MCS

Permanent URI for this community

Now showing 1 - 1 of 1

Sentiment Analysis of Resource-Poor Language (Roman Urdu) on Novel Dataset
(UMT,LAHORE, 2023) Abdul MoinArshad S2021027001 M. Waseem S2021027002 Anusha Naeem S2021027009
The internet has made it convenient to disseminate information about products, services, events, and political ideologies. However, most sentiment analysis research has predominantly focused on the English language, neglecting other languages such as Roman Urdu. Analyzing sentiment in Roman Urdu poses various challenges, primarily due to the absence of dedicated lexical re sources and the resulting information amalgamation. This study aims to address these challenges by developing a comprehensive dataset for Roman Urdu sentiment analysis and evaluating ma chine learning models for this purpose. The research explores commonly utilized methods for analyzing sentiments in Roman Urdu and Urdu. The outcomes of this study will contribute to the enhancement of Roman Urdu and sentiment analysis tools. A specifically tailored dataset for Roman Urdu is created, and the utilization of machine learning techniques significantly improves accuracy and performance. The proposed approach achieves 92% accuracy on the test data using machine learning