Filtering harmful sentences based on three-word co-occurrence
Proceedings of the 8th Annual Collaboration, Electronic messaging, Anti-Abuse and Spam Conference
Grouping co-occurrence filtering based on bayesian filtering
IEA/AIE'12 Proceedings of the 25th international conference on Industrial Engineering and Other Applications of Applied Intelligent Systems: advanced research in applied artificial intelligence
| Hi-index | 0.00 |
BBSs (Bulletin Board Systems) and Social Network Services (SNS) have been increasing in recenter years. In such systems, users can easily upload and share their own information via personal computers, and also cellular phones. However some information, such as adult content, is not appropriate for all users, notably children. Many SNS and BBS providing companies have been trying to monitor and check harmful information which comes from their users. At the current stage, these companies manually check the users’ sentences before being published. For these companies, even partial automatization of such watching tasks will reduce the huge labor cost. Based on the above motivation and background, in this paper, we focus on filtering harmful text information. In this paper, we have built a program, which can calculate three words co-occurrences, and demonstrate that three-word co-occurrence is more useful than two-word cooccurrence by experimentations.