摘要
介绍了M arkov逻辑网的理论模型、学习算法和推理算法,并将其应用于中文文本分类中。实验结合了判别式训练的学习算法,MC-SAT、吉布斯抽样和模拟退火等推理算法,结果表明基于M arkov逻辑网的分类方法能够取得比传统K邻近(KNN)分类算法更好的效果。
This paper introduced the theory, learning methods and inference algorithms of Markov logic network that was also applied to the Chinese text classification. With reference to the discriminative learning algorithm for Markov logic network weights, MC-SAT, Gibbs sampling and simulated tempering algorithm in experiments, it proves that the method based on Markov logic network is better than conventional K Nearset Neighbor (KNN) method in text classification.
出处
《计算机应用》
CSCD
北大核心
2009年第10期2729-2732,共4页
journal of Computer Applications
基金
重庆市自然科学基金资助项目(CSTC2008BB2021)
关键词
统计关系学习
机器学习
MARKOV逻辑网
文本分类
Statistical Relational Learning (SRL)
machine learning
Markov logic network
text classification