• Durga Prasad Manukonda
• Rohith Gowtham Kodali
• Dheeraj Guduri

This study aimed to investigate linguistics-based approach on sentiment analysis in Telugu language. Sentiment analysis is the process of classifying sentences as positive, negative, or neutral. One of the uses of this technology is to find the perceptions towards organizations and their services through their feedback in text format. With rapid growth of social media discussions on social networks, lots of data in the Telugu language is available. The new sentiment analyzer models are based on previous models made for the English language. These translate sentences with help of translation APIs into English and analyze the sentiments. One can ob- serve compounding errors by faulty machine translation. Modern translators are inefficient are unreliable. This model is specially designed for Telugu language with newly identified rules. This model has been successfully tested for the Telugu language. The corpus for this study was gathered by using BootCa we have manually tagged root words with their polarity. We have identified and generated some important rules for identifying the sentiment in the sentence. These rules have been explained and discussed at length along with examples.

NLP, SentiWordNet, Sentiment Analysis, BootCaT, Wikipedia database

"Phrase-Based Heuristic Sentiment Analyzer for the Telugu Language", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 3, page no.245-251, March-2019, Available :http://www.jetir.org/papers/JETIRAR06055.pdf

"Phrase-Based Heuristic Sentiment Analyzer for the Telugu Language", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 3, page no. pp245-251, March-2019, Available at : http://www.jetir.org/papers/JETIRAR06055.pdf