Context (archaeology)

Distributional semantics

Pointwise mutual information

Biology

Statistics

Similarity (geometry)

Variety (cybernetics)

Coherence (philosophical gambling strategy)

Semantic similarity

Set (abstract data type)

Computer science

Semantics (computer science)

Paleontology

Natural language processing

Mathematics

Image (mathematics)

Artificial intelligence

Geometry

Information retrieval

Programming language

Mutual information

Word (group theory)

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.

Evaluating Topic Coherence Using Distributional Semantics