Visual method and apparatus for enhancing search result navigation

• US 7,502,786 B2
• Filed: 01/04/2007
• Issued: 03/10/2009
• Est. Priority Date: 01/12/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Liu, Shixia, Su, Zhong
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Kim; Paul
• Application Number:
  US11/619,665
• Publication Number:
  US 20070162443A1
• Time in Patent Office: 796 Days
• Field of Search: 707/5, 707/4, 715/200
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A visual method for enhancing search result navigation, comprising: obtaining a first search result from a search engine; clustering the first search result to get clustering information; calculating the correlations between the clustering information and the ranked list of the first search result, and performing visualization processing on the clustering information; and displaying the visual cluster hierarchy and the ranked list of the first search result in a joint manner based on the correlations. When a cluster is selected, further searching is performed and the search result is clustered again. Using the present invention, through combining a traditional ranked list of search results and the visual cluster hierarchy of these search results to display them in a joint manner, a convenient way is provided for the web user to find the potential correlations between the visual cluster hierarchy and the ranked list of the search results. Besides, through dynamically searching and clustering more search results again, the web user may get more detailed and more accurate search results easily.

Claims

1. A visual method for enhancing search result navigation, comprising:

obtaining a first search result from a search engine;

clustering the first search result to get clustering information;

calculating the correlations between the clustering information and a ranked list of the first search result, and performing visualization processing on the clustering information; and

displaying the visual cluster hierarchy and the ranked list of the first search result in a joint manner based on the correlations,

wherein the first search result contains a predetermined number of search result entries in the search results produced by the search engine based on a query;

wherein the displaying visual cluster hierarchy and the ranked list of the first search result in a joint manner comprises one of the following;

when the pages of the ranked list of the first search result are displayed, the cluster in the visual cluster hierarchy that contains the most search result entries of the first search result is highlighted;

when the pages of the ranked list of the first search result are displayed, the cluster in the visual cluster hierarchy that contains the most search result entries of the first search result in the first page is highlighted;

when a search result entry in the ranked list of the first search result is selected, the cluster in the visual cluster hierarchy that contains the search result entry and the most search result entries of the first search result is highlighted;

when a search result entry in the ranked list of the first search result is selected, the cluster in the visual cluster hierarchy that contains the search result entry and the most search result entries of the first search result in the first page is highlighted; and

when a cluster in the visual cluster hierarchy is selected, the ranked list of the search result entries in the first search result that are contained in the cluster is displayed,

wherein the step of clustering the first search result applies the Suffix Tree Clustering algorithm;

wherein said method further comprises repeatable steps comprising;

selecting a cluster in the visual cluster hierarchy;

generating new query keywords and submitting them to the search engine;

producing a new search result based on the new query keywords by the search engine;

selecting a predetermined number of search result entries in the new search result to produce a second search result;

clustering the second search result to obtain sub-clustering information;

calculating the correlations between the sub-clustering information and the ranked list of the second search result, and performing visualization processing on the sub-clustering information;

displaying the visual sub-clustering information and the ranked list of the second search result in a joint manner based on the correlations; and

when a visual sub-clustering information item in the visual sub-clustering information is selected, repeating the repeatable steps,

wherein the visual cluster hierarchy and the visual sub-clustering information form a tree structure, wherein the clusters contained in the visual cluster hierarchy are taken as root nodes and the visual sub-clustering information items contained in the visual sub-clustering information are taken as branch nodes,

wherein the step of generating new query keywords comprises;

combining the current query keywords with the name of the selected cluster to generate new query keywords,

wherein the step of generating new query keywords comprises;

collecting relevant documents;

determining keywords in the relevant documents; and

combining the keywords with the current query to produce a new query,

wherein the relevant documents are the documents that have been read by the web user or the documents that belong to the selected cluster, and

wherein the step of determining keywords in the relevant documents comprises;

calculating weights of all words except stopwords in each document of the relevant documents with the following formula;

value_i=tf·

idf,

where value represents the weight of a word;

tf represents the frequency at which the word appears in the relevant documents;

idf=all_documents/keyword_documents,

where all documents represents the number of all the relevant documents, keyword documents represents the number of the relevant documents that contain this word; and

determining the words with high weights as keywords.