Unsupervised Technical Domain Terms Extraction using Term Extractor

TL;DR

This paper presents an unsupervised method for extracting domain-specific terms from text corpora, utilizing chunking, preprocessing, and ranking based on relevance and cohesion, aimed at improving automatic terminology extraction.

Contribution

It introduces a novel unsupervised approach that combines chunking, preprocessing, and ranking functions for domain term extraction, tailored for the TermTraction shared task.

Findings

Effective in extracting relevant domain terms

Outperforms baseline methods in shared task

Demonstrates robustness across different domains

Abstract

Terminology extraction, also known as term extraction, is a subtask of information extraction. The goal of terminology extraction is to extract relevant words or phrases from a given corpus automatically. This paper focuses on the unsupervised automated domain term extraction method that considers chunking, preprocessing, and ranking domain-specific terms using relevance and cohesion functions for ICON 2020 shared task 2: TermTraction.