SEARCHING, SORTING, AND DISPLAYING VIDEO CLIPS AND SOUND FILES BY RELEVANCE

• US 20090313236A1
• Filed: 06/13/2008
• Published: 12/17/2009
• Est. Priority Date: 06/13/2008
• Status: Active Grant

• Current Assignee: Relevant Holdings LLC
• Original Assignee: Inform, Inc. (CDC Corporation)
• Inventors: Hernacki, Paul Matthew, Peters, Gregory Alton
• Granted Patent: US 8,364,693 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A documents database has a plurality of documents, including but not limited to text files, video clips and sound files. Each document is associated with at least one category of a plurality of categories in a categories database, and each category has at least one keyword. A search request having at least one search term is received from a user, and a categories database is searched for categories having a keyword corresponding to the user search term to identify first level categories. The other keywords from the identified first level categories are retrieved and the documents database is searched for documents having a user search term or a retrieved keyword. The identified documents are then ranked and presented to the user. Other search expansion techniques, and display techniques, are also discussed.

Claims

1. A method for searching for documents in a documents database, the documents database having a plurality of documents, each document being associated with at least one category of a plurality of categories in a categories database, each category comprising at least one keyword, the method comprising:

receiving a search request from a user, the search request having at least one user search term;

searching a categories database for categories having at least keyword corresponding to at least one said user search term to identify at least one first level category;

identifying first level documents by searching the documents associated with at least one first level category for the presence of at least one user search term;

ranking the first level documents; and

presenting to the user at least part of one ranked document, at least some information about at least one ranked document, or a link to the document.

2. The method of claim 1 and, prior to presenting, further comprising:

if the number of first level documents is less than a predetermined number of documents then identifying at least one second level category related to at least one first level category;

identifying second level documents by searching the documents associated with at least one second level category for the presence of at least one user search term; and

wherein ranking is performed on the combination of the first level documents and the second level documents.

3. The method of claim 2 and further comprising defining two categories to be related if at least one of the following occurs:

an administrator designates that the two categories are related;

or

both categories have at least one keyword in common;

or

a document is received and the provider of the document has designated that the document is associated with both categories;

or

at least a predetermined number of user searches have been received which used search terms but none of the search terms were keywords in both categories.

4. The method of claim 1 and, prior to presenting the ranked documents, further comprising:

if the number of first level documents is less than a predetermined number of documents then identifying at least one second level category related to at least one first level category;

identifying second level documents by searching the documents associated with at least one second level category for the presence of at least one user search term;

if the number of the first level documents and the second level documents is greater than the predetermined number of documents then performing ranking on the combination of the first level documents and the second level documents;

if the total number of the first level documents and the second level documents is less than the predetermined number of documents then identifying at least one third level category related to at least one second level category;

identifying third level documents by searching the documents associated with at least one third level category for the presence of at least one user search term; and

wherein ranking is performed on the combination of the first level documents, the second level documents, and the third level documents.

5. The method of claim 1 and, prior to presenting the ranked documents, further comprising:

if the number of first level documents is less than a predetermined number of documents then identifying at least two second level categories;

ranking the second level categories;

identifying second level documents by searching the documents associated with the highest ranked second level category for the presence of at least one user search term;

if the total number of the first level documents and the second level documents is greater than the predetermined number of documents then performing ranking on the combination of the first level documents and the second level documents;

if the total number of the first level documents and the second level documents is less than the predetermined number of documents then identifying additional documents as second level documents by searching the documents associated with the second highest ranked second level category for at least one user search term;

if the total number of the first level documents and the second level documents is greater than the predetermined number of documents then performing ranking on the combination of the first level documents and the second level documents.

6. The method of claim 1 and further comprising applying event condition criteria to the documents to change the ranking of at least one document.

7. The method of claim 6 wherein the event condition criteria is applied to remove at least one document prior to ranking the documents.

8. The method of claim 6 wherein the event condition criteria is applied to change the ranking of at least one document after ranking the documents.

9. The method of claim 6 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has occurred then either discarding the document or changing the ranking of the document.

10. The method of claim 9 wherein the event condition is a date.

11. The method of claim 9 wherein the event condition is an election.

12. The method of claim 9 wherein the event condition is a sports game.

13. The method of claim 9 wherein the event condition is a championship game.

14. The method of claim 6 and further comprising discarding the document if the ranking of the document is below a predetermined threshold.

15. The method of claim 6 wherein the ranking of the document is reduced.

16. The method of claim 6 wherein the ranking of the document is abruptly reduced.

17. The method of claim 6 wherein the ranking of the document is reduced over a predetermined period of time.

18. The method of claim 6 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has not occurred then increasing the ranking of the document.

19. The method of claim 18 wherein the ranking of the document is increased over a predetermined period of time occurring before the event condition.

