Query refinement method for searching documents

• US 5,987,457 A
• Filed: 11/25/1997
• Issued: 11/16/1999
• Est. Priority Date: 11/25/1997
• Status: Expired due to Term

• Current Assignee: Hanger Solutions, LLC (IP Investments Group LLC)
• Original Assignee: Acceleration Software International Corporation
• Inventors: Ballard, Clinton L.
• Primary Examiner(s): Kulik, Paul V.
• Assistant Examiner(s): Robinson, Greta L.
• Application Number:
  US08/969,961
• Time in Patent Office: 721 Days
• Field of Search: 707/5, 707/3, 707/2, 707/1, 707/10
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9532 Query formulation
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A user views search results and subjectively determines if a document is desirable or undesirable. Only documents categorized by the user are analyzed for deriving a list of prospective keywords. The frequency of occurrence of each word of each document is derived. Keywords that occur only in desirable documents are good keywords. Keywords that occur only in undesirable documents are bad keywords. Keywords that occurs in both types are dirty keywords. The best keywords are the good keywords with the highest frequency of occurrence. The worst keywords are the bad keywords with the highest frequency of occurrence. A new query phrase includes the highest ranked good keywords and performs filtering using the highest ranked bad keywords. Key phrases are derived to clean dirty keywords into good key phrases. A key phrase also is derived from a good keyword and replaces the good keyword to narrow a search.

Claims

1. A method for refining an initial query phrase to search for documents of interest to a user, comprising the steps of:

categorizing at least one document found in a search using the initial query phrase as of interest based upon feedback from the user;

categorizing at least one other document found in the search using the initial query phrase as not of interest based upon feedback from the user;

generating a list of keywords by analyzing only the categorized documents;

ranking as first keywords, the keywords in the list of keywords which occur in only the documents of interest;

ranking as second keywords, the keywords in the list of keywords which occur in only the documents not of interest;

forming a refined query phrase to search for documents which include one or more of a plurality of the highest ranked first keywords, and to filter out documents which include any one or more of a plurality of the highest ranked second keywords.

2. The method of claim 1, wherein the step of ranking as first keywords comprises ranking by number of documents of interest that each respective first keyword occurs in;

and wherein the step of ranking as second keywords comprises ranking by number of documents not of interest that each respective second keyword occurs in.

3. The method of claim 2, wherein ranking among a plurality of first keywords having a common number of documents of interest that each one of said plurality of first keywords occurs in is by number of occurrences.

4. The method of claim 1, wherein the step of ranking as first keywords comprises ranking by number of occurrences of each respective first keyword, and wherein the step of ranking as second keywords comprises ranking by number of occurrences of each respective second keyword.

5. The method of claim 1, wherein of all the documents found in the search, the documents not categorized as `of interest` or `not of interest` are uncategorized.

6. The method of claim 1, wherein a user views a plurality of documents, and of all the documents viewed by the user, the documents not categorized as `of interest` by the user are categorized as `not of interest.`

7. The method of claim 1, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest based upon a ratio of the number of documents of interest that the third keyword occurs in to the number of documents not of interest that the third keyword occurs in;

wherein the step of forming further comprises forming a refined query phrase to search for documents which include one or more of a plurality of the highest ranked third keywords.

8. The method of claim 1, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest based upon a ratio of the number of occurrences in the documents of interest to the number of occurrences in the documents not of interest; and

wherein the step of forming further comprises forming a refined query phrase to search for documents which include one or more of a plurality of the highest ranked third keywords.

9. The method of claim 1, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest based upon a ratio of the number of occurrences in the documents of interest to the total number of occurrences in both the documents of interest and documents not of interest; and

wherein the step of forming further comprises forming a refined query phrase to search for documents which include one or more of a plurality of the highest ranked third keywords.

10. The method of claim 1, further comprising the steps of:

identifying as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest;

for at least one of the third keywords, replacing said at least one of the third keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said at least one of the third keywords; and

wherein the step of forming further comprises forming a refined query phrase to search for documents which include one or more of the plurality of key phrases.

11. The method of claim 10, wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one of the third keywords in a document of interest and combining said one of the third keywords with either one of a preceding word or following word in the respective occurrence.

