SYSTEMS AND METHODS OF RETRIEVING RELEVANT INFORMATION

• US 20080147649A1
• Filed: 02/14/2008
• Published: 06/19/2008
• Est. Priority Date: 01/10/2001
• Status: Active Grant

• Current Assignee: Looksmart Group, Inc.
• Original Assignee: Looksmart, Ltd (Looksmart Group, Inc.)
• Inventors: Yun, Yeogirl, Kim, Michael, Kim, Brian S., Dod, Anurag, Chung, Sudong
• Granted Patent: US 8,086,601 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

The present invention provides systems and methods of retrieving the pages according to the quality of the individual pages. The rank of a page for a keyword is a combination of intrinsic and extrinsic ranks. Intrinsic rank is the measure of the relevancy of a page to a given keyword as claimed by the author of the page while extrinsic rank is a measure of the relevancy of a page on a given keyword as indicated by other pages. The former is obtained from the analysis of the keyword matching in various parts of the page while the latter is obtained from the context-sensitive connectivity analysis of the links connecting the entire Web. The present invention also provides the methods to solve the self-consistent equation satisfied by the page weights iteratively in a very efficient way. The ranking mechanism for multi-word query is also described. Finally, the present invention provides a method to obtain the more relevant page weights by dividing the entire hypertext pages into distinct number of groups.

Claims

1-52. -52. (canceled)

53. A method of providing a search result set in response to a query, comprising:

a) ranking a first page, crawled from the Internet, for each of a first plurality of words on the page by combining

i) a first factor for use on the page of each of the plurality of words, with

ii) a second factor for use on other pages on the Internet of each of the plurality of words associated with an outbound link on the other pages to the first page;

b) ranking the other pages for each of another plurality of words on each of the other pages to form a database of pages ranked in accordance with words used on each of the pages in the database;

c) indexing the database for each of a plurality of selected words reflecting the ranking of each selected word on each of a plurality of pages; and

then

d) responding to a search query by searching the index to provide a search result set of pages from the database, each page of the set of pages ranked for words associated with the search query.

54. The method of claim 53 wherein ranking further comprises:

adjusting the first factor in accordance with a page weight for the first page;

adjusting the second factor in accordance with a page weight for each of said other pages; and

then

combining the adjusted first and second factors.

55. The method of claim 54 wherein adjusting the second factor further comprises:

adjusting the second factor in accordance with the number of outbound links to the first page.

56. The method of claim 54 wherein adjusting the first factor further comprises:

determining the page weight for the first page in accordance with a probability of a user viewing the first page as a result of viewing pages on the Internet in a random fashion.

57. The method of claim 54 wherein adjusting the first factor further comprises:

adjusting the first factor in accordance with a proximity on the first page between at least two of the first plurality of words.

58. The method of claim 54 wherein adjusting the first factor further comprises:

adjusting the first factor in accordance with a word order on the first page between at least two of the first plurality of words.

59. The method of claim 53 responding further comprises:

ordering the search result set of pages in accordance with the ranking of words associated with the query on the pages in the search result set.

60. The method of claim 53 wherein ranking a first page further comprises:

crawling the Internet without limitation to topic to select the first page.

61. A method of providing a search result set in response to a query, comprising:

receiving a query;

searching an index of words, prepared before the query was received, for words associated with the query to prepare a search result set; and

providing the search result set in an order in accordance with rankings in the index for each of the words associated with the query, the rankings combining a first factor based on use of the related word on a page selected from the Internet before the query was received and a second factor based on the use of the related word on other pages having outbound links to the selected pages.

62. The method of claim 61 further comprising:

adjusting the first and second factors in accordance with page weights for related pages determined in accordance with a probability of a user viewing the related page as a result of viewing pages on the Internet in a random fashion.

63. The method of claim 61 further comprising:

adjusting the first and second factors in accordance with a proximity between at least two of the words on one of the pages selected from the Internet.

64. The method of claim 61 further comprising:

adjusting the first and second factors in accordance with a word order of at least two of the words on one of the pages selected from the Internet.

65. The method of claim 61 further comprising:

adjusting the second factor in accordance with the number of other pages having outbound links to the selected pages.

66. A method of providing a search result set in response to a query, comprising:

building an index of words on pages selected in a random fashion from the Internet;

ranking each word in the index for use of the word on each of the pages on which the word is used;

adjusting the ranking for each word for each page in accordance with a number of other pages using said each word in association with an outbound link to said each page; and

then

receiving and responding to a query by searching the index for words associated with the query and ordering a search result set in response to the query in accordance with the rankings in the index of the associated words and pages on which the associated words are used.

67. The method of claim 66 wherein ranking each word further comprises:

adjusting the ranking in accordance with a page weight related to a probability of a user selecting the related page when viewing pages on the Internet in a random fashion.

68. The method of claim 66 wherein adjusting the ranking for each word for each page further comprises:

adjusting the ranking for each page with an outbound link with a page weight related to a probability of a user selecting the related page when viewing pages on the Internet in a random fashion.

69. The method of claim 66 wherein the ranking of each word is adjusted in accordance with a proximity to another word on the same page.

70. The method of claim 66 wherein the ranking of each word is adjusted in accordance with a word order to another word on the same page.