20. The method of claim 1 wherein the search request further comprises search setting limitations and, after searching the documents database, the method further comprises applying the search setting limitations to the documents to change the ranking at least one document to the search request.

21. The method of claim 20 wherein applying the search setting limitations comprises:

reading a search setting limitation associated with the search;

if a document does not meet the search limitation then discarding the document.

22. The method of claim 20 wherein the search setting limitation is at least one of an occurrence start date, an occurrence end date, a posted start date, a posted end date, a speech, a town-hall meeting, a debate, an advertisement, an interview, a candidate for a political office, a holder of a political office, a media document, a special interest group document, a political party document, a user-generated document, a type of political race, a year of a political race, a country, a state, a county, a city, a municipality, or a regional governmental body.

23. The method of claim 5 and, after ranking the second level categories, discarding second level categories which have a ranking below a predetermined ranking level.

24. The method of claim 5 wherein ranking the second level categories comprises ranking the second level categories by the weight of a relationship between a first level category and a related second level category.

25. The method of claim 24 wherein the weight is determined by at least one of:

a weight approved by a vote of administrators;

the number of keywords that the first level category and the related second level category have in common;

the number of documents that the first level category and the related second level category have in common;

or the number of user searches which do not have a search term which is common to the first level category and the related second level category.

26. A method of ranking documents produced in response to a search request to a search engine, the search request comprising a plurality of search terms, the method comprising:

scoring the documents in accordance with a predetermined scoring process;

altering the scoring of the documents based upon at least one characteristic selected from the following characteristics;

a normalized average rating for the document based upon ratings of the document by users who have accessed the document, the document type, the source of the document, the number of times the document has been accessed, the primary person featured in the document, the title of the primary person featured in the document, or the political office of the primary person featured in the document; and

ranking the documents in accordance with the results of the altered scoring.

27. The method of claim 26 wherein the document type is a predetermined one of the following:

a speech, a town-hall meeting, a debate, an advertisement, or an interview.

28. The method of claim 26 wherein the source of the document is a predetermined one of the following:

a candidate for a political office, the current holder of a political office, a former holder of a political office, a media organization, a special interest group, a political party, or from a user of the search engine.

29. The method of claim 26 wherein the primary person featured in the document is a predetermined one of the following:

a candidate for a political office, the current holder of a political office, or a former holder of a political office.

30. The method of claim 26 wherein the title of the primary person featured in the document is a predetermined one of the following:

President, Vice President, presidential press secretary, presidential cabinet member, other presidential administration person, other presidential officials, foreign head of state, foreign ambassador, United States senator, United States representative, a major national political figure, a major national political advisor, a major media figure, a military commander, a state governor, a state lieutenant governor, other major state political figure, a major state media figure, a state senator, a state representative, a state department secretary, a county commissioner, a city commissioner, a county official, a city official, a city mayor, or a city councilperson.

31. The method of claim 26 and further comprising applying event condition criteria to the documents to change the scoring of at least one document.

32. The method of claim 31 wherein the event condition criteria is applied to remove at least one document prior to ranking the documents.

33. The method of claim 31 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has occurred then either discarding the document or changing the scoring of the document.

34. The method of claim 33 wherein the event condition is a date.

35. The method of claim 33 wherein the event condition is an election.

36. The method of claim 33 wherein the event condition is a sports game.

37. The method of claim 33 wherein the event condition is a championship game.

38. The method of claim 31 and further comprising discarding the document if the scoring of the document is below a predetermined threshold.

39. The method of claim 31 wherein the scoring of the document is reduced.

40. The method of claim 31 wherein the scoring of the document is abruptly reduced.

41. The method of claim 31 wherein the scoring of the document is reduced over a predetermined period of time.

42. The method of claim 31 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has not occurred then increasing the scoring of the document.

43. The method of claim 42 wherein the scoring of the document is increased over a predetermined period of time occurring before the event condition.

44. A method of ranking documents produced by a search, comprising:

receiving search result documents, each document having an initial ranking;

applying event condition criteria to the documents to change the ranking of at least one document to the search request; and

providing re-ranked search result documents.

45. The method of claim 44 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has occurred then discarding the document.

46. The method of claim 45 wherein the event condition is a date.

47. The method of claim 45 wherein the event condition is an election.

48. The method of claim 44 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has occurred then reducing the ranking of the document to the search request.

49. The method of claim 48 and further comprising discarding the document if the ranking of the document is below a predetermined threshold.

50. The method of claim 48 wherein the ranking of the document is abruptly reduced.

51. The method of claim 48 wherein the ranking of the document is reduced over a predetermined period of time.

52. The method of claim 44 wherein applying event condition criteria comprises:

reading an event condition associated with a document;

if the event condition has not occurred then increasing the ranking of the document.

