Research of information recommendation system based on reading behavior

In modern knowledge explosion society, Humans’ ability to find the information they desire grows more slowly than the rate at which the new information becomes available. Recommendation system can reduce information overload and provide appropriate content to targeted user. This paper proposes a new recommending technique based on mining the user’s reading behavior, computing the user profile through 5W1H scheme. By comparing the user profile and the information representation, we can provide the customized content in suitable time to satisfy the user’s personalized requirement. For performance evaluation, we implement the technology to make experiments on real information and people. The results show that our technology and approach can acquire the user’s requirement, and recommend suitable information they really need.