SentMix-3L: A Bangla-English-Hindi Code-Mixed Dataset for Sentiment Analysis

Md Nishat Raihan, Dhiman Goswami, Antara Mahmud, Antonios Anastasopoulos, Marcos Zampieri
10/27/2023
cs.CL

No official code found. Best implementation: LanguageTechnologyLab/SentMix-3L (4 stars, AGPL-3.0, last commit 2023-11-08).

Abstract

Code-mixing is a well-studied linguistic phenomenon when two or more languages are mixed in text or speech. Several datasets have been build with the goal of training computational models for code-mixing. Although it is very common to observe code-mixing with multiple languages, most datasets available contain code-mixed between only two languages. In this paper, we introduce SentMix-3L, a novel dataset for sentiment analysis containing code-mixed data between three languages Bangla, English, and Hindi. We carry out a comprehensive evaluation using SentMix-3L. We show that zero-shot prompting with GPT-3.5 outperforms all transformer-based models on SentMix-3L.

View on arXivView PDF

Code Implementations(1)

LanguageTechnologyLab/SentMix-3LCommunity100%
AGPL-3.0

This is a dataset for the sentiment analysis task. It contains 1k natural code mixed data. The languages are Bangla-English-Hindi.

41Oct 17, 20232 years agoAGPL-3.0

Similar Papers

OffMix-3L: A Novel Code-Mixed Dataset in Bangla-English-Hindi for Offensive Language Identification

91% match1 repo32023

Comparative Study of Pre-Trained BERT Models for Code-Mixed Hindi-English Data

88% match2023

Leveraging Large Language Models for Code-Mixed Data Augmentation in Sentiment Analysis

88% match1 repo12024

Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text in Multilingual Encoders

86% match2026

Marathi-English Code-mixed Text Generation

86% match2023

Zero- and Few-Shot Prompting with LLMs: A Comparative Study with Fine-tuned Models for Bangla Sentiment Analysis

86% match2023

Cite this paper

@article{raihan2023sentmixl,
  title  = {SentMix-3L: A Bangla-English-Hindi Code-Mixed Dataset for Sentiment Analysis},
  author = {Md Nishat Raihan and Dhiman Goswami and Antara Mahmud and Antonios Anastasopoulos and Marcos Zampieri},
  year   = {2023},
  eprint = {2310.18023},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2310.18023v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2310.18023.svg)](https://codeofpaper.com/paper/2310.18023)

Discussion