BLM-17m: A Large-Scale Dataset for Black Lives Matter Topic Detection on Twitter

TL;DR

This paper introduces BLM-17m, a large-scale dataset of 17 million tweets related to the Black Lives Matter movement, aimed at facilitating topic detection and analysis of social media discourse during the George Floyd incident.

Contribution

The paper provides a new extensive dataset for Black Lives Matter topic detection on Twitter, along with baseline evaluations using TF-IDF and LDA methods.

Findings

TF-IDF and LDA baselines established for the dataset

Evaluation metrics applied with different k values

Dataset publicly available for research use

Abstract

Protection of human rights is one of the most important problems of our world. In this paper, our aim is to provide a dataset which covers one of the most significant human rights contradiction in recent months affected the whole world, George Floyd incident. We propose a labeled dataset for topic detection that contains 17 million tweets. These Tweets are collected from 25 May 2020 to 21 August 2020 that covers 89 days from start of this incident. We labeled the dataset by monitoring most trending news topics from global and local newspapers. Apart from that, we present two baselines, TF-IDF and LDA. We evaluated the results of these two methods with three different k values for metrics of precision, recall and f1-score. The collected dataset is available at https://github.com/MeysamAsgariC/BLMT.