Generating a document representation using semantic networks

• US 8,756,185 B2
• Filed: 12/14/2012
• Issued: 06/17/2014
• Est. Priority Date: 03/06/2009
• Status: Active Grant

• Current Assignee: Ifwe Inc.
• Original Assignee: Tagged Incorporated
• Inventors: Dawson, Karl, Metel, Piotr, Dzialo, Jaromir, Niewiadomska, Agata, Grochowski, Marek
• Primary Examiner(s): CHANG, LI WU
• Application Number:
  US13/714,937
• Publication Number:
  US 20140019385A1
• Time in Patent Office: 550 Days
• Field of Search: None
• US Class Current: 706/45
• CPC Class Codes: G06F 16/322 Trees
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition

Abstract

A method, system and computer program product for developing a semantic network are disclosed. The system includes a server and a client computer, and the method involves: creating a semantic network containing at least one root concept; performing a set of instructions associated with the semantic network, the set of instructions comprising: transmitting information about the semantic network from the client computer to a server; providing information about at least one resource to the server, the at least one resource containing concepts and relations associated with the at least one root concept; receiving information about a modified semantic network from the server; presenting the information about the modified semantic network to a user; receiving a response from the user; based on the response, further modifying the semantic network. The system interactively modifies semantic networks in response to user feedback, and produces personal semantic networks and document use histories.

Claims

1. A non-transitory computer-readable medium comprising computer program code for controlling one or more processors of a client computer to carry out the following steps:

creating a semantic network containing at least one root concept;

performing a set of instructions associated with the semantic network, the set of instructions comprising;

transmitting information about the semantic network from the client computer to a server;

providing information about at least one resource to the server;

receiving information about a modified semantic network from the server, the modified semantic network being an updated version of the semantic network having been updated based on the information about the at least one resource;

presenting the information about the modified semantic network to a user;

receiving a response from the user, the response including a numeric weight representative of a conceptual unattractiveness value indicating the user'"'"'s preferred direction for modifying the modified semantic network; and

based on the response, further modifying the modified semantic network.

2. The non-transitory computer readable medium of claim 1, wherein the at least one root concept is a user-selected topic mark from an existing semantic network, the existing semantic network being previously created and stored in the server.

3. The non-transitory computer readable medium of claim 1, wherein providing information about at least one resource to the server comprises:

querying a user for the at least one resource;

receiving a response from the user about the at least one resource; and

based on the response, transmitting information about the at least one resource from the client computer to the server.

4. The non-transitory computer-readable medium of claim 3, further comprising:

carrying out, in response to receiving a null response from the user about the at least one resource, an automated search in a set of resources available via a wide-area network;

rating results of the search for their relevance;

presenting the user with the rated results; and

querying the user for a selection from the rated results.

5. The non-transitory computer-readable medium of claim 1, comprising code for creating a data structure comprising concepts, relations between the concepts, and concept/relation metadata.

6. The non-transitory computer-readable medium of claim 1, further comprising code for creating concept/relation metadata that comprise at least one of:

a numeric weight associated with each concept;

a comment associated with at least one concept or at least one relation;

a tag for annotating the at least one concept;

an identifier for the tag;

a numeric weight associated with the tag;

a source indication of each concept;

an identifier for each concept or relation;

a timestamp for each concept or relation; and

a concept similarity directive.

7. The non-transitory computer-readable medium of claim 1, further comprising code for creating concept/relation metadata that comprise a numeric weight associated with each concept, a source indication of each concept, an identifier for each concept or relation, a timestamp for each concept or relation, and a concept similarity directive.

8. The non-transitory computer-readable medium of claim 1, further comprising code for creating concept/relation metadata that comprise a tag for annotating at least one concept, an identifier for the tag, and a numeric weight associated with the tag.

9. The non-transitory computer-readable medium of claim 1, further comprising code for creating a resource structure for the at least one resource, the resource structure comprising:

a unique identifier; and

resource metadata associated with the resource.

