Sentiment Analysis in Gujarati Language with Dictionary Approach

In recent years, due to the availability of voluminous data on the web for Indian languages, it has become an important task to analyze this data to retrieve useful information. Because of the growth of Indian language content, it is beneficial to utilize this explosion of data. In this research, the deliberate choice of Gujarati language for sentiment analysis stems from a commitment to capture the distinct linguistic and cultural characteristics of the Gujarat region. For the purpose of sentiment analysis, created a Gujarati tweets dataset by crawling on particular hashtags. Once the dataset is prepared, performed all necessary pre-processing steps on our Gujarati dataset of tweets. After that, perform the sentiment analysis using a dictionary approach. In this approach, created dictionaries of positive and negative words manually and also took the help of some research work as well. With the help of dictionaries, prepared the training dataset and testing dataset and apply different classification algorithms on it. Through this survey, we get to know different methods used for sentiment analysis as it is used to determine the sentiment or emotion expressed in a piece of text, whether positive or negative. Also, get to know how to count the sentiment score of Gujarati tweets. Additionally, we explore the significance of our dataset and its potential for fostering future research endeavors in Gujarati language analysis. This comprehensive analysis contributes to a deeper understanding of sentiment dynamics in Gujarati, with implications for diverse applications in natural language processing and cultural studies. There are many potential applications for sentiment analysis such as marketing and customer service, social media monitoring, political analysis, public health, financial analysis, disaster response, etc.