Clustering Prominent People and Organizations in Topic-Specific Text Corpora

TL;DR

This paper presents a novel method for clustering prominent people and organizations in topic-specific texts using named entity recognition and word embeddings, demonstrating effective semantic clustering through human and quantitative evaluation.

Contribution

The paper introduces a new clustering approach that combines named entity recognition with word embeddings to improve semantic grouping of entities in text corpora.

Findings

The method effectively clusters semantically similar entities.

Human judges rated the clustering quality positively.

Quantitative metrics confirmed the method's effectiveness.

Abstract

Named entities in text documents are the names of people, organization, location or other types of objects in the documents that exist in the real world. A persisting research challenge is to use computational techniques to identify such entities in text documents. Once identified, several text mining tools and algorithms can be utilized to leverage these discovered named entities and improve NLP applications. In this paper, a method that clusters prominent names of people and organizations based on their semantic similarity in a text corpus is proposed. The method relies on common named entity recognition techniques and on recent word embeddings models. The semantic similarity scores generated using the word embeddings models for the named entities are used to cluster similar entities of the people and organizations types. Two human judges evaluated ten variations of the method after it was run on a corpus that consists of 4,821 articles on a specific topic. The performance of the method was measured using three quantitative measures. The results of these three metrics demonstrate that the method is effective in clustering semantically similar named entities.