GENERATING SEARCH RESULTS BASED ON USER FEEDBACK

• US 20110004609A1
• Filed: 07/02/2009
• Published: 01/06/2011
• Est. Priority Date: 07/02/2009
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Holt, Alexander W., Emanuel, Barton W., Moran, Michael E., CHITIVELI, SRINIVAS V.
• Granted Patent: US 8,150,843 B2
• Time in Patent Office: Days
• US Class Current: 707/759
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

Systems, methods and articles of manufacture are disclosed for generating search results based on user feedback. A request may be received to generate search results retrieved using a search string. The request may include user feedback for one or more selected documents of the search results. Improved search results may be generated based on the search results and the feedback for one or more selected documents of the search results. The improved search results may be output to a graphical display device.

Claims

1. A computer-implemented method for processing search results, comprising configuring one or more processors to perform an operation comprising:

receiving an initial set of search results, wherein the initial set of search identifies a plurality of documents responsive to a search request, ranked in an initial ordering;

receiving a selection of one or more documents from the plurality of documents;

training a classifier using each selected document as a training example;

supplying one or more of the plurality of documents responsive to the search request documents to the trained classifier to obtain a measure of similarity between each document supplied to the trained classifier and a document category represented by the selected documents;

re-ranking the initial ordering of each document having a measure of similarity which exceeds a specified threshold; and

presenting the re-ranked search results to a user on a graphical display device.

2. The computer-implemented method of claim 1, wherein re-ranking the initial ordering comprises, ranking each document having a measure of similarity which exceeds a specified threshold ahead of other documents in the initial search result.

3. The computer-implemented method of claim 1, wherein the selection of one or more documents from the plurality of documents is received in response to monitoring what documents referenced in the initial search results are accessed by the user and prompting the user to indicate whether each accessed document is relevant to what the user is searching for.

4. The computer-implemented method of claim 1, wherein the search string includes a plurality of search terms.

5. The computer-implemented method of claim 1, wherein the classifier is one of a naï

ve Bayes classifier, a linear classifier, a latent semantic indexing classifier and an artificial neural network.

6. The computer-implemented method of claim 1, wherein the selection of one or more documents from the plurality of documents indicates at least one document relevant to what the user is actually searching for, and wherein the at least one document is used as a positive training example to train the classifier.

7. The computer-implemented method of claim 1, wherein the selection of one or more documents from the plurality of documents indicates at least one document not relevant to what the user is actually searching for, and wherein the at least one document is used as a negative training example to train the classifier.

8. The computer-implemented method of claim 1, further comprising:

supplying one or more documents not included in the initial search request documents to the trained classifier to obtain the measure of similarity between each document supplied to the trained classifier and the document category represented by the selected documents; and

augmenting the initial search results with any of the documents not included in the initial search results which have a measure of similarity which exceeds the specified threshold.

9. A computer program product, the computer program product comprising a computer usable medium having computer usable program code for processing search results, the code being configured for:

receiving an initial set of search results, wherein the initial set of search identifies a plurality of documents responsive to a search request, ranked in an initial ordering;

receiving a selection of one or more documents from the plurality of documents;

training a classifier using each selected document as a training example;

supplying one or more of the plurality of documents responsive to the search request documents to the trained classifier to obtain a measure of similarity between each document supplied to the trained classifier and a document category represented by the selected documents;

re-ranking the initial ordering of each document having a measure of similarity which exceeds a specified threshold; and

presenting the re-ranked search results to a user on a graphical display device.

10. The computer program product of claim 9, wherein re-ranking the initial ordering comprises, ranking each document having a measure of similarity which exceeds a specified threshold ahead of other documents in the initial search result.

11. The computer program product of claim 9, wherein the selection of one or more documents from the plurality of documents is received in response to monitoring what documents referenced in the initial search results are accessed by the user and prompting the user to indicate whether each accessed document is relevant to what the user is searching for.

12. The computer program product of claim 9, wherein the search string includes a plurality of search terms.

13. The computer program product of claim 9, wherein the classifier is one of a naï

ve Bayes classifier, a linear classifier, a latent semantic indexing classifier and an artificial neural network.

14. The computer program product of claim 9, wherein the selection of one or more documents from the plurality of documents indicates at least one document relevant to what the user is actually searching for, and wherein the at least one document is used as a positive training example to train the classifier.

15. The computer program product of claim 9, wherein the selection of one or more documents from the plurality of documents indicates at least one document not relevant to what the user is actually searching for, and wherein the at least one document is used as a negative training example to train the classifier.

16. The computer program product of claim 9, wherein the code is further configured for:

supplying one or more documents not included in the initial search request documents to the trained classifier to obtain the measure of similarity between each document supplied to the trained classifier and the document category represented by the selected documents; and

augmenting the initial search results with any of the documents not included in the initial search results which have a measure of similarity which exceeds the specified threshold.

17. A system, comprising:

a processor; and

a memory containing an application program configured for processing search results, which, when executed on the processor is configured to perform an operation comprising;

receiving an initial set of search results, wherein the initial set of search includes a plurality of documents responsive to a search request, ranked in an initial ordering,

receiving a selection of one or more documents from the plurality of documents,

training a classifier using each selected document as a training example,

supplying one or more of the plurality of documents responsive to the search request documents to the trained classifier to obtain a measure of similarity between each document supplied to the trained classifier and a document category represented by the selected documents,

re-ranking the initial ordering of each document having a measure of similarity which exceeds a specified threshold, and

presenting the re-ranked search results to a user on a graphical display device.

18. The system of claim 17, wherein re-ranking the initial ordering comprises, ranking each document having a measure of similarity which exceeds a specified threshold ahead of other documents in the initial search result.

19. The system of claim 17, wherein the selection of one or more documents from the plurality of documents is received in response to monitoring what documents referenced in the initial search results are accessed by the user and prompting the user to indicate whether each accessed document is relevant to what the user is searching for.

20. The system of claim 17, wherein the search string includes a plurality of search terms.

21. The system of claim 17, wherein the classifier is one of a naï

ve Bayes classifier, a linear classifier, a latent semantic indexing classifier and an artificial neural network.

22. The system of claim 17, wherein the selection of one or more documents from the plurality of documents indicates at least one document relevant to what the user is actually searching for, and wherein the at least one document is used as a positive training example to train the classifier.

23. The system of claim 17, wherein the selection of one or more documents from the plurality of documents indicates at least one document not relevant to what the user is actually searching for, and wherein the at least one document is used as a negative training example to train the classifier.

24. The system of claim 17, wherein the operation further comprises:

supplying one or more documents not included in the initial search request documents to the trained classifier to obtain the measure of similarity between each document supplied to the trained classifier and a document category represented by the selected documents; and

augmenting the initial search results with any of the documents not included in the initial search results which have a measure of similarity which exceeds the specified threshold.