Study Of Automatic Tagging System

We are now in the golden age of the Internet and big data.Every day,there are hundreds of millions of information has been generated on the Internet.As users,when we faced with dazzled information,how to find the information we are interested or we need is a problem of growing concern.At the same time,a large quantity of sites and information producers want their content to be delivered to the users.In news site,for example,both sides of the news want a efficient delivery method in such an era of information explosion.Tag was born to solve such problems.But facing with such large scale of news,traditional way which is assignment by human is too limited to meet the growing requirement.Therefore,Automatic tagging is drawing more and more attention.In this thesis,we study the problem of tag assignment.We design and implement a prototype system of an automatic tagging function.In the automatic tagging system,we implement two kinds of automatic tagging method,one is based on query expansion,the other is based on collaborative filtering.And the automatic tagging system also incorporates an automatic summarization module and a text classification module.Besides using classification method to do the documents classification,we also use it to improve the efficiency of the original algorithm for the automatic tagging.Finally,we developed a web-page-based system interface,users can interact with the system in their web browser.The automatic tagging system can take input through several kinds of ways.And after the automatic tagging process in the server,the system's web page would display the tagging results on the browser.