中文摘要：

随着移动互联网的发展，越来越多的读者开始使用移动设备来获取信息。情境作为推荐系统的重要因素不可或缺，但是在基于情境感知的推荐时，各情境因素对读者选择资源所起的重要性各不相同。本文提出一种基于情境感知的移动阅读个性化信息推荐模型，利用情境条件熵计算不同情境属性的权重，将其与传统的协同过滤推荐算法结合。最后，选取科学网博文开展实验研究，结果表明提出的推荐方法能为处于特定情境下的读者提供个性化的移动阅读服务。

英文摘要：

With the development of the mobile internet, more and more readers get access to portable device to obtain information. As an important factor, context is indispensable to recommendation system, but the importance of each contextual factor means different to readers when they single out the resources they are interested in. This paper proposes a context-aware recommendation model for mobile reading. Firstly, the value of contextual conditional entropy is calculated to measure the weight of the different context attributes. Based on it, a context-aware recommendation approach is proposed combined with the traditional collaborative filtering. Finally, the blogs of Science Net are chosen to conduct experiments, and the results show that the proposed approach can provide the personalized mobile reading services for readers in the particular contextual environments.