Sports-fanaticism formalism for sentiment analysis in Arabic text

Social media sites become a common means for sports fans to communicate. Understanding how social text contributes to increase or decrease sports-fanaticism is essential. The sports-fanaticism is a psychological emotion that carries a blind hatred (hatred for no reason) against the competitive teams combined with a blind love toward the favorite teams. It is one of the social problems that have negative impacts on the social fabric. Fanatic attitudes are controlled by sentiment where emotional attitudes overcome mental attitudes. This paper aims to formulate the sports Arabic text into fanatic and anti-fanatic contexts to build a classification model using sentiment analysis. Sentiment analysis is a field of artificial intelligence that uses natural language processing, computational linguistics, and text analysis to systematically identify, extract, and analyze people’s sentiment. The proposed classification model is built to automatically classify Arabic social texts into a fanatic and anti-fanatic emotion. The methodology is sequentially described as follows: First, social text is formalized into twenty-one fanatic and anti-fanatic contexts. Second, fanatic lexicon with 1780 terms is developed. Third, 919 K domain-specific tweets are collected and labeled. Fourth, machine learning algorithms are applied to build a classification model. As a result, we have developed a classification tool that automatically evaluates, and classifies a given Arabic text against sports-fanaticism. Such a tool and model will help social sites to automatically filter messages posted by visitors to avoid sports-fanaticism. This research is a primary work to automatically detect religious fanaticism, political fanaticism, and other forms of fanaticism that have a negative impact on society.