AU2005202353B2

AU2005202353B2 - Methods and apparatus for storing and retrieving knowledge

Info

Publication number: AU2005202353B2
Application number: AU2005202353A
Authority: AU
Inventors: Trevor Bryan Davies; Mark Hanlon; Chris Tromans
Original assignee: HYPERKNOWLEDGE MAN SERVICES AG
Current assignee: HYPERKNOWLEDGE MANAGEMENT SERVICES AG
Priority date: 1999-11-23
Filing date: 2005-05-31
Publication date: 2007-05-17
Anticipated expiration: 2020-11-22
Also published as: AU2005202353A1

Description

S The Commissioner of Patents F B RICE CO PO Box 200 Melbourne VIC S WODEN ACT 2606 Telephone: 03 8618 4100 eFacsimile: 03 8618 4199 Customer No: 2410000452 12 April 2007 Our Ref: 123600 SContact: Tracey Webb n twebb@fbrice.com.au C€3 PATENTS

AUSTRALIA

tt Patents Act 1990 sSa\ ci REQUEST TO AMEND A COMPLETE SPECIFICATION 2 P, Hyperknowledge Management Services AG Australian Patent Application 2005202353 (Divisional of AU 17978/01) Entitled: "Methods and apparatus for storing and retrieving knowledge" Hyperknowledge Management Services AG, of Sulgeneckstrasse 35, 3007 Bern, Switzerland; request the Commissioner for leave to amend the complete specification relating to 2005202353 in accordance with the accompanying Statement of Proposed Amendments responsive to the objections raised in the first report of Examiner Jyott Shamdasani. We also enclose marked up pages that indicate the location of the proposed amendments on the replacement pages.

An executed Notice of Entitlement is enclosed.

In the proposed amendments the following claims (currently on file) have been deleted: Dependent claims 7 and 8, independent claim 23 and dependent claims 24 to 30 thereon, dependent claims 37 to 41, independent claim 44, independent claim 45 and dependent claim 46 thereon, dependent claims 52 to 54 and dependent claims 57 to 67.

In item 2 of the report, the Examiner objected that the use of the term "complementary relationships" in claims 1, 16, 31,43, 45 and 47 was vague and therefore unclear. Furthermore in the Examiner's opinion, claims 14, 21, 23, 46 and 55 did not fully define the invention.

With regard to the remaining, renumbered claims, Independent claims 1, 7 (formerly claim 12 (formerly claim 14), 16 (formerly claim 14), 21 (formerly claim 31), 28 (formerly claim 43), 29 (formerly claim 47), and 33 (formerly claim 55) have been amended to more clearly define the invention.

We remind the Examiner that where there is perceived a lack of clarity it is necessary to consider all the disclosure of the specification in order to attempt to resolve that lack of clarity, and where there is doubt about the scope of the phrase the applicant should have the right to make a clarifying amendment to change what was unclear into what is his/her intended scope. We respectfully submit The Commissioner of Patents Our Ref: 123600 12 April 2007 that, as is clearly evident from the specification "complementary relationships" are those in the nature of opposites. See for example lines 29 of page 1, where it is stated that "relationships indicated by opposing directions are complementary" Examples, only, are given as North-South (What is-What) and East-West (how-Why). Nevertheless, to render the claims clear, the term "complementary" has been replaced with the term "opposing".

Moreover independent claims 1, 7, 12, 14 and 33 have further been amended to include the feature that the linked knowledge model defines at least two noun/verb phrase pairs linked by a pair of opposing relationships. It is thus respectfully submitted that each independent claim fully defines the invention. With respect, the feature noted by the Examiner, "a first noun/verb phrase lacking a second noun/verb phrase pair related by at least one of the two complimentary relationships defined between the noun/verb phrase pairs" is an optional feature (as claimed in dependent claims 13 and 20) and is therefore not an essential feature of the invention.

In light of the above amendments, the applicants respectfully submit that none of the cited art documents, Wu et. Al. (KDA: A knowledge base database assistant), Anick (Integrating natural language processing and information retrieval in a trouble shooting help desk") or Hedin (US 5,386,556) teach or suggest a system or method for storing and/or retrieving knowledge, or a software product including a medium on which is encoded a sequence of software instructions which when executed direct performance of a method, each having a linked knowledge model which defines at least two noun/verb phrase pairs linked by a pair of opposing relationships. In the absence of any teaching towards such a feature it is respectfully submitted that each independent claim and dependent claims thereon are novel and inventive in light of the cited art.

In light of the proposed amendments we respectfully request that the Examiner reconsider his objections.

Finally minor amendments have been made the specification on pages 2 through 6 to bring the language into conformity with the changes made to the claims.

It is submitted that the objections raised have now been met and we trust that this application is now in order for acceptance.

Addressfor Service and Correspondence F B RICE CO in Australia: Patent and Trade Mark Attorneys Level 23, 200 Queen St Melbourne VIC 3000 Attorney Code: RV Hyperknowledge Management Services AG Registered Patent Attorney The Commissioner of Patents PO Box 200 WODEN ACT 2606 F B RICE CO Melbourne VIC Telephone: 03 8618 4100 Facsimile: 03 8618 4199 Customer No: 2410000452 12 April 2007

PATENTS

Our Ref: 123600 Contact: Tracey Webb twebb@fbrice.com.au

AUSTRALIA

Patents Act 1990 FIRST STATEMENT OF PROPOSED AMENDMENTS Hyperknowledge Management Services AG Australian Patent Application 2005202353 (Divisional of AU 17978/01) Entitled: "Methods and apparatus for storing and retrieving knowledge" In the Description on File: Cancel pages 2 to 6 on file and replace with new enclosed pages 2 to 6.

In the Claims on File: Cancel the claim pages on file and replace with new enclosed claim pages.