12. The method of claim 1, further comprising for at least one of the first keywords, replacing said at least one of the first keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said at least one of the first keywords, and wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one of the first keywords and combining said one of the first keywords with either one of a preceding word or following word in the respective occurrence;

and wherein the step of forming further comprises forming a refined query phrase to search for documents which include one or more of the plurality of key phrases.

13. A method for refining an initial query phrase to search for web pages on the world wide web that are of interest to a user, comprising the steps of:

categorizing at least one web page found in a search using the initial query phrase as of interest based upon feedback from the user;

categorizing at least one other web page found in the search using the initial query phrase as not of interest based upon feedback from the user;

generating a list of keywords by analyzing only the categorized web pages;

ranking as first keywords, the keywords in the list of keywords which occur in only the web pages of interest;

ranking as second keywords, the keywords in the list of keywords which occur in only the web pages not of interest;

forming a refined query phrase to search for web pages which include one or more of a plurality of the highest ranked first keywords, and to filter out web pages which include any one or more of a plurality of the highest ranked second keywords.

14. The method of claim 13, wherein the step of ranking as first keywords comprises ranking by number of web pages of interest that each respective first keyword occurs in;

and wherein the step of ranking as second keywords comprises ranking by number of web pages not of interest that each respective second keyword occurs in.

15. The method of claim 14, wherein ranking among a plurality of first keywords having a common number of web pages of interest that each one of said plurality of first keywords occurs in is by number of occurrences.

16. The method of claim 13, wherein the step of ranking as first keywords comprises ranking by number of occurrences of each respective first keyword, and wherein the step of ranking as second keywords comprises ranking by number of occurrences of each respective second keyword.

17. The method of claim 13, wherein of all the web pages found in the search, the web pages not categorized as `of interest` or `not of interest` are uncategorized.

18. The method of claim 13, wherein a user views a plurality of web pages, and of all the web pages viewed by the user, the web pages not categorized as `of interest` by the user are categorized as `not of interest.`

19. The method of claim 13, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the web pages of interest and the web pages not of interest based upon a ratio of the number of web pages of interest that the third keyword occurs in to the number of web pages not of interest that the third keyword occurs in;

wherein the step of forming further comprises forming a refined query phrase to search for web pages which include one or more of a plurality of the highest ranked third keywords.

20. The method of claim 13, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the web pages of interest and the web pages not of interest based upon a ratio of the number of occurrences in the web pages of interest to the number of occurrences in the web pages not of interest; and

wherein the step of forming further comprises forming a refined query phrase to search for web pages which include one or more of a plurality of the highest ranked third keywords.

21. The method of claim 13, further comprising the step of:

ranking as third keywords, the keywords in the list of keywords which occur in both the web pages of interest and the web pages not of interest based upon a ratio of the number of occurrences in the web pages of interest to the total number of occurrences in both the we pages of interest and web pages not of interest; and

wherein the step of forming further comprises forming a refined query phrase to search for web pages which include one or more of a plurality of the highest ranked third keywords.

22. The method of claim 13, further comprising the steps of:

identifying as third keywords, the keywords in the list of keywords which occur in both the web pages of interest and the web pages not of interest;

for at least one of the third keywords, replacing said at least one of the third keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said at least one of the third keywords; and

wherein the step of forming further comprises forming a refined query phrase to search for web pages which include one or more of the plurality of key phrases.

23. The method of claim 22, wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one of the third keywords in a web page of interest and combining said one of the third keywords with either one of a preceding word or following word in the respective occurrence.

24. The method of claim 13, further comprising for at least one of the first keywords, replacing said at least one of the first keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said at least one of the first keywords, and wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one of the first keywords and combining said one of the first keywords with either one of a preceding word or following word in the respective occurrence;

and wherein the step of forming further comprises forming a refined query phrase to search for web pages which include one or more of the plurality of key phrases.

25. A system for refining an initial query phrase to search for documents of interest to a user, comprising:

a display device;

an input device;

means for accessing a document domain;

processing means for categorizing at least one document found in a search using the initial query phrase as of interest based upon feedback from the user;

processing means for categorizing at least one other document found in the search using the initial query phrase as not of interest based upon feedback from the user,

processing means for ranking as first keywords, words which occur in only the documents of interest;

processing means for ranking as second keywords, words which occur in only the documents not of interest;

processing means for forming a refined query phrase to search for documents which include one or more of a plurality of the highest ranked first keywords, and to filter out documents which include any one or more of a plurality of the highest ranked second keywords.

