GM-CTSC at SemEval-2020 Task 1: Gaussian Mixtures Cross Temporal Similarity Clustering

TL;DR

This paper presents an unsupervised approach using Gaussian Mixture Models on temporal word embeddings to detect lexical semantic change, achieving improved results over threshold-based methods in the SemEval-2020 task.

Contribution

The paper introduces a novel Gaussian Mixture Model-based algorithm for detecting semantic change in words over time using temporal embeddings.

Findings

Gaussian Mixture Models improved detection accuracy.

Combination of GMM with Temporal Referencing yielded best results.

Performance varied across different embedding algorithms.

Abstract

This paper describes the system proposed for the SemEval-2020 Task 1: Unsupervised Lexical Semantic Change Detection. We focused our approach on the detection problem. Given the semantics of words captured by temporal word embeddings in different time periods, we investigate the use of unsupervised methods to detect when the target word has gained or loosed senses. To this end, we defined a new algorithm based on Gaussian Mixture Models to cluster the target similarities computed over the two periods. We compared the proposed approach with a number of similarity-based thresholds. We found that, although the performance of the detection methods varies across the word embedding algorithms, the combination of Gaussian Mixture with Temporal Referencing resulted in our best system.