In this era of social networking, any field of work relies heavily on social interaction and information flow. Social interaction forms the foundation of everyday and permanent phenomena that must be understood from different perspectives to provide smarter and more effective solutions for progress and productivity. In this research, we analyze the characteristics and features of social interaction from two perspectives - content and theme - using various case studies and providing multidimensional coverage on a large scale. Content analysis is a granular analysis that explores the structure, relationships and conversational interaction of textual materials. Theme analysis is a pattern analysis of communication data that focuses on an overall perspective. Our goal is to conduct an in-depth analysis of social interaction and research it from a collective perspective of different interaction forms. For the first perspective, we focus on content-based analysis that centers on textual information as the primary source of research. Text on Weibo is considered to be full of sentiment and emotion. By implementing deterministic categorization methods and combining multidimensional positivity and arousal spaces, we quantitatively predict the emotion of the text, which poses challenges such as classification accuracy and label inconsistency. This research attempts to overcome the aforementioned limitations by proposing a text-based emotion identification probability model named TERMS, which can simultaneously reduce the error rate of deterministic emotion classifiers and the impact of rating discrepancies caused by annotators. For the second perspective, theme analysis, we experiment with social interaction data (node degree, call volume and frequency) in various countries around the world and analyze their impact on population. Population growth, also known as urbanization, is an increasingly serious issue that brings about changes in urban areas, such as economic, infrastructure, and social interaction attributes. Studying these changes requires scientific understanding. By configuring a network called a contact network and using power law rules to analyze changes in close social interactions, we record the impact of population on social interaction in different cities around the world and present them in various comprehensive and individual levels. Therefore, for this purpose, case studies conducted in theme and content analysis emphasize the features of social interaction and explore social interaction with different scales of environment and parameters to demonstrate the nature of social interaction used to achieve any goals.