Abstract
The field of sentiment analysis has experienced significant growth in recent years due to its applications in various domains such as news headlines, online product purchase, marketing, and reputation management. With the rise of social media and online shopping platforms, there is a wealth use-generated data available. This has led manufacturing, sales, and marketing companies to seek global feedback on their practices and products from these sources. In the context of Sindhi language, millions of phrases are shared daily on news media sites, Twitter, Facebook, and other platforms. However, the exclusion of sentiment analysis for Sindhi language limits the utilization of this vast amount of data, focusing primarily on the resource-rich English language. This systematic review aims to collect and evaluate published research related to Sindhi language sentiment analysis, specifically focusing on pre-processing, feature extraction, classification methods. The study offers a comprehensive analysis of research conducted on Sindhi text for product evaluation, covering key areas, such as relevant corpora acquisition, data preprocessing, feature extraction, classification techniques, methodologies, limitations, and future directions. Each reviewed article is assessed and classified based on specified criteria. The findings of this review provide valuable insights and propose several approaches for future investigations in this area.
Keywords
Natural language processing, Sentiment analysis, Sindhi corpus, Sindhi text, Systematic review, Text pre-processing
Subject Area
Computer Science
Article Type
Article
First Page
1676
Last Page
1691
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite this Article
Soomro, Safdar Ali; Yuhaniz, Siti Sophiayati; Dootio, Mazhar Ali; Murtaza, Ghulam; and Mughal, Muhammad Hussain
(2025)
"A Systematic Review on Sentiment Analysis for Sindhi Text,"
Baghdad Science Journal: Vol. 22:
Iss.
5, Article 26.
DOI: https://doi.org/10.21123/bsj.2024.10954