Similarity-Based Models of Word Cooccurrence Probabilities

TL;DR

This paper introduces a similarity-based approach to estimate probabilities of unseen word combinations in NLP, improving language modeling and disambiguation accuracy by leveraging distributional word similarities.

Contribution

It presents novel probabilistic models that use word similarity to better estimate probabilities of unseen word pairs, outperforming traditional methods.

Findings

20% perplexity reduction in language modeling

Significant speech recognition error reduction

Up to 40% improvement in disambiguation task

Abstract

In many applications of natural language processing (NLP) it is necessary to determine the likelihood of a given word combination. For example, a speech recognizer may need to determine which of the two word combinations ``eat a peach'' and ``eat a beach'' is more likely. Statistical NLP methods determine the likelihood of a word combination from its frequency in a training corpus. However, the nature of language is such that many word combinations are infrequent and do not occur in any given corpus. In this work we propose a method for estimating the probability of such previously unseen word combinations using available information on ``most similar'' words. We describe probabilistic word association models based on distributional word similarity, and apply them to two tasks, language modeling and pseudo-word disambiguation. In the language modeling task, a similarity-based model is used to improve probability estimates for unseen bigrams in a back-off language model. The similarity-based method yields a 20% perplexity improvement in the prediction of unseen bigrams and statistically significant reductions in speech-recognition error. We also compare four similarity-based estimation methods against back-off and maximum-likelihood estimation methods on a pseudo-word sense disambiguation task in which we controlled for both unigram and bigram frequency to avoid giving too much weight to easy-to-disambiguate high-frequency configurations. The similarity-based methods perform up to 40% better on this particular task.