AUTOMATIC EXPERT IDENTIFICATION, RANKING AND LITERATURE SEARCH BASED ON AUTHORSHIP IN LARGE DOCUMENT COLLECTIONS

• US 20060248076A1
• Filed: 04/20/2006
• Published: 11/02/2006
• Est. Priority Date: 04/21/2005
• Status: Active Grant

• Current Assignee: Case Western Reserve University
• Original Assignee: Case Western Reserve University
• Inventors: Mehregany, Mehran, Zhang, Guo-Qiang, Troy, Adam D.
• Granted Patent: US 8,280,882 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3347 using vector based model
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

Disclosed is an author-centric search that facilitates identifying a source commonly associated with a topic by, for example, providing a ranked listing of experts in a field of knowledge related to a search phrase. The search phrase can be captured and parsed into the individual words (e.g., substrings) of the search phrase. Based on occurrences of the words in one or more documented communications, statistics can be generated to determine the relevancy of each documented communication in relation to the search phrase. Further, additional statistics can be generated describing the occurrence of multiple words in a documented communication and/or a distance of words between the search phrase words in a documented communication. The statistics can be utilized to generate expert scores. The expert scores can be sorted for and/or displayed to the user.

Claims

1. A computer implemented search system, comprising:

a component that calculates a relevancy score for at least one information source associated with an author; and

a module that generates an author score based in part on an expert weight and the relevancy score.

2. The system of claim 1, the at least one information source comprising one of a web page, a documented communication, a record, a publication, an article, an essay, a compilation, and a thesis.

3. The system of claim 1, the expert weight comprising a predefined value utilized to create a statistical preference for at least one characteristic of all the information sources.

4. The system of claim 1, the author comprising one of an individual, an institution, a publisher, a venue, a journal, a conference, a group, an entity, and a combination thereof.

5. The system of claim 1, further comprising:

an input component that receives a search phrase;

a synonym component that creates one or more synonym substrings related to the search phrase;

a parse component that parses the search phrase into one or more substrings and creates a substring set, the substring set comprising the one or more synonym substrings and the one or more substrings; and

a search result display component that formats and displays search results based in part on one or more author scores.

6. The system of claim 1, further comprising an array builder that initializes a value array for each information source and a weight array, the value array and the weight array comprising (n*(n+1)/2 array cells wherein n comprises the total number of substrings in the substring set.

7. The system of claim 6, further comprising an array filler that inputs an occurrence count for each of the substrings of the substring set in a first array based on the appearance of one of the substrings of the substring set in the information.

8. The system of claim 7, the array filler further inputs a distance count in at least one empty array cell of the count array for each possible combination of substrings of the substring set, the distance count comprising a minimum count of strings that appear between the substrings of each combination that occur in the at least one information source associated with the count array of the empty array cell, the minimum count subtracted from a desired constant number.

9. The system of claim 7, the array filler further inputs a predefined weight into each of the array cells of the weight array, the predefined weight relates to a level of significance assigned to one of the occurrence count and the distance count in a corresponding array cell of each count array associated with each information source.

10. The system of claim 1, the component calculates a dot product of the count array of each information source and the weight array to compute the relevancy score of the at least one information source.

11. A method for an author-centric search, comprising:

initializing a first data structure and a second data structure for each of a plurality of documented communications;

utilizing the first data structure and the second data structure to compute a relevancy score for each of the plurality of documented communications; and

determining a score for an author based in part on the relevancy score for each of the plurality of documented communications associated with the author.

12. The method of claim 11, further comprising:

prompting a user to enter a search string;

parsing the search string into one or more words;

filling at least one memory space of the first data structure for each documented communication with data based on the occurrence of the one or more words in the documented communication;

filling at least one memory space of the second data structure for each documented communication with a weighted value that signifies a statistical preference for the data in the corresponding memory space of the first data structure; and

executing a mathematical function with the data and the weighted values of the first and second data structures for each documented communication in order to compute the relevancy score for the documented communication.

13. The method of claim 12, the data further comprising combination data, the combination data based in part on a minimum count of information between words for each possible combination of the one or more words of the search string.

14. The method of claim 11, determining a score comprises calculating an aggregate of relevancy scores for each of the plurality of documented communications associated with the author.

15. The method of claim 11, determining a score further comprises establishing a communications weight to signify a statistical preference for one or more aspects of the plurality of documented communications.

16. The method of claim 15, the one or more aspects of the plurality of documented communications comprise at least one of a date of publication, a venue, a type, and a number of authors.

17. The method of claim 11, determining a score further comprises executing a mathematical function that utilizes the aggregate for each author and the communications weight to arrive at the score.

18. The method of claim 11, further comprising displaying a ranked listing of one or more author scores.

19. The method of claim 18, the displaying further comprising presenting the name of each author related to each listed author score and a hyperlink to each documented communication associated with the each author.

20. A system for an author-centric search, comprising:

means for initializing a first data structure and a second data structure for each of a plurality of documented communications;

means for utilizing the first data structure and the second data structure to compute a relevancy score for each of the plurality of documented communications; and

means for determining a score for an author based in part on the relevancy score for each of the plurality of documented communications associated with the author.