سيدعرفان فرهانيان

The rapid evolution of internet-based applications such as websites and social networks has led individuals to generate an large volume of opinions, reviews, and news on various topics, including products, politics, daily activities, and more. Accessing public opinions through such platforms can provide valuable insights. Sentiment analysis involves the use of Natural Language Processing (NLP), text analysis, and computational linguistics to identify and extract subjective information from diverse sources. It can also be considered a process for identifying or classifying sentiment polarity into three main categories: positive, negative, and neutral. This thesis focuses on the feasibility of selecting an appropriate sentiment polarity detection model based on the size, structure, and type of input text (tweets, product reviews, or post comments). The research includes 9 datasets, 8 sentiment polarity analysis models, and 3 decision-making models. First, the datasets are preprocessed and cleaned. Then, a decision-making model selects one of the sentiment polarity analysis models to label the sentiment polarity of the given text. The objective of this thesis is to identify the most suitable sentiment polarity analysis model for Persian texts, achieving optimal performance through implementation and eva‎luation of the proposed solution. In this study, the decision-making model identifies the most appropriate sentiment polarity analysis model based on the input text, and the selected model performs the prediction.

پيش‌بيني قطبيت متن , تحليل احساسات , متن‌كاوي , پردازش زبان طبيعي

text polarity prediction , sentiment analysis , text mining , natural language processing

Seyed Erfan Farhanian

Dr Hasan Naderi