1 Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China|2 Hewlett Packard Labs China, Beijing 100084, China; 
3 Graduate University of Chinese Academy of Sciences, Beijing 100190, China.

Recent years have witnessed an increasing interest in transfer learning. This paper deals with the classification problem that the target-domain with a different distribution from the source-domain is totally unlabeled, and aims to build an inductive model for unseen data. Firstly, we analyze the problem of class ratio drift in the previous work of transductive transfer learning, and propose to use a nor-malization method to move towards the desired class ratio. Furthermore, we develop a hybrid regu-larization framework for inductive transfer learning. It considers three factors, including the distribution geometry of the target-domain by manifold regularization, the entropy value of prediction probability by entropy regularization, and the class prior by expectation regularization. This framework is used to adapt the inductive model learnt from the source-domain to the target-domain. Finally, the experiments on the real-world text data show the effectiveness of our inductive method of transfer learning. Mean-while, it can handle unseen test points.

Supported by the National Science Foundation of China (Grant Nos. 60435010, 60675010), National High Technology Research and Development of China (Grant Nos. 2006AA01Z128, 2007AA01Z132), National Basic Research Priorities Pro-gramme (Grant No. 2007CB311004) and National Science and Technology Support Plan (Grant No. 2006BAC08B06)