Summary of the Invention Embodiments of the present invention help a user to find knowledge pertinent to a user, rather than finding unconnected, raw information. Knowledge is used herein to refer to information which has been organized according to context and meaning, while information is used herein to refer to raw information which is unorganized or organized according to 'other criteria, such as input format, vocabulary used, source, etc.

To find knowledge, embodiments of the present invention receive information as natural language and produce therefrom a linked knowledge model. The linked knowledge model can then be navigated to extract knowledge, according to the relationships "How", "Why", "What" and "What is" discussed above, using a user interface, which visually facilitates making the connections between pieces of information.

Embodiments of the present invention can increase a user's knowledge by augmenting, the knowledge base available to the user in the form of a linked knowledge model. Gaps in the linked knowledge model and edges or boundaries of the linked knowledge model are susceptible to automatic detection. Embodiments of the invention can therefore detect such gaps and edges and search for missing information on an ad hoc, regularly scheduled or other basis.

Some embodiments of the invention can include user interfaces which automatically analyze the knowledge sought in inquires, for example received by email, and prepare responses to those inquiries based on the contents of a dynamic linked knowledge model accessible to the user. Moreover, the contents of queries and other received information can be related to the information already contained in the linked knowledge model, to expand the model, automatically. The linked knowledge model can also be presented in various useful formats, depending on the type of knowledge sought.

According to one aspect of the invention, there is a system for storing and retrieving knowledge, comprising a text analyzer having an input which receives information as natural language and which produces at an output annotated information, a parser having an input which receives the annotated information from the text analyzer and which produces and stores in a memory a linked knowledge model thereof, the linked knowledge model defining at least two noun/verb phrase pairs linked by a pair of opposing relationships, and a user interface presenting at least one pair of opposing (complementary) relationships to an operator, through which the operator can navigate the linked knowledge model to retrieve knowledge. The system can further comprise a knowledge research agent having an input receiving the linked knowledge model and having an output on which the knowledge research agent produces a query compatible with an external search engine. The Hyperknowledge agent may automatically augment the linked knowledge model with new information found by executing the query on the external search engine. In the system, the parser can further comprise a Rules Engine executing a plurality of rules defining a structure of knowledge embedded in information expressed in natural language. The linked knowledge model may include connected noun/verb phrase pairs, and the plurality of rules may define how sentences in natural language relate to connected noun/verb phrase pairs in the linked knowledge model. The rules may include at least one semantic rule, at least one syntactic rule or at least one context rule.

According to a second aspect of the invention, there is a method for retrieving knowledge. The method comprises presenting to an operator a selector by which the operator indicates one of a pair of opposing relationships and a query input by which the operator indicates query text, and combining the one of the pair of opposing relationships indicated and the query text to form a query by which knowledge is retrieved whereby the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships. According to the method of this aspect of the invention, the selector may further define four directions characterized by opposing pairs of directions, each direction representing a relationship. The selector may further be presented in a graphical form resembling a compass rose. The method may further comprise storing knowledge in a linked knowledge model including noun/verb phrase pairs, searching the linked knowledge model for a first noun/verb phrase pair defined by the query text, and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of opposing relationships to the first noun/verb phrase pair. Finally, searching may further comprise parsing a natural language query text into a searchable Boolean expression.

According to a third aspect of the invention, there is a method for storing knowledge, comprising parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs, linking the at least two connected noun/verb phrase pairs to a pair of opposing relationships, identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model, and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model. According to this method, the identified portions of the linked knowledge model may further include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two opposing relationships defined between the noun/verb phrase pairs.

According to other aspects of the invention, each of the foregoing aspects of the invention may be embodied as a software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct the performance of one of the methods discussed above, or which cause a general purpose computer to act as a special purpose system as defined above.

For instance, according to a fourth aspect, the invention is a software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: presenting to an operator a selector by which the operator indicates one of a pair of opposing relationships and a query input by which the operator indicates query text; and combining the one of the pair of opposing relationships indicated and the query text to form a query by which knowledge is retrieved, whereby the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships.

In a fifth aspect, the invention is a software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model, where at least two noun/verb phrase pairs are linked by a pair of opposing relationships.

In a sixth aspect, the invention is a software product including a machine readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of opposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts making available to the operator and presenting to the operator, wherein in act of making available to the operator the multitude of knowledge sets is selected from those that are in opposing relationships with at least one of already selected knowledge sets.

In a seventh aspect, the invention is a system for retrieving knowledge, comprising: a repository of representations of relationships between at least two units of information; and a user interface for accessing at least one representation of at least one relationship between the at least two units of information, wherein the user interface is further adapted to provide access to subsequent relationships.

In a eighth aspect, the invention is a method for knowledge retrieval, comprising: creating a database of pairs of opposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator a selector and presenting to the operator the at least one knowledge set, wherein in the making available to the operator, multitude of knowledge sets is selected from those that are in opposing relationships with at least one of already selected knowledge sets.

In a ninth aspect, the invention is a machine-readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of opposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator and presenting to the operator, wherein in act of snaking available to the operator, the multitude of knowledge sets is selected from those that are in opposing relationships with at least one of already selected knowledge sets.

In a tenth aspect, the invention is a method of doing business, comprising: creating a database of pairs of opposing relationships between knowledge sets; providing access to the database of pairs of the opposing relationships between the knowledge sets, wherein an entity obtaining access to the database is adapted to executing a second method, comprising: presenting to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of providing access to the database and presenting to the operator, wherein in act of presenting to the operator, the multitude of knowledge sets is selected from those that are in opposing relationships with at least one of already selected knowledge sets.

In a eleventh aspect, the invention is a method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: providing a repository of relationships between at least two units of information; identifying at least one relationship relevant to the document from the relationships between at least two units of information; and displaying a user interface for accessing at least one representation of the at least one relationship relevant to the document, wherein the user interface is further adapted to provide access to subsequent relationships.