26. The system of claim 25, wherein of all the documents found in the search, the documents not categorized as `of interest` or `not of interest` are uncategorized.

27. The system of claim 25, wherein a user views a plurality of documents, and of all the documents viewed by the user, the documents not categorized as `of interest` by the user are categorized as `not of interest.`

28. The system of claim 25, wherein the processing means for ranking as first keywords comprises means for ranking by number of documents of interest that each respective first keyword occurs in;

and wherein the processing means for ranking as second keywords comprises means for ranking by number of documents not of interest that each respective second keyword occurs in.

29. The system of claim 28, wherein ranking among a plurality of first keywords having a common number of documents of interest that each one of said plurality of first keywords occurs in is by number of occurrences.

30. The system of claim 25, wherein the processing means for ranking as first keywords comprises means for ranking by number of occurrences of each respective first keyword, and wherein the processing means for ranking as second keywords comprises means for ranking by number of occurrences of each respective second keyword.

31. The system of claim 25, further comprising:

processing means for ranking as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest based upon a ratio of the number of documents of interest that the third keyword occurs in to the number of documents not of interest that the third keyword occurs in;

wherein the processing means for forming forms the refined query phrase to encompass documents which include one or more of a plurality of the highest ranked third keywords.

32. The system of claim 25, further comprising:

processing means for ranking as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest based upon a ratio of the number of occurrences in the documents of interest to the total number of occurrences in both the documents of interest and documents not of interest; and

wherein the processing means for forming forms the refined query phrase to encompass documents which include one or more of a plurality of the highest ranked third keywords.

33. The system of claim 25, further comprising:

processing means for identifying as third keywords, the keywords in the list of keywords which occur in both the documents of interest and the documents not of interest;

processing means for replacing at least one of the third keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said at least one of the third keywords; and

wherein the processing means for forming forms the refined query phrase to encompass documents which include one or more of the plurality of key phrases.

34. A method for refining an initial query phrase to search for documents of interest to a user, comprising the steps of:

categorizing at least one document found in a search using the initial query phrase as of interest based upon feedback from the user;

generating a list of words occurring in the at least one document categorized as of interest;

ranking the words in said list of words as keywords;

for at least one of the ranked keywords, replacing said one of the ranked keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said one ranked keyword, and wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one ranked keyword and combining said one ranked keyword with either one of a preceding word or following word in the respective occurrence of said keyword in a document of interest;

forming a refined query phrase to search for documents which include one or more of a plurality of the keywords and key phrases.

35. The method of claim 34, wherein the step of ranking comprises ranking the words in said list of words by number of documents of interest that each respective word occurs in.

36. The method of claim 35, wherein ranking among a plurality of words having a common number of documents of interest that each one of said plurality of words occurs in is by number of occurrences.

37. The method of claim 34, wherein the step of ranking comprises ranking the words in said list of words by number of occurrences in the documents categorized as of interest.

38. The method of claim 35, wherein the step of ranking comprises filtering out words in said list of words which occur in a list of words excluded from being keywords.

39. A system for refining an initial query phrase to search for documents of interest to a user, comprising:

a display device;

an input device;

means for accessing a document domain;

processing means for categorizing at least one document found in a search using the initial query phrase as of interest based upon feedback from the user;

processing means for ranking keywords which occur in the documents of interest;

for at least one of the ranked keywords, processing means for replacing said one of the ranked keywords with a plurality of key phrases, wherein each one of the plurality of key phrases includes said one ranked keyword, and wherein each one of the plurality of key phrases is obtained by finding an occurrence of said one ranked keyword and combining said one ranked keyword with either one of a preceding word or following word in the respective occurrence of said keyword in a document of interest; and

processing means for forming a refined query phrase to search for documents which include one or more of a plurality of the keywords and key phrases.

40. The system of claim 39, wherein the processing means for ranking comprises means for ranking the words in said list of words by number of documents of interest that each respective word occurs in.

41. The system of claim 40, wherein the processing means for ranking comprises means for ranking among a plurality of words having a common number of documents of interest that each one of said plurality of words occurs in is by number of occurrences.

42. The system of claim 39, wherein the processing means for ranking comprises means for ranking the words in said list of words by number of occurrences in the documents categorized as of interest.

43. The system of claim 42, wherein the processing means for ranking comprises means for filtering out words in said list of words which occur in a list of words excluded from being keywords.