Research and Design an Enhanced Data Mining Algorithm Based on GA and VSM

摘要

Classification technology is one of research topic in data mining. Its low accuracy and high recall rates are impede its further application. It is especially true in the application of mass data. Integrating genetic algorithm with vector space model (VSM), a multi-topic classification method of vector space model based on GA is proposed. The algorithm withdraws characteristic vectors for a given application and calculates the similarities of classification characteristic vectors. Comparing the value of every classification similarity with dynamic threshold, several different classifications are classified to the multi-topic. Through applying to Web test clarification, the algorithm is proven to achieve high accuracy and low recalling.