In a twelfth aspect, the invention is a method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: selecting a token from the document; retrieving from the repository of knowledge pairs of opposing relationships between knowledge sets that are relevant to the token selected from the document; identifying at least one pair of complimentary relationships relevant to the token selected from the document if any opposing relationships were retrieved during the act of retrieving; and generating a visual representation of the at least one pair of opposing relationships.

In a thirteenth aspect, the invention is a system for retrieving knowledge relevant to a document, comprising: THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:- I. A system for storing and retrieving knowledge, comprising: a text analyzer having an input which receives information as natural language and which produces at an output annotated information; a parser having an input which receives the annotated information from the text analyzer and which produces and stores in a memory a linked knowledge model thereof, the linked knowledge model defining at least two noun/verb phrase pairs linked by a pair of opposing relationships; and a user interface presenting at least one pair of opposing relationships to an operator, through which the operator can navigate the linked knowledge model to retrieve knowledge.

2. The system of claim 1, further comprising a knowledge research agent having an input receiving the linked knowledge model and having an output on which the knowledge research agent produces a query compatible with an external search engine.

3. The system of claim 2, wherein the knowledge research agent automatically augments the linked knowledge model with new information found by executing the query on the external search engine.

4. The system of claim 1, the parser further comprising a rules engine executing a plurality of rules defining a structure of knowledge embedded in information expressed in natural language.

The system of claim 4, the linked knowledge model including connected noun/verb phrase pairs, and wherein the plurality of rules define how sentences in natural language relate to connected noun/verb phrase pairs in the linked knowledge model.

6. The system of claim 5, wherein the rules includes at least one semantic rule, or at least one syntactic rule or at least one context rule.

7. A method for retrieving knowledge, comprising: presenting to an operator a selector by which the operator indicates one of a pair of opposing relationships and a query input by which the operator indicates query text; and combining the one of the pair of opposing relationships indicated and the query text to form a query by which knowledge is retrieved, whereby the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships.

8. The method of claim 7, wherein the selector defines four directions characterized by opposing pairs of directions, each direction representing a relationship.

9. The method of claim 8, wherein the selector is presented in a graphical form resembling a compass rose.

The method according to any one of the preceding claims, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searching the linked knowledge model for a first noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of opposing relationships to the first noun/verb phrase pair.

11. The method of claim 10, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

12. A method for storing knowledge, comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; linking the at least two connected noun/verb phrase pairs to a pair of opposing relationships; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model.

13. The method of claim 12, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two opposing relationships defined between noun/verb phrase pairs.

14. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: presenting to an operator a selector by which the operator indicates one of a pair of opposing relationships and a query input by which the operator indicates query text; and combining the one of the pair of opposing relationships indicated and the query text to form a query by which knowledge is retrieved whereby, the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships.

15. The software product of claim 14, wherein the selector defines four directions characterized by opposing pairs of directions, each direction representing a relationship.

16. The software product of claim 14, wherein the selector is presented in a graphical form resembling a compass rose.

17. The software product of any one of claims 14 to 16, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searching the linked knowledge model for a fast noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of opposing relationships to the first noun/verb phrase pair.

18. The software product of any one of claims 14 to 17, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

19. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model, where at least two noun/verb phrase pairs are linked by a pair of opposing relationships.

The software product of claim 19, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two opposing relationships defined between noun/verb phrase pairs.

21. A method for knowledge retrieval, comprising: creating a database of pairs of opposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator a selector and presenting to the operator the at least one knowledge set, wherein in the making available to the operator, multitude of knowledge sets are selected from those that are in opposing relationships with at least one of already selected knowledge sets.

22. The method of claim 21, wherein the selector represents at least one relationship characterized by a opposing pair of directions.

23. The method of claim 22, wherein the selector defines two relationships represented by two axes, each axis representing a relationship.

24. The method of claim 22, wherein the selector is presented in a graphical form representing a compass rose.

The method of claim 22, further comprising presenting to the operator the selector when the operator positions a representation of a pointing device over a representation of a link enabling the knowledge retrieval.

26. The method of claim 25, further comprising presenting to the operator a second selector, by which the operator can select a noun/verb pair corresponding to at least one of the relationships retrieved from the database.

27. The method of any one of claims 22 to 26, further comprising: selecting, by the operator, at least one of the previously selected knowledge sets; and displaying to the operator the at least one of the previously selected knowledge sets and at least one knowledge set related to the at least one of the previously selected knowledge sets.

28. A software product including a machine readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of opposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts making available to the operator and presenting to the operator, wherein in act of making available to the operator the multitude of knowledge sets is selected from those that are in opposing relationships with at least one of already selected knowledge sets.

29. A method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: selecting a token from the document; retrieving from the repository of knowledge, pairs of opposing relationships between knowledge sets that are relevant to the token selected from the document; identifying at least one pair of opposing relationships relevant to the token selected from the document if any opposing relationships were retrieved during the act of retrieving; and generating a visual representation of the at least one pair of opposing relationships.

30. The method of claim 29, further comprising: accumulating pairs of relevant opposing relationships by repeating acts of selecting, retrieving, and identifying for additional tokens in the document; and generating a visual representation of at least one relevant opposing relationship from the pairs of relevant opposing relationships.

31. The method of claim 30, further comprising removing duplicates from the pairs of relevant opposing relationships.

32. The method of claim 30, wherein the act of generating the visual representation further comprises generating a visual representation of a link between the at least one relevant opposing relationship and a token to which it is relevant.

