Method and system utilizing a personalized user model to develop a search request

• US 9,594,850 B2
• Filed: 10/10/2014
• Issued: 03/14/2017
• Est. Priority Date: 11/17/2011
• Status: Active Grant

• Current Assignee: SRI International, Inc.
• Original Assignee: SRI International, Inc.
• Inventors: Donneau-Golencer, Thierry D.
• Primary Examiner(s): Wong, Leslie
• Application Number:
  US14/511,999
• Publication Number:
  US 20150046435A1
• Time in Patent Office: 886 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/9032 Query formulation

Abstract

A method and system for developing a computer-executable query relating to a search request issued by a user includes determining whether a search term of the search request has a user-specific meaning, connotation, context, or association based on an analysis of electronic content associated with the user and/or an analysis of interactions of the user with electronic content that is accessible to the user through one or more computing devices. If the search term has a user-specific meaning, connotation, context, or association, the method and system can incorporate the user-specific meaning, connotation, context, or association into the search request and/or the computer-executable query.

Claims

1. An information retrieval system comprising, embodied in one or more non-transitory machine accessible storage media:

an electronic content and interaction analyzer module to cause one or more computing devices to generate user-specific semantic information for a plurality of pieces of stored content and interactions data;

a query developer module to cause the one or more computing devices to;

with the user-specific semantic information, interpret a search request; and

based on the interpretation of the search request, develop a search query reflecting the user-specific semantic information; and

a query executer module to cause the one or more computing devices to;

execute the search query to obtain query results; and

output a representation of the query results to an output device of the one or more computing devices;

wherein the electronic content and interaction analyzer module is to extract a plurality of instances of a word from the stored content and interactions data, associate different user-specific semantic information with different instances of the extracted word, and store data indicative of the associations of different user-specific semantic information with different instances of the extracted word in a user-specific model, and wherein the query developer module is to interface with the user-specific model to develop the search query.

2. The information retrieval system of claim 1, wherein the query developer module is to formulate a plurality of alternative search requests based on the user-specific semantic information, output the alternative search requests, receive an interactive selection of at least one of the alternative search requests, and formulate the search query based on the selected alternative search request.

3. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to periodically update the user-specific semantic information in response to new content and interactions data.

4. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to create an index of most important words in the stored content and interactions data and store the index in a user-specific model, and wherein the query developer module is to interface with the user-specific model to develop the search query.

5. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to perform a similarity analysis of pieces of the stored content and interactions data, calculate similarity scores indicative of similarities between pieces of the stored content and interactions data, and store the similarity scores in a user-specific model, and wherein the query developer module is to interface with the user-specific model to develop the search query.

6. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to assign weights to pieces of the stored content and interactions data based on one or more of frequency of use and recency of use, and store the weights in a user-specific model, and wherein the query developer module is to interface with the user-specific model to develop the search query.

7. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to associate different content or activity types with different pieces of the stored content or interactions data and store data indicative of the associations of different content or activity types with different pieces of the stored content or interactions data in a user-specific model, and wherein the query developer module is to interface with the user-specific model to develop the search query.

8. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to identify words that are related to words or information in the stored content and interactions data, classify the relationship between the identified words and the words or information in the stored content and interactions data according to relationship types, and store the related words and data indicative of the relationship types in a user-specific model, and wherein the query developer is to include one or more of the related words in the search query.

9. The information retrieval system of claim 1, wherein the electronic content and interaction analyzer module is to associate words in the stored content and interactions data with topics, and store data indicative of the associations of words and topics in a user-specific model, and the query executer is to interface with the user-specific model to identify a plurality of alternative topics.

10. The information retrieval system of claim 1, wherein the query developer module comprises a query interpreter module to determine user-specific meanings of individual search terms of a search request, and the query developer module is to include the user-specific meanings of the individual search terms of the search request in the search query.

11. The information retrieval system of claim 10, wherein the query developer module comprises a query augmenter module to analyze the user-specific meanings of search terms output by the query interpreter module and identify a new search term related to an individual search term of the search request, and the query developer module is to include the new search term in the search query.

12. The information retrieval system of claim 10, wherein the query interpreter module comprises a plurality of search term analyzer modules including at least two of:

a topic analyzer module, a heat map analyzer module, a semantic analyzer module, a context analyzer module, and a query history analyzer module, each of the search term analyzer modules is to interface with a user-specific model to perform a different type of analysis of the search request, and the query interpreter module utilizes output from the search term analyzer modules to annotate the search request.

13. The information retrieval system of claim 10, wherein the query developer module comprises a query corrector module to interface with a user-specific model to determine a user-specific correction of a search term and annotate the search request to include the user-specific correction of the search term.

14. The information retrieval system of claim 1, wherein the query executer module is to identify a plurality of alternative topics by analyzing the query results;

output the plurality of alternative topics to an output device of the one or more computing devices;

receive an interactive selection of at least one of the plurality of alternative topics via an input device of the one or more computing devices; and

output query results associated with the selected topic to the output device.

15. One or more non-transitory computer accessible storage media comprising a plurality of instructions that in response to being executed, result in a computing device:

generating, for a user, user-specific semantic information from stored content and interactions data of the user;

receiving a search request from the user;

based on the search request, developing a search query including at least some of the user-specific semantic information;

executing the search query;

with the user-specific semantic information, analyzing results of the execution of the search query; and

presenting output in response to the search query in accordance with the analysis of the results of the execution of the search query;

wherein the developing search query further comprises to extract a plurality of instances of a word from the stored content and interactions data, associate different user-specific semantic information with different instances of the extracted word, and store data indicative of the associations of different user-specific semantic information with different instances of the extracted word in a user-specific model, and interface with the user-specific model to develop the search query.

16. The one or more non-transitory computer accessible media of claim 15, wherein said analyzing results of the execution further comprises executing a clustering algorithm to cluster the results of the execution of the search query into a plurality of alternative topics, wherein the plurality of alternative topics are determined at least in part based on the user-specific semantic information.

17. The one or more non-transitory computer accessible media of claim 15, wherein said analyzing results of the execution further comprises identifying a plurality of alternative topics based on the user-specific semantic information;

outputting the plurality of alternative topics; and

receiving an interactive selection of one or more of the plurality of alternative topics; and

said presenting output comprises outputting query results associated with the selected topic.

18. A computing device comprising:

at least one processor,

an input device coupled to the at least one processor;

an output device coupled to the at least one processor; and

at least one memory device coupled to the at least one processor, the at least one memory device having stored therein a plurality of instructions that when executed by the at least one processor cause the at least one processor to;

periodically, over time, for a user, generate user-specific semantic information from stored content and interactions data of the user;

receive a search request;

develop a plurality of alternative search queries, each alternative search query corresponding to the search request and including a different set of the user-specific semantic information; and

output data indicative of results of executing each of the alternative search queries;

wherein developing a plurality of alternative search queries further comprises the steps to;

extract a plurality of instances of a word from the stored content and interactions data;

associate different user-specific semantic information with different instances of the extracted word;

store data indicative of the associations of different user-specific semantic information with different instances of the extracted word in a user-specific model; and

interface with the user-specific model to develop the search query.

19. The computing device of claim 18, wherein to output data indicative of results of executing each of the alternative search queries comprises displaying a plurality of different topics, each topic indicative of the results of execution of one of the alternative search queries.