حمیدرضا سوهانی

Nowadays, data is considered one of the most valuable assets of humanity. The availability of reliable data is essential and vital for all organizations, industries, companies, factories, and various operations. This importance stems from the fact that all decision-making, conclusions, predictions, and evaluations are based on data. In recent years, with the rapid growth of the internet, social networks, and various data collection and storage methods, we are confronted with vast volumes of diverse data, many of which hold significant analytical and evaluative value.The advancement of information and communication technologies has led to the continuous production and storage of large volumes of data. These data include user information, online behaviors, financial transactions, sensor data, and numerous other sources. Consequently, the need for analyzing and utilizing these data has increased significantly. Simultaneously, the rapid development of artificial intelligence, machine learning, and deep learning methods, along with substantial progress in hardware for computation and algorithm execution, has intensified the demand across various industries for this field of science. Therefore, the development and examination of machine learning algorithms have become increasingly necessary. These algorithms can help us leverage existing data to extract valuable insights that play a crucial role in decision-making and prediction processes.One of the fundamental and important approaches in machine learning and data analysis is data classification, which is a subset of supervised learning. Through this method, it is possible to identify patterns in the available data and predict future trends and behavior. For instance, in fields such as marketing, medicine, finance, and even sports, data analysis and classification can help achieve more accurate and beneficial outcomes.In this thesis, we aim to investigate and compare some of the most prominent classification algorithms and their application in predicting the results of football matches. In this research, various classification algorithms such as Random Forest, Support Vector Machine (SVM), and others will be examined. The objective is to determine the most accurate and effective method for predicting football match outcomes.Additionally, we aim to analyze the sentiments of fans of different teams over specific time periods using text mining models, by analyzing their tweets on Twitter. We will then examine the relationship between these sentiments and the actual match results.