Open Access

ARTICLE

RUSAS: Roman Urdu Sentiment Analysis System

Kazim Jawad¹, Muhammad Ahmad², Majdah Alvi³, Muhammad Bux Alvi^3,*

1 Faculty of Computer Science and Engineering, Frankfurt University of Applied Sciences, Frankfurt am Main, 60318, Germany
2 Technical Writer and Researcher, Proteus Technologies LLC, Islamabad, 04405, Pakistan
3 Faculty of Engineering, The Islamia University of Bahawalpur, Bahawalpur, 63100, Pakistan

* Corresponding Author: Muhammad Bux Alvi. Email:

(This article belongs to the Special Issue: Advance Machine Learning for Sentiment Analysis over Various Domains and Applications)

Computers, Materials & Continua 2024, 79(1), 1463-1480. https://doi.org/10.32604/cmc.2024.047466

Received 06 November 2023; Accepted 17 March 2024; Issue published 25 April 2024

Abstract

Sentiment analysis, the meta field of Natural Language Processing (NLP), attempts to analyze and identify the sentiments in the opinionated text data. People share their judgments, reactions, and feedback on the internet using various languages. Urdu is one of them, and it is frequently used worldwide. Urdu-speaking people prefer to communicate on social media in Roman Urdu (RU), an English scripting style with the Urdu language dialect. Researchers have developed versatile lexical resources for features-rich comprehensive languages, but limited linguistic resources are available to facilitate the sentiment classification of Roman Urdu. This effort encompasses extracting subjective expressions in Roman Urdu and determining the implied opinionated text polarity. The primary sources of the dataset are Daraz (an e-commerce platform), Google Maps, and the manual effort. The contributions of this study include a Bilingual Roman Urdu Language Detector (BRULD) and a Roman Urdu Spelling Checker (RUSC). These integrated modules accept the user input, detect the text language, correct the spellings, categorize the sentiments, and return the input sentence’s orientation with a sentiment intensity score. The developed system gains strength with each input experience gradually. The results show that the language detector gives an accuracy of 97.1% on a close domain dataset, with an overall sentiment classification accuracy of 94.3%.

---

Keywords

---