Evaluating Topic Coherence Using Distributional Semantics
Evaluating Topic Coherence Using Distributional Semantics
复制标题
DOI：
--
复制
发表时间：
2013-03
期刊：
影响因子：
--
通讯作者：
Nikolaos Aletras;Mark Stevenson
中科院分区：
文献类型：
--
作者：
Nikolaos Aletras;Mark Stevenson
本文介绍了分布式语义相似性方法，用于自动测量主题模型生成的一组单词的连贯性。我们利用维基百科作为参考语料库来识别上下文特征并收集频率，构建语义空间来表示每个主题词。主题词和上下文特征之间的相关性是使用逐点互信息（PMI）的变体来测量的。主题一致性是通过测量使用各种度量计算的这些向量之间的距离来确定的。对三个数据集的评估表明，基于分布的度量优于该任务的最先进方法。
This paper introduces distributional semantic similarity methods for automatically measuring the coherence of a set of words generated by a topic model. We construct a semantic space to represent each topic word by making use of Wikipedia as a reference corpus to identify context features and collect frequencies. Relatedness between topic words and context features is measured using variants of Pointwise Mutual Information (PMI). Topic coherence is determined by measuring the distance between these vectors computed using a variety of metrics. Evaluation on three data sets shows that the distributional-based measures outperform the state-of-the-art approach for this task.