一种基于用户偏好的学习推荐方法

传统的推荐算法一定程度上为学习者提供了自适应的学习服务,但忽略了用户的学习兴趣偏好,难以提供学习者满意的推荐服务.为了提高学习推荐的效率,对用户的偏好进行进算,根据兴趣偏好对基本用户进行聚类,然后根据用户之间的兴趣相似性初步预测目标用户的兴趣度,进而给用户推荐兴趣度较高的学习服务.实验结果表明,该方法可显著地提高推荐质量.

Learning Recommendation Method Based on User Preference

To some extent, the traditional recommendation algorithm can provide adaptive learning services for the users, but it couldn＇t provide satisfactory recommended service for Web consumers timely,and the user interest had been ignored. To provide satisfactory recommended service, we should understand and cal- culate the user interest degree and the basal users are clustered. At the same time, user interests matrix is established. The user＇s interestingness is predicted by the interestingness of neighbors whose preferences are similar to the target user,and then the learning pages with higher interestingness to Web users are rec- ommendedThe experimental results show that this method can provid better recommendation results.