# Interpretable Multi Labeled Bengali Toxic Comments Classification using Deep Learning

Source: [Interpretable Multi Labeled Bengali Toxic Comments Classification using Deep Learning](<https://www.sota2.com/research/paper/68fa6f291e209131cea44261>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:22:26.632Z

Authors: Tanveer Ahmed Belal, G. M. Shahariar, Md. Hasanul Kabir

Published: 2023-04-08T19:28:26.000Z

## Abstract

This paper presents a deep learning-based pipeline for categorizing Bengali toxic comments, in which at first a binary classification model is used to determine whether a comment is toxic or not, and then a multi-label classifier is employed to determine which toxicity type the comment belongs to. For this purpose, we have prepared a manually labeled dataset consisting of 16,073 instances among which 8,488 are Toxic and any toxic comment may correspond to one or more of the six toxic categories - vulgar, hate, religious, threat, troll, and insult simultaneously. Long Short Term Memory (LSTM) with BERT Embedding achieved 89.42% accuracy for the binary classification task while as a multi-label classifier, a combination of Convolutional Neural Network and Bi-directional Long Short Term Memory (CNN-BiLSTM) with attention mechanism achieved 78.92% accuracy and 0.86 as weighted F1-score. To explain the predictions and interpret the word feature importance during classification by the proposed models, we utilized Local Interpretable Model-Agnostic Explanations (LIME) framework. We have made our dataset public and can be accessed at - https://github.com/deepu099cse/Multi-Labeled-Bengali-Toxic-Comments-Classification

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Toxicity Classification on Jigsaw (test)](<https://www.sota2.com/research/sota/toxicity-classification-on-jigsaw-test.md>): Toxicity Classification; Jigsaw (test); Accuracy: 93.6.

- [Toxicity Classification on Toxicity Dataset (test)](<https://www.sota2.com/research/sota/toxicity-classification-on-toxicity-dataset-test.md>): Toxicity Classification; Toxicity Dataset (test); Test Accuracy: 93.64.

- [Toxic Comment Classification on Jigsaw Toxic Comment Classification](<https://www.sota2.com/research/sota/toxic-comment-classification-on-jigsaw-toxic-comment-classification.md>): Toxic Comment Classification; Jigsaw Toxic Comment Classification; Accuracy: 93.64.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2304.04087>)

[PDF](<https://arxiv.org/pdf/2304.04087.pdf>)
