Scoring concepts for contextual personalized information retrieval
First Claim
1. A computer-implemented method for generating search results from an input query, the method comprising:
- using a computer system to execute method steps comprising;
receiving an input query for a search comprising target concepts;
determining a structured set of criteria associated with criteria values, wherein the input query is normalized into the structured set of criteria, and wherein the criteria specify how to filter and score the target concepts, and wherein the criteria value is a tuple that corresponds to the criteria;
mapping the structured set of criteria and associated criteria values to nodes in a knowledge base comprising a semantic network relationship among concepts and an index of a plurality of documents, wherein each node represents a target concept associated with a target value, and wherein the target concept is an instance of a category;
matching nodes for the target values of the target concepts to nodes for selection aspects of the criteria and criteria values using transitivity across the semantic network between these nodes;
selecting the set of target concepts having the associated target values that match the selection aspects of the criteria and the criteria values in the matching of the nodes;
computing a partial score on each of the target values in the selected target concepts based on scoring aspects of the criteria and criteria values, each partial score scoring each of the target values against the criteria values as descendants of the criteria values in the semantic network, and each partial score measuring a degree of match between the target values used in computing that partial score and the criteria values on which that partial score is based;
computing at least one weight for each of the criteria values;
combining the partial scores for the target values that matched a given criteria value to generate a single partial score for that criteria value that is based on the at least one weight for that criteria value, wherein each selected target concept has a set of single partial scores for each criteria value;
determining a total score for each selected target concept by integrating the set of single partial scores of the selected target concept; and
applying the total scores for the target concepts to generate search results for the user in a ranked order, the search results including one or more of the documents indexed.
7 Assignments
0 Petitions
Accused Products
Abstract
Information retrieval systems face challenging problems with delivering highly relevant and highly inclusive search results in response to a user'"'"'s query. Contextual personalized information retrieval uses a set of integrated methodologies that can combine automatic concept extraction/matching from text, a powerful fuzzy search engine, and a collaborative user preference learning engine to provide accurate and personalized search results. The system can include constructing a search query to execute a search of a database. The system can parse an input query from a user conducting the search of the database into sub-strings, and can match the sub-strings to concepts in a semantic concept network of a knowledge base. The system can further map the matched concepts to criteria and criteria values that specify a set of constraints on and scoring parameters for the matched concepts.
-
Citations
47 Claims
-
1. A computer-implemented method for generating search results from an input query, the method comprising:
using a computer system to execute method steps comprising; receiving an input query for a search comprising target concepts; determining a structured set of criteria associated with criteria values, wherein the input query is normalized into the structured set of criteria, and wherein the criteria specify how to filter and score the target concepts, and wherein the criteria value is a tuple that corresponds to the criteria; mapping the structured set of criteria and associated criteria values to nodes in a knowledge base comprising a semantic network relationship among concepts and an index of a plurality of documents, wherein each node represents a target concept associated with a target value, and wherein the target concept is an instance of a category; matching nodes for the target values of the target concepts to nodes for selection aspects of the criteria and criteria values using transitivity across the semantic network between these nodes; selecting the set of target concepts having the associated target values that match the selection aspects of the criteria and the criteria values in the matching of the nodes; computing a partial score on each of the target values in the selected target concepts based on scoring aspects of the criteria and criteria values, each partial score scoring each of the target values against the criteria values as descendants of the criteria values in the semantic network, and each partial score measuring a degree of match between the target values used in computing that partial score and the criteria values on which that partial score is based; computing at least one weight for each of the criteria values; combining the partial scores for the target values that matched a given criteria value to generate a single partial score for that criteria value that is based on the at least one weight for that criteria value, wherein each selected target concept has a set of single partial scores for each criteria value; determining a total score for each selected target concept by integrating the set of single partial scores of the selected target concept; and applying the total scores for the target concepts to generate search results for the user in a ranked order, the search results including one or more of the documents indexed. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28)
-
29. A computer program product having a non-transitory computer-readable storage medium having computer program instructions recorded thereon for generating search results from an input query, the computer program instruction comprising instructions for:
-
receiving an input query for a search comprising target concepts; determining a structured set of criteria associated with criteria values, wherein the input query is normalized into the structured set of criteria, and wherein the criteria specify how to filter and score the target concepts, and wherein the criteria value is a tuple that corresponds to the criteria; mapping the structured set of criteria and associated criteria values to nodes in a knowledge base comprising a semantic network relationship among concepts and an index of a plurality of documents, wherein each node represents a target concept associated with a target value, and wherein the target concept is an instance of a category; matching nodes for the target values of the target concepts to nodes for selection aspects of the criteria and criteria values using transitivity across the semantic network between these nodes; selecting the set of target concepts having the associated target values that match the selection aspects of the criteria and the criteria values in the matching of the nodes; computing a partial score on each of the target values in the selected target concepts based on scoring aspects of the criteria and criteria values, each partial score scoring each of the target values against the criteria values as descendants of the criteria values in the semantic network, and each partial score measuring a degree of match between the target values used in computing that partial score and the criteria values on which that partial score is based; computing at least one weight for each of the criteria values; combining the partial scores for the target values that matched a given criteria value to generate a single partial score for that criteria value that is based on the at least one weight for that criteria value, wherein each selected target concept has a set of single partial scores for each criteria value; determining a total score for each selected target concept by integrating the set of single partial scores of the selected target concept; and applying the total scores for the target concepts to generate search results for the user in a ranked order, the search results including one or more of the documents indexed. - View Dependent Claims (30, 31, 32, 33, 34, 35, 36, 37, 38, 39)
-
-
40. A computer system for generating search results from an input query, the system comprising:
-
at least one processor; a computer-readable storage medium storing software components for execution by the at least one processor, the components comprising; a contextual search manager for; receiving an input query for a search comprising target concepts; determining a structured set of criteria associated with criteria values, wherein the input query is normalized into the structured set of criteria, and wherein the criteria specify how to filter and score the target concepts and wherein the criteria value is a tuple that corresponds to the criteria; mapping the structured set of criteria and associated criteria values to nodes in a knowledge base comprising a semantic network relationship among concepts and an index of a plurality of documents, wherein each node represents a target concept associated with a target value, and wherein the target concept is an instance of a category in the knowledge base, and wherein the target value is an instance of an attribute of the category; a criteria evaluator for; matching nodes for the target values of the target concepts to nodes for selection aspects of the criteria and criteria values using transitivity across the semantic network between these nodes; selecting the set of target concepts having the associated target values that match the selection aspects of the criteria and the criteria values in the matching of the nodes; computing a partial score on each of the target values in the selected target concepts based on scoring aspects of the criteria and criteria values, each partial score scoring each of the target values against the criteria values as descendants of the criteria values in the semantic network, and each partial score measuring a degree of match between the target values used in computing that partial score and the criteria values on which that partial score is based; computing at least one weight for each of the criteria values; combining the partial scores for the target values that matched a given criteria value to generate a single partial score for that criteria value that is based on the at least one weight for that criteria value, wherein each selected target concept has a set of single partial scores for each criteria value; a contextual search evaluator for; determining a total score for each selected target concept by integrating the set of single partial scores of the selected target concept; and applying the total scores for the target concepts to generate search results for the user in a ranked order, the search results including one or more of the documents indexed. - View Dependent Claims (41, 42, 43, 44, 45, 46, 47)
-
Specification