Spam Filtering based on Knowledge Transfer Learning

Xing Wang, Bin-Xing Fang, Hui He, Hong-Li Zhang
2015 International Journal of Security and Its Applications  
Spam is a serious problem not only the number of floods but also more and more volatile type. It has caused a great impact on people's daily lives. Especially fraud spam, even cause huge losses to companies or individuals possibility. Therefore, it is imminent to filter spam efficiently. Existing spam filtering mechanism is mainly based on the character and content of the spam message. However, once the spam filter uses in other user's mailbox, the existing spam filtering techniques can not be
more » ... ell adapted. In this paper, we propose the adaptive spam filtering method for the above shortcomings. The method uses the unlabeled spam data that from other user or domain to enhance the adaptive and opposability of the anti-spam system. We use the transfer learning model to build the spam filtering system. A transfer learning model can use the untagged data, and migrate knowledge between different filter model, and improve the active collaboration of the filter.
doi:10.14257/ijsia.2015.9.10.31 fatcat:woxtawexdfddnpo5necjptsbdq