33. A system for retrieving knowledge relevant to a document, comprising: a repository of representations of opposing relationships between at least two units of information, where each unit of information is definable by a noun/verb phrase pair; a knowledge matcher identifying at least one opposing relationship relevant to the document from the opposing relationships between the at least two units of information; and a user interface enabling access to at least one representation of the at least one opposing relationship relevant to the document, wherein the user interface is further adapted to provide access to subsequent opposing relationships.

34. The system of claim 33, wherein the mechanism for identifying the at least one opposing relationship relevant to the document further comprises a module for 29 identifying at least one opposing relationship relevant to a token in the document based on a context of the word.

Dated this twelfth day of April 2007 Hyperknowledge Management Services

AG

Patent Attorneys for the Applicant: F B RICE CO MARKED-UP COPY 2 Summary of the Invention Embodiments of the present invention help a user to find knowledge pertinent to a user, rather than finding unconnected, raw information. Knowledge is used herein to refer to information which has been organized according to context and meaning, while information is used herein to refer to raw information which is unorganized or organized according to other criteria, such as input format, vocabulary used, source, etc.

To find knowledge, embodiments of the present invention receive information as natural language and produce therefrom a linked knowledge model. The linked knowledge model can then be navigated to extract knowledge, according to the relationships "Why "What" and "What is" discussed above, using a user interface, which visually facilitates making the connections between pieces of information.

Embodiments of the present invention can increase a user's knowledge by augmenting the knowledge base available to the user in the form of a linked knowledge model. Gaps in the linked knowledge model and edges or boundaries of the linked knowledge model are susceptible to automatic detection. Embodiments of the invention can therefore detect such gaps and edges and search for missing information on an ad hoc, regularly scheduled or other basis.

According to one aspect of the invention, there is- a system for storing and retrieving knowledge, comprising a text analyzer having an input which receives information as natural language and which produces at an output annotated information, parser having an t ~w cheieives thp annoatdi~ o ation~othj t~anayzer, and which produces and stores in a memol& ainke'd knowIe'g model thereo ,and a user interface presenting at least one pair Jf~opeetr)eainhp oa prtr V /W through which the operator can navigate the linked knowledge model to retrieve o knowledge. The system can further comprise a knowledge research agent having an input receiving the linked knowledge model and having an output on which the knowledge research agent produces a query compatible with an external search engine.

The Hyperknowledge agent may automatically augment the linked knowledge model with new information found by executing the query on the external search engine. In the system, the parser can further comprise a Rules Engine executing a plurality of rules defining a structure of knowledge embedded in information expressed in natural language. The linked knowledge model may include connected noun/verb phrase pairs, and the plurality of rules may define how sentences in natural language relate to connected noun/verb phrase pairs in the linked knowledge model. The rules may include at least one semantic rule, at least one syntactic rule or at least one context rule.

According to a second aspect of the invention, there is a method for retrieving knowledge. The method comprises presentin to an operator a selector by which the operator indicates one of a pair of 2 relationships and a query input by which the operator indicates query text, and combining the one of the pair of A retihips indicte an he quey text toorm a ue by ch knowledges eve cordingothe rn of this aspect o the nnin, the selector may further define four directions characterized by', ,t pairs of directions, each direction representing a relationship. The selector may further be '/a presented in a graphical form resembling a compass rose. The method may further comprise storing knowledge in a linked knowledge model including noun/verb phrase pairs, searching the linked knowledge model for a first noun/verb phrase pair defined by the query text, and retrieving a second noun/verb phrase air from the linked knowledge model bearing the indicated one of the pair of relationships to the first noun/verb phrase pair. Finally, searching may further comprise parsing a natural language query text into a searchable Boolean expression.

According to a third aspect of the invention, there is a method for storing owledge, o nga a natur a documlent into a li kedknwld mode or'oA e" in, k w d 1o model 'cing conn tectn ^se p' en'g t knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model, and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model. According to this method, the identified portions of the linked knowledge model may further include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two relationships defined between the noun/verb phrase pairs.

For instance, according to a fourth aspect, the invention is a software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: reseting to an operator a selector by which the operator indicates one of a pair of c relationships and a query input by which the operator indicates query text; and u co binig the one of the prf e tionships indicated and the query text to torm a queryy whc nowledge is retrieved A x In a fifth aspect, the invention is a software product including'a ac 'ne readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions ofjthe link knowledge modeL I n a sixth aspect, the invntion is a software producdincluding a machine readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method cprising creating a database of pairs of o relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts making available to the operator and presenting to the operator, wherein in act of making available to the operator the multitude of knowledge sets is selected from those that are in relationships with at least one of already selected knowledge sets.

In a seventh aspect, the invention is a system for retrieving knowledge, comprising: a repository of representations of relationships between at least two units of information; and a user interface for accessing at least one representation of at least one -relationship between the at least two units of information, wherein the user interface is further adapted to provide access to subsequent relationships.

In a eighth aspect, the invention is a method for knowledge retrieval, comprising: creating a database of pairs of c relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator a selector and presenting to the operator the at least one knowledge set, wherein in the making available to the operator, multitude of knowledge sets is selected from those that are in-A relationships with at least one of already selected knowledge sets.

In a ninth aspect, the invention is a machine-readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of o.Jl~ relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator and presenting to the operator, wherein in act of snaking available to the operator, the multitude of knowledge sets is selected from those that are in nrn ary relationships with at least one of already selected knowledge sets.

In a tenth aspect, the invention is a method of doing business, comprising: creating a database of pairs of.nlie11er relationships between knowledge sets; providing access to the database of pairs of the fe. relationships between the knowledge sets, wherein an entity obtaining access to the database is adapted to executing a second method, comprising: presenting to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of providing access to the database and presenting to the operator, wherein in act of presenting to the operator, the multitude of knowledge sets is selected from those that are in relationships with at least one of already selected knowledge sets.

