interacts with the functionality of the system.

12. System for providing information to a user, which system is configured to: maintain user profiles of respective users; cause a search engine for searching content to process a search query associated with a first user, wherein the search query is based at least in part on information obtained through interaction of the first user with the functionality of the system; and to make an overview of information available to the first user, the information including data pertaining to content identified by the search engine in response to the search query, characterised in that the system is configured to adapt the information included in the overview based on a determination of relevance of information included in the user profiles of a plurality of further users to information derived at least in part from the search query.

13. System according to claim 12, configured to execute a method according to any one of claims 1-11.

14. Computer programme including a set of instructions capable, when incorporated in a machine-readable medium, of causing a system having information processing capabilities to perform a method according to any one of claims 1-11.