ClaC: Semantic Relatedness of Words and Phrases

TL;DR

This paper introduces three methods for measuring the semantic relatedness of words and phrases, combining semantic networks, distributional models, and a hybrid approach, with the hybrid achieving notable accuracy.

Contribution

It presents a novel hybrid approach that combines semantic network and distributional similarity models for improved phrasal semantic relatedness measurement.

Findings

Hybrid approach achieved 77.4% F-measure

Semantic network and distributional models compared

Hybrid outperforms individual models

Abstract

The measurement of phrasal semantic relatedness is an important metric for many natural language processing applications. In this paper, we present three approaches for measuring phrasal semantics, one based on a semantic network model, another on a distributional similarity model, and a hybrid between the two. Our hybrid approach achieved an F-measure of 77.4% on the task of evaluating the semantic similarity of words and compositional phrases.