In a twelfth aspect, the invention is a method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: selecting a token from the document; retrieving from the repository of knowledge pairs of cg florelationships between knowledge sets that are relevant to the token selected from the document; identifying at least one pair of complimentary relationships relevant to the token selected from the document if any M relationships were retrieved during the act of retrieving; and generating a visual representation of the at least one pair of e 'n'*ay relationships.

In a thirteenth aspect, the invention is a system for retrieving knowledge relevant to a document, comprising: THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:- 1. A system for storing and retrieving knowledge, comprising: a text analyzer having an input which receives information as natural language and which produces at an output annotated information; a parser having an input which receives the annotated information from the text analyzer and which produces and stores in a memory a linked knowledge model thereof, the linked knowledge model defining at least two noun/verb phrase pairs linked by a pair of opposing relationships; and a user interface presenting at least one pair of emplementaryopposing relationships to an operator, through which the operator can navigate the linked knowledge model to retrieve knowledge.

7. The system of claim 5, wherein the rules include at least one syntactic rule.

8. The system of claim 5, wherein the rules include at least one context rule.

97. A method for retrieving knowledge, comprising: presenting to an operator a selector by which the operator indicates one of a pair of comple:metaryopposing relationships and a query input by which the operator indicates query text; and combining the one of the pair of eemplemeitaryopposing relationships indicated and the query text to form a query by which knowledge is retrieved, whereby the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships.

-08. The method of claim 97, wherein the selector defines four directions characterized by complementaopposing pairs of directions, each direction representing a relationship.

The method of claim -08, wherein the selector is presented in a graphical form resembling a compass rose.

4-2 10. The method according to any one of the preceding claims-9, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searching the linked knowledge model for a first noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of eemplemenopposing relationships to the first noun/verb phrase pair.

-4411. The method of claim 4-210, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

412. A method for storing knowledge, comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; linking the at least two connected noun/verb phrase pairs to a pair of opposing relationships; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model.

4-513. The method of claim 12-4, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two eomplementafopposing relationships defined between noun/verb phrase pairs.

614. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: presenting to an operator a selector by which the operator indicates one of a pair of eomplementaryopposing relationships and a query input by which the operator indicates query text; and combining the one of the pair of eemplementopposing relationships indicated and the query text to form a query by which knowledge is retrieved, whereby the knowledge is defined by at least two noun/verb phrase pairs linked by a pair of opposing relationships.

-415. The software product of claim 4-614, wherein the selector defines four directions characterized by eemplementaryopposing pairs of directions, each direction representing a relationship.

4-816. The software product of claim -1-614, wherein the selector is presented in a graphical form resembling a compass rose.

4-917. The software product of any one of claims-14 to 16, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searching the linked knowledge model for a fast noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of eemplementaryopposing relationships to the first noun/verb phrase pair.

S2018. The software product of any one of claims 14 to 17-14, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

192-. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model, where at least two noun/verb phrase pairs are linked by a pair of opposing relationships.

-220. The software product of claim 2--19, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two eofplementaryopposing relationships defined between noun/verb phrase pairs.

23. A system for retrieving knowledge, comprising:.

a repository of representations of relationships between at least two units of information; and a user interface for accessing at least etone representation of at least one relationship between the at least two units of information, wherein the user interface is further adapted to provide access to subsequent relationships.

21. The system of claim 23, wherein the representations of relationships include at least one of the following: representations of "How" and "Why" relationship, and representations of "What" and "What is" relationship.

The system of claim 24, wherein the user interface defines four directions characterized by complementary pairs of directions, each axis representing a relationship.

26. The systcem of claim 25, wherein the user interface is a likeness of a eompass rose with at least two axis representinig at least two OF the representations of 27. The systemi of claim 25, wher-ein~ the user- initerface is activated bypoiiin presen tationl Of a curSorF ovrei the pr ntto of thje t%'o relationfsips eln a display.

28. The system of claimi 27, wherein the user initerface is Pfrther adapted to display-a fepresentatien of previeusly displayed rcesttin of relatioiiships- 29. The system~ of claimi 28, wher~ein the user inter faee is faiher: adapted to prov'ide retrieval of at least one rpsnaun of- a relationshp Hirm the previously displayed reprsenatins of the relationships.

The system~ of elaini 23, w'hereinl the uiser interface is furthef-adapted to display a renRcsental ian of a noni-br nhi-ase nair that eaeh eorresnond to a set of' iformation in ra-treiatahihp7 -3412 1. A method for knowledge retrieval, comprising: creating a database of pairs of ee fpi-mentaiyopposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator a selector and presenting to the operator the at least one knowledge set, wherein in the making available to the operator, multitude of knowledge sets is selected from those that are in eeiplimnen'afyqDp"sing relationships with at least one of already selected knowledge sets.

-&222. The method of claim 3-12 1, wherein the selector represents at least one relationship characterized by a eemppkmentafyopposing pair of directions.

k e, Cc, In Cc, 3423. The method of claim 3222, wherein the selector defines two relationships represented by two axes, each axis representing a relationship.

3424. The method of claim 3-222, wherein the selector is presented in a graphical form representing a compass rose.

3-525. The method of claim 3-222, wherein the aet further comnprisinges presenting to the operator the selector when the operator positions a representation of a pointing device over a representation of a link enabling the knowledge retrieval.

3626. The method of claim -3-525 further comprising presenting to the operator a second selector, by which the operator can select a noun/verb pair corresponding to at least one of the relationships retrieved from the database.

37. The iinetLhd of claini 36, wrvein the aet of preseinting to the Operatorf the secnd seleetair furither- eeomprises preening- 'he operator the seconld selectr wyheni the operatorrollS E! Frepesentation f a pontr a Frepr entation or the selected knowledge set.

