Big social data analysis with computational intelligent algorithms

Social networks have long been used by various disciplines that are interested in social science or who feel the need to use this concept in their own right. This concept has undergone qualitative and quantitative changes in the field of application in the new century. New information technologies have led to the rapid and effective growth of social networks. These platforms, originally intended for communication between individuals or groups, have now evolved into a platform where millions of people reflect their political views, personal sentiments, or economic preferences. The amount of instant data they produce has reached incredible proportions. These data can be called social network data. The analysis of this data is within the scope of social network analysis. There are currently many social network analysis problems and their solution methods. But nowadays new problems arise with online social networks and various methods are used to solve these problems. These methods provide a solution to a certain extent but still need to be developed. In this thesis, online social networks, social network analysis and twenty-one online social network analysis problems are discussed in detail in a single study for the first time in the literature. Stance Detection problem, which is one of the most popular and interesting online social network analysis problems, was first considered as an optimization problem in this study. During the optimization study, three heuristic optimization algorithms were used, one of which was proposed for the first time, a new hybrid algorithm and the results of thirty-two classification algorithms used in the literature were compared. The method used is based on ensuring accurate classification of social network data using optimization technique. Although the proposed method is very new, promising results have been achieved. Key Words: Social Network, Social Network Analysis, Social Network Problems, Social Network Analysis Methods.