Akademik Veri Yönetim Sistemi
3th International Prizren Scientific Research and Innovation Congress, Prizren, Kosova, 25 - 26 Nisan 2026, cilt.1, sa.1, ss.319-327, (Tam Metin Bildiri)
This study examines the use of artificial intelligence techniques in the analysis of social behavior from a big data and machine learning perspective. With the acceleration of digitalization, the use of social media platforms, online interactions, and sensor-based systems has become widespread. The amount of data generated through these tools has also increased significantly. This data provides a good resource for understanding the behavioral patterns of individuals and societies. However, analyzing this high-volume and heterogeneous data structure requires advanced computational techniques beyond classical statistical methods. In this context, the study proposes an analytical framework based on the integration of big data analytics and machine learning algorithms. In the proposed approach, data preprocessing, feature extraction, and modeling stages are systematically addressed. Social trends, behavioral clusters, and anomalies are identified using supervised and unsupervised learning methods. Furthermore, sentiment analysis is performed on text-based data using natural language processing techniques to quantitatively evaluate social perceptions and trends. Performance evaluation is conducted using metrics such as accuracy, precision, and generalization ability, and the findings demonstrate that artificial intelligence-based approaches achieve high success in social data analysis. Machine learning methods have been shown to offer significant advantages, particularly in uncovering complex and nonlinear relationships. Furthermore, the potential of artificial intelligence and big data technologies to transform analytical processes in the social sciences has been demonstrated. The proposed approach provides a powerful tool that can be used in numerous application areas, such as public policy development, social media analysis, crisis management, and predicting societal trends.