SOCIAL DATA COLLECTING SYSTEM USING MAHOUT MACHINE

摘要

Provided is a mahout-based social data collecting system capable of filtering advertisement-oriented social information by simply changing or distorting keywords, thereby improving the accuracy through machines repeated studying. Also, the system constantly analyzes and learns the filtered advertisement-oriented social information to collect social information where advertisement oriented social information is filtered. To achieve the purpose, the present invention collects social information through a web crawler and a mahout module determining advertisement features in advertisement-oriented information by accumulating advertisement oriented social information as learning data. Additionally, the present invention includes an information collecting module having a filtering module for performing filtering by applying the advertisement-oriented information to social information. Finally, the mahout module can update the filtering module of the information collecting module by analyzing advertisement features of social information chosen by a manager among social information filtered in the information collecting module.