38. The mlethod o4 claim~ 32, whereini the act of fflking available to the operatorfurther eo mrisc prenting to the operator the seleetr whein the operator selects a link enabling the knlegelee alQ] Th lmA~ oF~lhimn 3Q frthr eerisin r~ntn to\ the nrtr a secnd r-~ sclcetor, by whieh the oper-ator can select a noun/verb pair corr-esponding to at least oee of the relationships retrieved fromn the database.

The mnethod of elaini :38, whereiin the act of preseniting to the Operator the second seleetor fur~ther coiprises preseniting the oper~ator the seconid selector- when the op oratorclickS on a rereenai ofthe selected knoewledge set.

4 1. The method of claim 32, furither: eonmpris..ingpeetn the operator with a reprsenatin of at least two knoewledge sets r-etrieved in steps of makinlg available to the operator and pr~esenitinig to the operator- the Afrst slco.

4227. The method of any one of claims 3-222 to 26, further comprising: a selecting, by the operator, at least one of the previously selected knowledge sets; and Sdisplaying to the operator the at least one of the previously selected knowledge sets and at least one knowledge set related to the at least one of the previously selected

C

r knowledge sets.

C(,

0 4328. A software product including a machine readable medium on which is encoded itr a sequence of software instructions which, when executed, direct performance of a method comprising: I creating a database of pairs of eemplimentaryopposing relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts making available to the operator and presenting to the operator, wherein in act of making available to the operator the multitude of knowledge sets is selected from those that are in eemplmeniaryopposing relationships with at least one of already selected knowledge sets.

44. A machine readable medium on which is encoded a sequence of sofware instructions which, when executed, direct performance of a method comprising: creating a database of pairs of complirnentaryrelationships between knowledge -laking available to an operator a se.letor, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and rep.ating acts of r making available to the operator and presenting to the operator, wherein in act of snaking available to the operator, the multitude of knwledge sets is selected from those that are in comaplimentaryopposing relationships with at least one of alrcady selected 1knowledge sets.

A method of doing busines; .omprising: eireating a database of pairs of eomplimentary relationships betwee'n knowledge.

setsi pro)Viding aeSS to) the database of pairS 6f the eEopliHmentaf=Felationships between the knoe dge sets, wherein an cntity obtaining ac cess to the database is adapted to exctn eond mcnthod, com:prising:presentinlg tO an operator a selector, by which the oper-ato- can select a first knowledge set from a first multitude of kn~owledge sets retrieved fromi the database; responsive to selection of the at least one knowledge set, prsnigto the operator: the at least one knowledge set; and rcpeating acts of pr~oviding access to the database and presenting to the operaltor, wherint -1Of presentinlg to the operator:, the multitude of knowledge sets sseleeted fromi those that reein nplinientair relationships with at least one of already selected knowledge sets.

46. A method of retrieving knowledge relevant to a docuiment from a repository oe knowledge,copiig providing a repository o~f relationships between at least two units o~f infonna-in identifying at least one relationship relevant to the docuiment from the relationships between at least two units of information; and displaying a user interfaee bi- a cessing at least one representation of the at least one relationship relevant to the doc-ument, wherein the uiser intfce is, futher adapted to provide access to subsequent relationships.

4-29. A method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: selecting a token from the document; retrieving from the repository of knowledge pairs of eempIimieintafyopposing relationships between knowledge sets that are relevant to the token selected from the document; identifying at least one pair of eemip-tinentafyppo-sing relationships relevant to the token selected from the document if any eai 1Thimetafyqppqsing relationships were retrieved during the act of retrieving; and generating a visual representation of the at least one pair of eeffiliffiefitaYOPPOSing relationships.

351 4-830. The method of claim 4-729, further comprising: accumulating pairs of relevant omplimentaryopposing relationships by repeating acts of selecting, retrieving, and identifying for additional tokens in the document; and generating a visual representation of at least one relevant eompiementayopposing relationship from the pairs of relevant eemplimentaryopposing relationships.

49. The method of claim 47, wherein the token is at least one of: an abbreviation, a word, a sentence, and a paragraph.

5031. The method of claim 4830, further comprising removing duplicates from the pairs of relevant eemplmentaryopposing relationships.

5--32. The method of claim 4830, wherein the act of generating the visual representation further comprises generating a visual representation of a link between the at least one relevant complimentaryopposing relationship and a token to which it is relevant.

52. The method of claim 47, wherein pairs of complimentary relationships are noun/verb phrase pairs.

53-- The method of claim 52, wherein the act of identifying the at least one complimentary relationship further comprises identifying at least one noun/verb phrase 54. The method of claim 53, wherein a noun part of the at least one noun/verb phrase pair can be found in a vicinity of the token selected from the document.

5-533. A system for retrieving knowledge relevant to a document, comprising: a repository of representations of opposing relationships between at least two units of information, where each unit of information is definable by a noun/verb phrase pair; a knowledge matcher identifying at least one opposing relationship relevant to the document from the opposing relationships between at least two units of information; and a user interface enabling access to at least one representation of the at least one opposing relationship relevant to the document, wherein the user interface is further adapted to provide access to subsequent opposing relationships.

5634. The system of claim 5533, wherein the mechanism for identifying the at least one opposing relationship relevant to the document further comprises a module for identifying at least one opposing relationship relevant to a token in the document based on a context of the word.

0 57. The system of Claim 56, wheriein the token is at least one of: a wfrd, anH abbreviationi, a senitencee, and a paragiraph.

58. AA systc fr fstoring and retrievifng nowledge, substantially as herein described with reference to the acompanying rngs.

59. A miethod foar retrieving knowledge acording to claim 9, substantially as hcrei described with referene to the acconpanying d.rans.

A mcthod fsr tring knowledge acording to claimi 14, substantially as herein described with refe rene to the acompanying d a 1rawings.

