At present, there are some problems in the social security of employees in enterprises, such as incomplete security and low reliability. According to this background, this paper studies a topological method for enterprise social security data analysis based on data adaptive analysis strategy and strong learning data stream coprocessing. According to this background, this paper studies a topology method of enterprise social security data analysis based on data adaptive analysis strategy and strong learning data flow and designed a convolutional neural network method based on mutual interference deepening strategy and advanced learning classification mode convolution neural network based on mutual interference deepening strategy and intensive learning classification pattern. According to the coverage error strategy of different types of data, the high-precision matching analysis of enterprise employee social security data is realized, and the Cartesian formula is used to correct the error and topology analysis of the analysis results. According to the experimental data and results, the topology method of enterprise social security data analysis is based on the experimental data and results. We can know that the topological method of enterprise social security data analysis is based on convolutional neural network. Enterprise social security theory can effectively improve the scope and speed of social security. This theory effectively completes the high-precision matching of different types of data and indirectly improves employees’ cognition of enterprise values.