A survey on Web Contents Classification System

Currently, the amount of adult (pornographic) content on the Internet is increasing rapidly. This makes an automatic detection of adult content a more challenging task, when eliminating access to ill-suited websites. It is easy for children to access pornographic webpages due to the freely available adult content on the Internet. It creates a problem for parents wishing to protect their children from such unsuitable content.In 2005, the European Parliament launched a large program called “Safer Use of the Internet”, particularly for young people. Some webpages contain a huge amount of combined data related to healthcare (information on diseases, mental health, and physical fitness) and sexual knowledge (medicine for sexual health, birth control, treatment during pregnancy, etc.). In this system, we focus on the recognition of Web adult content. A fuzzy-ontology/SVM–based adult content detection system is proposed to automate the classification of pornographic versus medical websites. The proposed mechanism offers an adult content detection system that classifies webpages into normal, pornographic, or medical webpages using extracted web content features. The adult Web page bag recognition is carried out using multi-instance learning based on the combination of classifying texts, images and videos in Web pages. Additional Key Words and Phrases: Semantic knowledge, fuzzy ontology, SVM (Support Vector Machine), adult content identification, skin patch modeling, recognition of adult images, recognition of adult videos, recognition of adult Web page bags,k-NN (nearest neighbor).