Distributional Measures as Proxies for Semantic Relatedness

TL;DR

This paper critically evaluates distributional measures of semantic relatedness, identifies their limitations, proposes improved measures aligned with human judgments, and compares them with ontology-based approaches to advance understanding in the field.

Contribution

It provides a detailed analysis of major distributional measures, introduces new measures addressing their drawbacks, and offers an extensive comparison with ontology-based methods.

Findings

Distributional measures have limitations in capturing human-like semantic relatedness.

New measures proposed show improved alignment with human judgments.

Comparison reveals strengths and weaknesses of distributional versus ontology-based approaches.

Abstract

The automatic ranking of word pairs as per their semantic relatedness and ability to mimic human notions of semantic relatedness has widespread applications. Measures that rely on raw data (distributional measures) and those that use knowledge-rich ontologies both exist. Although extensive studies have been performed to compare ontological measures with human judgment, the distributional measures have primarily been evaluated by indirect means. This paper is a detailed study of some of the major distributional measures; it lists their respective merits and limitations. New measures that overcome these drawbacks, that are more in line with the human notions of semantic relatedness, are suggested. The paper concludes with an exhaustive comparison of the distributional and ontology-based measures. Along the way, significant research problems are identified. Work on these problems may lead to a better understanding of how semantic relatedness is to be measured.