10. The non-transitory computer-readable medium of claim 9, wherein the resource metadata comprises at least one of:

a source indication of the resource;

a reference to all or part of the resource;

a resource title;

a numeric weight associated with the resource;

an identifier for the resource; and

a timestamp for the resource.

11. The non-transitory computer-readable medium of claim 9, wherein the resource metadata comprises a reference to all or part of the resource and a numeric weight associated with the resource.

12. The non-transitory computer-readable medium of claim 9, wherein the resource metadata comprises a source indication of the resource, a reference to all or part of the resource, a numeric weight associated with the resource, and a timestamp for the resource.

13. The non-transitory computer-readable medium of claim 1, further comprising code for providing control parameters to the server, the control parameters comprising at least one of:

a limit on new connections to be contained in the modified semantic network;

a limit on a depth of new connections to be contained in the modified semantic network;

a concept similarity threshold to be applied in processing semantic structures;

a limit on a number of new resources from which semantic structures are to be extracted; and

an absolute limit on the maximum depth of relations in the modified semantic network.

14. The non-transitory computer-readable medium of claim 1, further comprising code for providing control parameters to the server, the control parameters comprising a limit on new connections to be contained in the modified semantic network, a limit on a depth of new connections to be contained in the modified semantic network, and an absolute limit on the maximum depth of relations in the modified semantic network.

15. The non-transitory computer-readable medium of claim 1, further comprising code for providing control parameters to the server, the control parameters comprising a concept similarity threshold to be applied in processing semantic structures, an absolute limit on the maximum depth of relations in the modified semantic network, and a limit on a number of new resources from which semantic structures are to be extracted.

16. The non-transitory computer-readable medium of claim 1, wherein the code implements a plug-in for a web browser.

20. The non-transitory computer-readable medium of claim 1 further comprising the server non-transitory computer-readable medium of claim 17.

17. A non-transitory computer-readable medium comprising computer program code for controlling one or more processors of a server to carry out the following steps:

receiving information about a semantic network comprising at least one root concept from a client computer;

determining at least one resource based on the information about the semantic network by carrying out an automated search in a set of resources available via a wide-area network and receiving results from the automated search;

extracting semantic structures from the at least one resource;

processing the semantic structures for determining a measure of significance for the semantic network;

modifying the semantic network, based on the processed semantic structures; and

transmitting information about the modified semantic network to the client computer.

18. The non-transitory computer-readable medium of claim 17, wherein processing the semantic structures comprises:

sending at least one query from the server computer to a knowledge source;

receiving at least one response from the knowledge source by the server computer; and

rating the semantic structures for the measure of significance based on the response.

19. The non-transitory computer-readable medium of claim 18, wherein the knowledge source comprises at least one of:

an ontology;

a database;

a knowledgebase; and

services, brokers and agents accessible through a wide-area network.

21. A system for developing a semantic network, the system comprising a client computer and a server,

the client computer comprising:

a component configured to create a semantic network comprising at least one root concept;

a component configured to transmit information about the semantic network from the client computer to the server computer;

a component configured to provide information about at least one resource to the server computer;

a component configured to provide, to the server computer, control parameters including a limit on new connections to be contained in a modified semantic network;

a component configured to receive information about the modified semantic network from the server computer;

a component configured to present the information about the modified semantic network to a user;

a component configured to receive a response from the user;

a component configured to further modify the modified semantic network based on the response;

a component configured to receive commands from the user on whether to repeat modification of the semantic network;

the server comprising;

a component configured to receive information about a semantic network comprising at least one root concept from the client computer;

a component configured to receive information about at least one resource from the client computer;

a component configured to extract semantic structures from the at least one resource;

a component configured to process the semantic structures, for determining a measure of significance for the semantic network;

a component configured to modify the semantic network, based on the processed semantic structures; and

a component configured to transmit information about the modified semantic network to the client computer.

22. The system of claim 21, wherein the server comprises a plurality of server computers configured to extract the semantic structures from the at least one resource and to process the semantic structures.

