Social network data mining (SNDM) technology shows great application potential in crime analysis and prevention. This study focuses on revealing the characteristics, laws, and trends of criminal behavior through an in-depth analysis of criminal information in social networks. Using data mining techniques such as association rule mining, cluster analysis, and community discovery, the key information and organizational structure of criminal networks are successfully mined, which provides a powerful means of investigation and prevention for public security departments. It is found that important criminal clues are hidden in the user communication data in social networks, and the communication mode and hidden information between criminals can be revealed through association rule mining technology. Cluster analysis helps to identify gangs and hot spots with similar criminal behaviors, which provides important clues for further investigation. In addition, community discovery technology further reveals the internal structure and membership relationship of criminal gangs, which is helpful in deeply understanding the operation mode of criminal organizations and the spread path of criminal acts. Based on historical data and mining results, this study also constructs a crime trend prediction model, which provides timely early warning information for public security departments and helps to take measures to prevent and crack down on criminal acts in advance. On the whole, this study not only enriches the application theory of SNDM technology in the field of crime analysis but also provides new ideas and tools for actual crime investigation and prevention.