English

Transfer Learning for Low-Resource Sentiment Analysis

Computation and Language 2023-04-11 v1

Abstract

Sentiment analysis is the process of identifying and extracting subjective information from text. Despite the advances to employ cross-lingual approaches in an automatic way, the implementation and evaluation of sentiment analysis systems require language-specific data to consider various sociocultural and linguistic peculiarities. In this paper, the collection and annotation of a dataset are described for sentiment analysis of Central Kurdish. We explore a few classical machine learning and neural network-based techniques for this task. Additionally, we employ an approach in transfer learning to leverage pretrained models for data augmentation. We demonstrate that data augmentation achieves a high F1_1 score and accuracy despite the difficulty of the task.

Keywords

Cite

@article{arxiv.2304.04703,
  title  = {Transfer Learning for Low-Resource Sentiment Analysis},
  author = {Razhan Hameed and Sina Ahmadi and Fatemeh Daneshfar},
  journal= {arXiv preprint arXiv:2304.04703},
  year   = {2023}
}

Comments

14 pages - under review at ACM TALLIP

R2 v1 2026-06-28T09:57:46.375Z