Rahim Dehkharghani - Faculty member Department of Computer Engineering University of Bonab Bonab, Iran

Sentiment analysis is the task of estimating the sentiment polarity which aims to extract the polarity embedded in user entered data (for example, in social media. The data isusually in textual format. Each piece of text requires language preprocessing steps to make it prepared for further processing such as machine translation, text to speech or sentiment analysis. The lack of this preprocessing phase will cause in inaccurate results. In order to achieve higher performance, natural language processing (NLP) techniques are required. These NLP techniques include preprocessing steps such as segmentation and tokenization, and then addressing other issues such as negation, intensification, conditional sentences, and sarcasm detection. This paper comprehensively investigates the NLP issues in sentiment analysis, which has not yet covered sufficiently in the literature. Addressing the above mentioned NLP issues in sentiment analysis significantly increases the effectiveness of polarity extraction task from text, which is comprehensively studied in this work

sentiment analysis, natural language processing, machine learning, polarity classification