53. The method of claim 53 wherein the ranking of the document is increased over a predetermined period of time occurring before the event condition.

54. The method of claim 53 wherein the event condition is a date.

55. The method of claim 53 wherein the event condition is an election.

56. A method of ranking categories, each category having at least one keyword, each category having a plurality of associated documents, a search engine identifying relevant categories based upon a comparison of category keywords with search terms in a search request provided to the search engine, the method comprising:

assigning a score to each category using the relationship

$\mathrm{Issue}<br><br><br><br>\mathrm{Rank}<br><br>\left(X\right)=k<br><br><br><br>1*\frac{{\mathrm{NoS}}_x}{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br><br><br>{\mathrm{NoS}}_i}+k<br><br><br><br>2*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}}{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}}+k<br><br><br><br>3*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}*{\mathrm{NoS}}_i}{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}*{\mathrm{NoS}}_i}+k<br><br><br><br>4*{\mathrm{NoV}}_x+k<br><br><br><br>5*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}*{\mathrm{NoV}}_i}{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}*{\mathrm{NoV}}_i}$

wherein;

X denotes a category;

k1 is a coefficient representing the number of searches made regarding the category X;

NoSx is the number of searches made regarding the category X;

N is the number of category defined;

NoSi is the number of searches made for each category I;

k2 is a coefficient representing the weight of the relationship between the category X and all other categories;

Wix is the weighted relationship between the category X and another category I;

k3 is a coefficient representing the weighted number of searches made against related category;

k4 is a coefficient representing the number of video clips related to the category X;

NoVx is the number of video clips related to the category X; and

k5 is a coefficient representing the number of video clips related to the category related to the category X; and

ranking the categories based upon the assigned scores.

57. The method of claim 56 and further comprising:

assigning to a document the highest score of the categories relevant to the document; and

ranking the documents based upon the assigned scores.

58. A method of ranking persons, each person being associated with at least one document, a search engine identifying relevant documents based upon a search request for a person, the method comprising:

assigning a score to each person using the relationship

$\mathrm{People}<br><br><br><br>\mathrm{Rank}<br><br>\left(Y\right)=\mathrm{kp}*{\mathrm{BoI}}_y+\mathrm{kn}*\frac{{\mathrm{NoS}}_y}{\underset{i=1}{\overset{M}{\sum <br><br>}}<br><br><br><br>{\mathrm{NoS}}_i}+k<br><br><br><br>d*\mathrm{PoE}<br><br>\left(\mathrm{DateOfElection}\right)$

wherein;

Y denotes a person;

kp is a coefficient for the broadness of impact rank;

Boly is the broadness of impact rank of the person Y;

kn is a coefficient for the number of searches made regarding the person Y;

NoSy is the number of searches made regarding the person Y;

NoSi is the number of searchers made regarding each person;

kd is a coefficient for the proximity of an election involving the person Y; and

PoE(DateOfElection) is a relevance factor based upon the proximity of an election involving the person Y;

$\mathrm{PoE}<br><br>\left(\mathrm{DateOfElection}\right)=\{\begin{array}{c}\frac{k_{\mathrm{PoE}<br><br><br><br>1}}{\begin{array}{c}\mathrm{DateOfElection}-\\ \mathrm{CurrentDate}\end{array}},\mathrm{CurrentDate}<<br><br>\mathrm{DateOfElection},\mathrm{in}<br><br><br><br>\mathrm{months}\\ 1,\mathrm{CurrentDate}=\mathrm{DateOfElection}\\ \frac{k_{\mathrm{PoE}<br><br><br><br>2}}{\begin{array}{c}\mathrm{CurrentDate}-\\ \mathrm{DateOfElection}\end{array}},\mathrm{CurrentDate}><br><br>\mathrm{DateOfElection},\mathrm{in}<br><br><br><br>\mathrm{days}\end{array}$

kpoE1 is a coefficient for proximity of elections for the time before the date of the election involving the person Y;

kpoE2 is a coefficient for proximity of elections for the time after the date of the election involving the person Y; and

ranking the persons based upon the assigned scores.

59. The method of claim 58 and further comprising:

assigning to a document the highest score of the persons relevant to the document; and

ranking the documents based upon the assigned scores.

60. The method of claim 58 wherein the coefficient for the broadness of impact rank is a function of the type of political office.

61. The method of claim 60 wherein the broadness of impact rank of the person Y is determined at least in part by whether the person Y is a current holder of the political office, a former holder of the political office, or a candidate for the political office.

