摘要
How to extract social relations from the text content in internet is a problem. A supervised method based on machine learning algorithm has been used to solve the problem. Based on the characteristics of social relationship, the appropriate rules have been made for feature extraction. Based on the result of feature extraction, two methods have been proposed which are support vector machine (SVM) and the maximum entropy model for the relation extraction experiment. The results show that support vector machine algorithm is better than the maximum entropy model.
源语言 | 英语 |
---|---|
页(从-至) | 713-718 |
页数 | 6 |
期刊 | Journal of Theoretical and Applied Information Technology |
卷 | 46 |
期 | 2 |
出版状态 | 已出版 - 12月 2012 |