LGGF: Lexicon-Guided Gated Fusion for Bangla Hazard News Classification

Published: 8 September 2026| Version 1 | DOI: 10.17632/6ybfcn6n96.1
Contributors:
,
,

Description

BanglaClimateNews (raw) is a publicly released Bangla-language news collection for hazard classification, spanning six classes: Cold Wave, Cyclone, Earthquake, Flood, Heatwave, and Waterlogging. Articles were scraped from three Bangla news outlets (Prothom Alo, Kaler Kantho, Jugantor) covering January 2019–December 2024, restricted to weather/disaster/national desks matched by a seed keyword. This upload contains the raw, unprocessed scrape — six Excel files, one per class (ColdWave.xlsx, Cyclon.xlsx, Earthquake.xlsx, Flood.xlsx, Heatwave.xlsx, Waterlogging.xlsx), each with two columns: newsheadline and summary (the article's lead paragraph as published on the source page). No deduplication, near-duplicate filtering, cleaning, or train/val/test splitting has been applied — those steps, along with the chi-square hazard lexicon and the LGGF model, are described in the associated paper: "LGGF: Lexicon-Guided Gated Fusion for Bangla Hazard News Classification" (COMPAS 2026). Released for academic research on Bangla hazard/disaster news classification, lexicon mining, and low-resource NLP. Please cite the associated paper if you use this data.

Files

Institutions

Categories

Natural Language Processing

Licence