23. A method for developing a semantic network, the method comprising:

receiving into a memory of a system a user'"'"'s input,

the user'"'"'s input comprising;

a semantic network containing a root concept; and

a list of one or more resources associated with the root concept;

the system comprising;

a server; and

a client computer;

providing to the user, in response to the user'"'"'s input, a set of one or more concept-relation-concept (CRC) triples;

receiving from the user a response associated with the set of one or more CRC triples, the response including a numeric weight for one or more concepts in the set of one or more CRC triples, the numeric weight representative of a conceptual unattractiveness value indicating the user'"'"'s preferred direction for modifying the semantic network;

modifying the semantic network based on the response; and

repeating modification of the semantic network according to commands from the user, wherein modification is repeated until all possible CRC triples have been provided based on the list of one or more resources.

24. The method of claim 23, wherein providing to the user, in response to the user'"'"'s input, a set of one or more concept-relation-concept (CRC) triples comprises:

determining a semantic similarity between concepts associated with different CRC triples within the set of one or more CRC triples;

producing one or more connections between the concepts having sufficient the semantic similarity; and

presenting to the user the set of one or more CRC triples and associated the one or more connections, thereby generating an expanded semantic network associated with the user'"'"'s input.

25. The method of claim 23, further comprising providing tags and/or terms for each concept in the set of one or more CRC triples.

26. The method of claim 24, wherein determining a semantic similarity comprises:

selecting at least two concepts from the set of one or more CRC triples, each of the at least two concepts being from a different CRC triple;

associating concept/relation metadata with the at least two concepts;

determining a probability of similarity based on the concept/relation metadata associated with the at least two concepts;

determining whether the probability of similarity is equal to or greater than a threshold, the threshold being set by the user; and

creating at least one connection between the at least two concepts if the probability of similarity is equal to or greater than the threshold.

27. The method of claim 26, wherein the concept/relation metadata comprises a concept similarity directive.

28. The method of claim 23, wherein the response further includes:

a removal request for one or more concepts in the set of one or more CRC triples, the removal request indicative of the user'"'"'s preferred direction for modifying the semantic network.

29. The method of claim 23, wherein modifying the semantic network based on the response comprises:

updating the set of one or more CRC triples with new CRC triples according to the response; and

providing to the user an updated set of one or more CRC triples,

wherein modifying the semantic network based on the response is repeated until all possible new CRC triples have been provided based on the list of one or more resources.

30. A method, performed by a system comprising a server and a client computer, for developing a personal semantic network, the method comprising:

producing a set of one or more document semantic networks associated with a document in response to natural language sentences within the document;

receiving, into a memory of the system, input commands from a user for generating one or more document output representations by utilizing the set of one or more document semantic networks associated with the document;

producing a set of one or more semantic statements in response to the input commands associated with the one or more document output representations;

generating a personal semantic network comprising one or more semantic relationships between the set of one or more semantic statements and the set of one or more document semantic networks, wherein the one or more semantic relationships are determined by a semantic similarity between the one or more semantic statements and the one or more document semantic networks;

storing into the memory of the system a document use history comprising the input commands received in association with the one or more document output representations; and

modifying the personal semantic network in response to the document use history.

31. The method of claim 30, wherein the personal semantic network is accessible to one or more additional users depending on sharing privileges granted by the user.

32. The method of claim 30, further comprising:

generating a first personal semantic network associated with a first user;

generating a second personal semantic network associated with a second user;

establishing one or more connections between the first personal semantic network and the second personal semantic network, wherein the one or more connections are determined by the semantic similarity between the first personal semantic network and the second personal semantic network; and

producing a personal output representation of a given document based on the one or more connections, thereby enabling sharing of information associated with the given document between the first user and the second user.

33. The method of claim 32, wherein producing a personal output representation based on the one or more connections comprises:

generating the personal output representation for the first user based on the first user'"'"'s input commands for generation of a given representation associated with the given document; and

presenting to the second user the personal output representation in response to the second user'"'"'s request, thereby enabling information sharing without having to allow the second user access to the given document.