Keyword index pruning

A method for using a web-based system for personalized search using keyword index pruning, for content acquisition, further comprising the steps of:

• an initiating keyword phrase is submitted by the search user to an external search engine, which then returns search results for review by the search user;
• the user then implicitly or explicitly scores the relevancy of the results;
• the system profiles the search user, keywords, and results based on the search interests and the associated responses of keywords and results within a search session;
• after eliminating the keyword tokens that do not meet the effectiveness criteria as described above, the profiles are then used to generate new keywords by matching the search user subject profile to the keyword profiles and then using the top-matching keyword tokens to generate new trial keyword phrases for subsequent search;
• this process is iterated and refined as the search user provides feedback on the subsequent search results.

A method for making a computer implemented web-based system for personalized search using keyword index pruning, comprising the steps of:

• providing a computer implemented web-based service system for searching, collecting relevancy feedback, and acquiring content by keywords;
• storing subject and object, including keywords and search results, in a database;
• matching and ranking said subject and keyword profiles to target content profiles to be optimal to keywords;
• deriving subject relevancy ratings of objects wherein optimal subject and object vectors are derived from said subject's relevancy ratings of objects;
• processing and identifying ineffective keywords;
• and
• eliminating said ineffective keywords.

A web-based method for personalized search, comprising the steps of:

• providing a service system for searching, collecting relevancy feedback, and acquiring content;
• wherein said provided system automatically identifies and eliminates ineffective keywords:
• providing subjects and objects represented in vector form;
• calculating optimal subject and object vectors derived from said subject's relevancy ratings of objects;
• and
• providing search results to, and collecting relevancy ratings from, an external application;
• whereby said system does not need to know anything about object content, does not need to know anything about subject demographics, and uses a fixed size profile for subjects and objects for scalable processing,
• all accomplished in a mentor-less and self-optimizing fashion, scalable to large numbers of users, keywords, and objects.

A method for using a system for personalized search, comprising the steps of:

• submitting keywords;
• automatically identifying and eliminating ineffective keywords;
• presenting search results;
• selecting a search result;
• rating the relevancy of the search result;
• updating user, keyword, and search result profiles;
• re-matching keyword and content profiles and re-displaying search results;
• and
• clustering and synthesizing keywords;
• whereby said system generates targeted search results to each user and self-optimizes keyword profiles to optimize the relevancy of the search results for each user.