Sentiment Analysis of Urdu Language Using Machine Learning Models

Abstract

The purpose of the study was to develop sentiment analysis tool and system for Urdu language through the use of Machine Learning models and deep learning algorithms. In modern age of social media communication and interaction, the anlysis of opinion of the users and customer’s of business products, political discourses, religious and cultural crtiques and debates to explore their sentiments abouth the services, products, scenarios and stories are much needed. The methodology used by the researchers was to create dataset from tweets and posts on social media platforms in Urdu language. The researchers then used pre-processing and Lemmatization data to make it suitable for trainig and testing when using machine learning algorithms such as Support Vector Machine, Naïve Bayes, BiLSTM and other realted models.The sentiment analyzer categorized the sentiments in Urdu language text with nine categroies and thus the resulted tool or system was tested for its effectivness and efficeincy with more than 50% accuracy and high level of validaity were found. Thus, the sentiment analysis of Urdu language using machine learning models was deceloped to bbridge the gap in sentiment analysis for Urdu language in Pakistan.