Sentiment Analysis of Reviews in Natural Language: Roman Urdu as a Case Study

Abstract

Opinion Mining from user reviews is an emerging field. Sentiment Analysis of Natural Language text helps us in finding the opinion of the customers. These reviews can be in any language e.g. English, Chinese, Arabic, Japanese, Urdu, and Hindi. This research presents a model to classify the polarity of the review(s) in Roman Urdu text (reviews). For the purpose, raw data was scraped from the reviews of 20 songs from Indo-Pak Music Industry. In this research a new dataset of 24000 reviews of Roman Urdu text is created. Nine Machine Learning algorithms-Naïve Bayes, Support Vector Machine, Logistic Regression, K-Nearest Neighbors, Artificial Neural Networks, Convolutional Neural Network, Recurrent Neural Networks, ID3 and Gradient Boost Tree, are attempted. Logistic Regression outperformed the rest, based on testing and cross validation accuracies that are 92.25% and 91.47% respectively.

RETRACTED ARTICLE: An ensemble deep learning classifier for sentiment analysis on code-mix Hindi–English data

“…Urdu is one of the popular languages spoken in the Indian subcontinent; Qureshi et al ( 2022 ) have explored Urdu reviews written in Roman Script. They have explored YouTube comments in Romanised Urdu on various Machine Learning models.…”

Roman Urdu Sentiment Analysis of Reviews on PSL Anthems

“…Sentiment Analysis is a way to categorize people's opinions as positive, negative and neutral according to their subjectivity [1]. A large number of research on sentiment analysis is already done in English, Arabic, Chinese, Japanese, Dutch, Farsi and many more [16]. Limited research is witnessed on resource-poor languages e.g.…”

“…To evaluate the model different evaluation metrics i.e., Accuracy, Recall, Precision and F-score [16], are calculated. Rest of the details are given in results and discussion section.…”

Aspect Level Songs Rating Based Upon Reviews in English

“…People are using the internet for business and social correspondence [3]. In past, there were some specialized companies to collect reviews and feedback, for decision making, regarding the product in hand through the market survey but it is an old-fashioned way to collect feedback [4]. People are becoming habitual in buying and selling products online.…”