A Comparison of Semantic Similarity Methods for Maximum Human Interpretability

TL;DR

This paper compares three semantic similarity methods incorporating semantic information for improved interpretability and accuracy, finding that tf-idf cosine similarity performs best for short news texts.

Contribution

It introduces and evaluates three semantic similarity methods that integrate semantic information, enhancing interpretability and accuracy over word-only approaches.

Findings

Tf-idf cosine similarity outperforms other methods for short texts.

Semantic methods provide more interpretable similarity results.

Word embedding-based methods offer alternative semantic similarity measures.

Abstract

The inclusion of semantic information in any similarity measures improves the efficiency of the similarity measure and provides human interpretable results for further analysis. The similarity calculation method that focuses on features related to the text's words only, will give less accurate results. This paper presents three different methods that not only focus on the text's words but also incorporates semantic information of texts in their feature vector and computes semantic similarities. These methods are based on corpus-based and knowledge-based methods, which are: cosine similarity using tf-idf vectors, cosine similarity using word embedding and soft cosine similarity using word embedding. Among these three, cosine similarity using tf-idf vectors performed best in finding similarities between short news texts. The similar texts given by the method are easy to interpret and can be used directly in other information retrieval applications.