SYSTEM AND METHOD FOR QUERY EXPANSION USING TOOLTIPS

• US 20100076984A1
• Filed: 03/26/2009
• Published: 03/25/2010
• Est. Priority Date: 03/27/2008
• Status: Abandoned Application

• Current Assignee: Hotgrinds Canada
• Original Assignee: Hotgrinds Canada
• Inventors: Howell, Matthias, Papadopoullos, Alkis, Vogel, Claude
• Application Number:
  US12/412,316
• Publication Number:
  US 20100076984A1
• Time in Patent Office: Days
• US Class Current: 707/749
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence

Abstract

A system and method for query expansion allows the refinement and expansion of a keyword query search by combining a key concept with semantically related concepts proposed by the system and associated with that key concept. The semantically related concepts may be grouped together in a cluster, which is then presented to the user in the form of a tooltip. Once a semantically related concept is selected from the cluster, a new search is launched. This new search may use as query terms the combination of at least part of the key concept and the selected semantically related concept to thereby expand the original search.

Claims

1. A method for expanding a keyword query search returning a set of documents related to the keyword query, the search expanding method comprising:

characterizing one keyword in the returned set of documents;

for each characterized keyword, proposing a cluster of concepts semantically related thereto;

upon selection of a semantically related concept in the cluster associated to one keyword, launching a new search using a query containing at least the selected semantically related concept and at least one keyword thereby expanding the original keyword query search.

2. The search expanding method as recited in claim 1 further comprising processing the set of documents related to the query to index and classifying the set of documents into categories of different concepts.

3. The search expanding method as recited in claim 2, wherein the indexing and the classifying the set of documents include:

parsing and tokenizing text objects contained in the set of documents;

extracting precursors from the parsed text objects in the set of documents, the precursors corresponding to the keywords; and

linking the extracted precursors to corresponding topic-specific taxonomies and ontologies so as to determine concepts semantically related to the extracted precursors, the semantically related concepts forming the cluster.

4. The search expanding method as recited in claim 1, wherein characterizing one keyword includes highlighting the keyword in the returned set of documents.

5. The search expanding method as recited in claim 4, wherein characterizing one keyword includes displaying the plurality of keywords in a different format, in the returned set of documents.

6. The search expanding method as recited in claim 1, wherein proposing the cluster of concepts semantically related to the characterized keyword includes displaying the semantically related concepts using a tooltip.

7. The search expanding method as recited in claim 6, wherein displaying the tooltip includes adjusting a size of the tooltip according to a relative location of the keyword in the set of documents.

8. The search expanding method as recited in claim 7, wherein adjusting the size of the tooltip includes centering horizontally and vertically the tooltip within a screen.

9. The search expanding method as recited in claim 7, wherein adjusting the size of the tooltip includes defining a scrollable area.

10. The search expanding method as recited in claim 1, wherein the selection of the semantically related concept in the cluster includes clicking on a hyperlink corresponding to the semantically related concept.

11. The search expanding method as recited in claim 1, wherein launching the new search comprises combining the selected semantically related concept with the original query so as to form a new search query.

12. A system for expanding a keyword query search returning a set of documents related to the keyboard query, the search expanding system comprising:

means for characterizing one keyword in the returned set of documents;

means for proposing, for each characterized keyword, a cluster of concepts semantically related thereto;

means for selecting a semantically related concept in the cluster, associated to one keyword; and

means for launching a new search using a query containing at least the selected semantically related concept and at least one keyword as the query, thereby expanding the original keyboard entry search.

13. A system for expanding a keyword query search returning a set of documents related to the keyword entry, the search expanding system comprising:

a processor so configured as to characterize one keyword in the returned set of documents;

a display so configured as to display of a cluster of concepts semantically related to each characterized keyword;

a pointer so configured as to select a semantically related concept in the cluster; and

a query server for launching a new search with a query including at least the selected semantically related concept and at least one keyword, thereby expanding the original keyword query search.

14. The search expanding system as recited in claim 13 further comprising a semantic indexer that indexes and classifies the set of documents into categories of different concepts.

15. The search expanding system as recited in claim 14, wherein the semantic indexer includes a parser and a tokenizer so configured as to respectively parse and tokenize text objects contained in the set of documents.

16. The search expanding system as recited in claim 15, wherein the semantic indexer includes an extractor so configured as to extract precursors from the parsed text objects in the set of documents, the precursors corresponding to the keyword.

17. The search expanding system as recited in claim 16, wherein the semantic indexer is so configured as to link the extracted precursors to corresponding topic-specific taxonomies and ontologies so as to determine concepts semantically related to the extracted precursors, the semantically related concepts forming the cluster.

18. The search expanding system as recited in claim 13, wherein the processor uses highlighting to characterize each keyword in the processed set of documents.

19. The search expanding system as recited in claim 13, wherein the display uses a tooltip to display the cluster of semantically related concepts.

20. The search expanding system as recited in claim 19, wherein the tooltip has an adjustable size.

21. The search expanding system as recited in claim 19, wherein the tooltip includes a scrollable area.

22. The search expanding system as recited in claim 13, wherein the pointer is so configured as to select the semantically related concept by clicking on a hyperlink corresponding to the selected semantically related concept.

23. The search expanding system as recited in claim 13, wherein the query server launches a new search by combining the selected semantically related concept with the original query so as to form a new search query.