Adopting Transition Point Technique for Persian Sentiment Analysis

Citation

Dashtipour K, Gogate M, Gelbukh A & Hussain A (2021) Adopting Transition Point Technique for Persian Sentiment Analysis. In: TBC. ICOTEN 2021: International Congress of Advanced Technology and Engineering, Virtual, 04.07.2021-05.07.2021. Piscataway, NJ, USA: IEEE.

Abstract
Sentiment analysis is used to analyses people’s opinions, views and emotions towards different entities such as products, organizations, companies and events. People’s opinions are important for most others during their decision-making process. For example, if someone wants to buy a product, they might want to know more about that product and the experiences of others with that product. Sentiment analysis is able to classify the reviews based on their polarity; even if reviews are expressed in a sentence or document, sentiment analysis is used to classify it into positive, negative or neutral reviews. In this paper, we proposed a framework using TF-IDF and transition point to detect polarity in Persian movie reviews. The proposed approach has been evaluated using different classifiers such as SVM, Naive Bayes, MLP and CNN. The experimental results show the transition point is more effective in comparison with traditional feature such as TF-IDF.

Keywords
Sentiment Analysis; Persian; Machine Learning

Notes
Output Status: Forthcoming

• Accepted manuscript