Punjabi Text Sentiment Analysis using Conjunction, Disjunction, and Negation Words

In modern times, the digital world has expanded in such a way that people can express opinions and thoughts with great ease. Earlier, English was the sole language to communicate over the internet. However, in today’s scenario, users on the internet have more personal choices to share content in vernacular languages. This leads to an increase in the usage of Punjabi content on the web for blogs, sharing opinions, providing feedback, and recommendations. Hence, Sentiment Analysis in Punjabi text is one of the growing study areas that aims to understand and clarify emotions and intentions into positive, negative, or neutral polarity. This paper comprehensively examines the evolving techniques, detailing foundational processes like data gathering and pre-processing. Also, a dataset in the Punjabi language was compiled and pre-processed, followed by the development of a lexicon for conjunctive, disjunctive, and negation words. The paper also elaborates on challenges connected with sentiment analysis in Punjabi, such as a lack of digital resources like wordnets and labelled corpora, morphological variations, and the problem of code-mixing with English.