6 I. A so3ftware product inceludinig a miachine readable medium on w.hich is cncoded a sequence ef software instructions whichi when executed dircet performlance ofa 0 miethod in accordancee with claim~f 16 or claim~ 21 or claim~ 43, substantially as herein described With ref~eee to the aeccompanyinig drawns 62. A system for retrieving laiowedge, substantially as herein described with reference to the accomfpanying drawings.

63. A mcthod foFr knowlcdg ri-etrie.,al arfdinig to claim 3 1, substantially as herei described with r-eference to the accompanying drawings 61. A machinc readable miedium on efi which is encoded a sequence of soiwar-e instr~uctionis which, when exeuated, direct performanec of a method in accordancee with claim 44, substantially as herein described with refernie to the acpanying drawings.

0 65. A mnethod of doing busincess acoriding to clai 145, substanitially as herein described vith reference to the accompanying draw...

66. A method Of retrieving knowledge relevant to a docuiment foma raepository o knowledge according to claim 416 or claim 47, substantially, as herein described with reference to the accompanying drawinfgs.

67. A system for retrieving know~ledge relevant to a document, substantially as -41, V-p,,F pp Sp thp ~nr~nr l rlr~iirinrr~ "VV"115 J11".7 Yr I

Claims

1. A system for storing and retrieving knowledge, comprising: a text analyzer having an input which receives information as natural language and which produces at an output annotated information; a parser having an input which receives the annotated information from the text analyzer and which produces and stores in a memory a linked knowledge model thereof; and a user interface presenting at least one pair of complementary relationships to an operator, through which the operator can navigate the linked knowledge model to retrieve knowledge.

5. The system of claim 4, the linked knowledge model including connected noun/verb phrase pairs, and wherein the plurality of rules define how sentences in natural language relate to connected noun/verb phrase pairs in the linked knowledge model.

6. The system of claim 5, wherein the rules include at least one semantic rule.

8. The system of claim 5, wherein the rules include at least one context rule.

9. A method for retrieving knowledge, comprising: presenting to an operator a selector by which the operator indicates one of a pair of complementary relationships and a query input by which the operator indicates query text; and combining the one of the pair of complementary relationships indicated and the query text to form a query by which knowledge is retrieved.

The method of claim 9, wherein the selector defines four directions characterized by complementary pairs of directions, each direction representing a relationship.

11. The method of claim 10, wherein the selector is presented in a graphical form resembling a compass rose.

12. The method of claim, 9, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searcling the linked knowledge model for a first noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of complementary relationships to the first noun/verb phrase pair.

13. The method of claim 12, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

14. A method for storing knowledge, comprising: parsing a natural language document into a linked knowledge model including corrected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model.

The method of claim 14, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two complementary relationships defined between noun/verb phrase pairs.

16. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: presenting to an operator a selector by which the operator indicates one of a pair of complementary relationships and a query input by which the operator indicates query text; and combining the one of the pair of complementary relationships indicated and the query text to form a query by which knowledge is retrieved.

17. The software product of claim 16, wherein the selector defines four directions characterized by complementary pairs of directions, each direction representing a relationship.

18. The software product of claim 16, wherein the selector is presented in a graphical form resembling a compass rose.

19. The software product of claim 16, further comprising: storing knowledge in a linked knowledge model including noun/verb phrase pairs; searching the linked knowledge model for a fast noun/verb phrase pair defined by the query text; and retrieving a second noun/verb phrase pair from the linked knowledge model bearing the indicated one of the pair of complementary relationships to the first noun/verb phrase pair.

The software product of claim 16, wherein searching further comprises parsing a natural language query text into a searchable Boolean expression.

21. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method comprising: parsing a natural language document into a linked knowledge model including connected noun/verb phrase pairs; identifying portions of the linked knowledge model in which a level of knowledge represented is lower relative to other portions of the linked knowledge model; and executing searches of external information for additional knowledge related to the identified portions of the linked knowledge model.

22. The software product of claim 21, wherein the identified portions of the linked knowledge model include a first noun/verb phrase pair lacking a second noun/verb phrase pair related by at least one of two complementary relationships defined between noun/verb phrase pairs.

23. A system for retrieving knowledge, comprising: a repository of representations of relationships between at least two units of information; and a user interface for accessing at least one representation of at least one relationship between the at least two units-of information, wherein the user interface is further adapted to provide access to subsequent relationships.

24. The system of claim 23, wherein the representations of relationships include at least one of the following: representations of "How" and "Why" relationship, and representations of "What" and "What is" relationship.

26. The system of claim 25, wherein the user interface is a likeness of a compass rose with at least two axis representing at least two of the representations of relationships'

27. The system of claim 25, wherein the user interface is activated by positioning presentation of a cursor over the presentation of the two relationships on a display.

28. The system of claim 27, wherein the user interface is further adapted to display a representation of previously-displayed representations of relationships.

29. The system of claim 28, wherein the user interface is further adapted to provide retrieval of at least one representation of a relationship from the previously displayed representations of the relationships.

The system of claim 23, wherein the user interface is fuirther adapted to display a representation of a noun/verb phrase pair that each correspond to a set of information in a relationship.

31. A method for knowledge retrieval, comprising: creating a database of pairs of complimentary relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator a selector and presenting to the operator the at least one knowledge set, wherein in the making available to the operator, multitude of knowledge sets is selected from those that are in complimentary relationships with at least one of already selected knowledge sets.

32. The method of claim 31, wherein the selector represents at least one relationship characterized by a complimentary pair of directions.

33. The method of claim 32, wherein the selector defines two relationships represented by two axes, each axis representing a relationship.

34. The method of claim 32, wherein the selector is presented in a graphical form representing a compass rose.

The method of claim 32, wherein the act further comprises presenting to the operator the selector when the operator positions a representation of a pointing device over a representation of a link enabling the knowledge retrieval.

