Method for recommending best information in real time by appropriately obtaining gist of web page and user's preference

Abstract

Disclosed is an information recommendation method for providing a construction method for a classified word database capable of rapidly accommodating changes in associations between words. The disclosed information recommendation method basically is based on the finding that, by analyzing occurrence frequency information of an arbitrary word in a Web site having an arbitrary classified word in real-time and obtaining the real-time degree of similarity between the classified word and the arbitrary word, it is possible to construct a database that is capable of being sensitive in responding to changes in associations between words. The disclosed information recommendation method relates to a method of extracting classified words indicating a main idea of text included in a Web page, accumulating the classified words, and constructing a real-time tag database; an information recommendation method for using the tag database to reflect the main idea of the Web page in real-time; and an information matching method of advertisements for using the tag database to reflect the main idea of the Web page and tastes of users in real-time.