This paper describes a Web filtering system "WebGuard", which aims to automatically detect and filter adult content on the Web. WebGuard uses Web crawler to extract relevant data from the Web, combines the textual content, the image content, and the URL name of a Web page to construct its feature vector. WebGuard uses data mining techniques to classify URLs into two classes: suspect URLs and normal URLs. The suspect URLs are stored in a database, which is constantly and automatically updated in order to reflect the highly dynamic evolution of the Web. When working, WebGuard simply captures a user?s URL, matches it with the suspect URLs stored in the database and takes an appropriate action - filtering or blocking - according to the result of the analysis. Our preliminary results show that it can detect and filter adult content effectively.