62. A method of indicating documents related to categories and persons, comprising:

identifying categories to which each document is related;

identifying the persons to which each document is related;

ranking the categories;

ranking the persons;

establishing an array comprising a plurality of cells, a first predetermined cell indicating the number of documents which are related to both the highest ranked category and the highest ranked person;

a second predetermined cell indicating the number of documents which are related to both the second highest ranked category and the highest ranked person;

a third predetermined cell indicating the number of documents which are related to both the highest ranked category and the second highest ranked person;

a fourth predetermined cell indicating the number of documents which are related to both the second highest ranked category and the second highest ranked person;

establishing links from the cells to the documents; and

displaying the array.

63. The method of claim 62 wherein the rank of a category (X) is determined by the following:

$\mathrm{Issue}<br><br><br><br>\mathrm{Rank}<br><br>\left(X\right)=k<br><br><br><br>1*\frac{{\mathrm{NoS}}_x}{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br><br><br>{\mathrm{NoS}}_i}+k<br><br><br><br>2*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}}{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}}+k<br><br><br><br>3*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}*{\mathrm{NoS}}_i}{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}*{\mathrm{NoS}}_i}+k<br><br><br><br>4*{\mathrm{NoV}}_x+k<br><br><br><br>5*\frac{\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ix}}*{\mathrm{NoV}}_i}{\underset{j=1}{\overset{N}{\sum <br><br>}}<br><br>\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>W_{\mathrm{ij}}*{\mathrm{NoV}}_i}$

where;

k1 is a coefficient representing the number of searches made regarding the category X;

NoSx is the number of searches made regarding the category X;

N is the number of category defined;

NoSi is the number of searches made for each category I;

k2 is a coefficient representing the weight of the relationship between the category X and all other category;

Wix is the weighted relationship between the category X and another category J;

k3 is a coefficient representing the weighted number of searches made against related category;

k4 is a coefficient representing the number of video clips related to the category X;

NoVx is the number of video clips related to the category X; and

k5 is a coefficient representing the number of video clips related to the category related to the category X.

64. The method of claim 62 wherein the rank of a person (Y) is determined by the following:

$\mathrm{People}<br><br><br><br>\mathrm{Rank}<br><br>\left(Y\right)=\mathrm{kp}*{\mathrm{BoI}}_y+\mathrm{kn}*\frac{{\mathrm{NoS}}_y}{\underset{i=1}{\overset{M}{\sum <br><br>}}<br><br><br><br>{\mathrm{NoS}}_i}+\mathrm{kd}*\mathrm{PoE}<br><br>\left(\mathrm{DateOfElection}\right)$

where;

kp is a coefficient for the broadness of impact rank;

Boly is the broadness of impact rank of the person Y;

kn is a coefficient for the number of searches made regarding the person Y;

NoSy is the number of searches made regarding the person Y;

NoSi is the number of searchers made regarding each person;

kd is a coefficient for the proximity of an election involving the person Y; and

PoE(DateOfElection) is a relevance factor based upon the proximity of an election involving the person Y;

$\mathrm{PoE}<br><br>\left(\mathrm{DateOfElection}\right)=\{\begin{array}{c}\frac{k_{\mathrm{PoE}<br><br><br><br>1}}{\begin{array}{c}\mathrm{DateOfElection}-\\ \mathrm{CurrentDate}\end{array}},\mathrm{CurrentDate}<<br><br>\mathrm{DateOfElection},\mathrm{in}<br><br><br><br>\mathrm{months}\\ 1,\mathrm{CurrentDate}=\mathrm{DateOfElection}\\ \frac{k_{\mathrm{PoE}<br><br><br><br>2}}{\begin{array}{c}\mathrm{CurrentDate}-\\ \mathrm{DateOfElection}\end{array}},\mathrm{CurrentDate}><br><br>\mathrm{DateOfElection},\mathrm{in}<br><br><br><br>\mathrm{days}\end{array}$

kpoE1 is a coefficient for proximity of elections for the time before the date of the election involving the person Y;

kpoE2 is a coefficient for proximity of elections for the time after the date of the election involving the person Y.

65. The method of claim 64 wherein the coefficient for the broadness of impact rank is a function of the type of political office.

66. The method of claim 65 wherein the broadness of impact rank of the person Y is determined at least in part by whether the person Y is a current holder of the political office, a former holder of the political office, or a candidate for the political office.

67. The method of claim 62 wherein the documents related to a cell are ranked, the documents are video files, and the ranking of the documents is based at least in part on at least one of the following:

whether the video is of a speech, a town-hall meeting, a debate, an advertisement, or an interview; and

whether the source of the video is a candidate for or the holder of a political office, a media organization, a special interest group, or a political party.

68. The method of claim 62 wherein the array is a linear array of rows and columns, the first row being a header row indicating a category, the first column being a header column indicating a person, and at least some of the cells of the array indicating the number of documents corresponding to both the rank of the category of that column in which the cell is located and the rank of the person of that row in which the cell is located.