Features of word similarity

TL;DR

This paper compares various computational models of word similarity and association, evaluating their ability to predict human rating data and highlighting the complexity of factors involved beyond semantic relatedness.

Contribution

It systematically assesses 28 models combining surface and semantic features, providing evidence that word similarity involves more than semantic relatedness alone.

Findings

Models show limited cross-validated performance

Word similarity ratings are influenced by multiple factors

Development of psychological process models is needed

Abstract

In this theoretical note we compare different types of computational models of word similarity and association in their ability to predict a set of about 900 rating data. Using regression and predictive modeling tools (neural net, decision tree) the performance of a total of 28 models using different combinations of both surface and semantic word features is evaluated. The results present evidence for the hypothesis that word similarity ratings are based on more than only semantic relatedness. The limited cross-validated performance of the models asks for the development of psychological process models of the word similarity rating task.