36. The method of claim 35, further comprising presenting to the operator a second selector, by which the operator can select a noun/verb pair corresponding to at least one of the relatiohships retrieved from the database.

37. The method of claim 36, wherein the act of presenting to the operator the second selector further comprises presenting the operator the second selector when the operator rolls a representation of a pointer over a representation of the selected knowledge set.

38. The method of claim 32, wherein the act of making available to the operator further comprises presenting to the operator the selector when the operator selects a link enabling the knowledge retrieval.

39. The method of claim 38, further comprising presenting to the operator a second selector, by which the operator can select a noun/verb pair corresponding to at least one of the relationships retrieved from the database.

The method of claim 38, wherein the act of presenting to the operator the second selector fuirther comprises presenting the operator the second selector when the operator clicks on a representation of the selected knowledge set.

41. The method of claim 32, further comprising presenting the operator with a representation of at least two knowledge sets retrieved in steps of making available to the operator and presenting to the operator the first selector.

42. The method of claim 32, further comprising: selecting, by the operator, at least one of the previously selected knowledge sets; and displaying to the operator the at least one of the previously selected knowledge sets and at least one knowledge set related to the at least one of the previously selected knowledge sets.

43. A software product including a machine readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of complimentary relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts making available to the operator and presenting to the operator, wherein in act of making available to the operator the multitude of knowledge sets is selected from those that are in complimentary relationships with at least one of already selected knoivledge sets.

44. A machine-readable medium on which is encoded a sequence of software instructions which, when executed, direct performance of a method comprising: creating a database of pairs of complimentary relationships between knowledge sets; making available to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of making available to the operator and presenting to the operator, wherein in act of snaking available to the operator, the multitude of knowledge sets is selected from those that are in complimentary relationships with at least one of already selected knowledge sets.

45. A method of doing business, comprising: creating a database of pairs of complimentary relationships between knowledge sets; providing access to the database of pairs of the complimentary relationships between the knowledge sets, wherein an entity obtaining access to the database is adapted to executing a second method, comprising: presenting to an operator a selector, by which the operator can select a first knowledge set from a first multitude of knowledge sets retrieved from the database; responsive to selection of the at least one knowledge set, presenting to the operator the at least one knowledge set; and repeating acts of providing access to the database and presenting to the operator, wherein in act of presenting to the operator, the multitude of knowledge sets is selected from those that are in complimentary relationships with at least one of already selected knowledge sets.

46. A method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: providing a repository of relationships between at least two units of information; identifying at least one relationship relevant to thedocument from the relationships between at least two units of information; and displaying a user interface for accessing at least one representation of the at least one relationship relevant to the document, wherein the user interface is further adapted to provide access to subsequent relationships.

47. A method of retrieving knowledge relevant to a document from a repository of knowledge, comprising: selecting a token from the document; retrieving from the repository of knowledge pairs of complimentary io relationships between knowledge sets that are relevant to the token selected from the document; identifying at least one pair of complimentary relationships relevant to the token selected from the document if any complimentary relationships were retrieved during the act of retrieving; and generating a visual representation of the at least one pair of complimentary relationships.

48. The method of claim 47, further comprising: accumulating pairs of relevant complimentary relationships by repeating acts of selecting, retrieving, and identifying for additional tokens in the document; and generating a visual representation of at least one relevant complimentary relationship from the pairs of relevant complimentary relationships.

The method of claim 48, further comprising removing duplicates from the pairs of relevant complimentary relationships.

51. The method of claim 48, wherein the act of generating the visual representation further comprises generating a visual representation of a link between the at least one relevant complimentary relationship and a token to which it is relevant.

53. The method of claim 52, wherein the act of identifying the at least one complimentary relationship further comprises identifying at least one noun/verb phrase pair.

54. The method of claim 53, wherein a noun part of the at least one noun/verb phrase pair can be found in a vicinity of the token selected from the document.

A system for retrieving knowledge relevant to a document, comprising: a repository of representations of relationships between at least two units of information; a knowledge matcher identifying at least one relationship relevant to the document from the relationships between at least two units of information; and a user interface enabling access to at least one representation of the at least one relationship relevant to the document, wherein the user interface is further adapted to provide access to subsequent relationships.

56. The system of claim 55, wherein the mechanism for identifying the at least one relationship relevant to the document further comprises a module for identifying at least one relationship relevant to a token in the document based on a context of the word.

57. The system of claim 56, wherein the token is at least one of: a word, an abbreviation, a sentence, and a paragraph.

58. A system for storing and retrieving knowledge, substantially as herein described with reference to the accompanying drawings.

59. A method for retrieving knowledge according to claim 9, substantially as herein described with reference to the accompanying drawings.

A method for storing knowledge according to claim 14, substantially as herein described with reference to the accompanying drawings.

61. A software product including a machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method in accordance with claim 16 or claim 21 or claim 43, substantially as herein described with reference to the accompanying drawings.

62. A system for retrieving knowledge, substantially as herein described with reference to the accompanying drawings.

63. A method for knowledge retrieval according to claim 31, substantially as herein described with reference to figures 2 to 7 of the accompanying drawings.

64. A machine readable medium on which is encoded a sequence of software instructions which when executed direct performance of a method in accordance with claim 44, substantially as herein described with reference to the accompanying drawings.

A method of doing business according to claim 45, substantially as herein described with reference to the accompanying drawings.

66. A method of retrieving knowledge relevant to a document from a repository of knowledge according to claim 46 or 47, substantially as herein described with reference to the accompanying drawings. 32

67. A system for retrieving knowledge relevant to a document, substantially as herein described with reference to figures 2 to 7 of the accompanying drawings. Dated this thirty-first day of May 2005 Trevor Bryan Davies Patent Attorneys for the Applicant: F B RICE CO