Transfer Learning Based on SVD for Spam Filtering
Jiana Meng, Hongfei Lin, Yu-hai Yu · 2010
At present most email spam filtering methods assume that the training data from a source domain and the test data from a target domain follow the same distribution. However, in many cases this assumption may not be hold. In this paper we propose a transfer learning method based on singular value decomposition (SVD) for solving spam filtering problem. We compute the similarity between target particular features and common features with singular value decomposition method in order to learn a common feature representation. Then we rebuild a vector space model (VSM) of the training and the test data. The final label predictions are decided by a traditional machine learning method. The empirical results on three data sets show that our method is effective.