With the explosive growth of data, people's demand for data analysis has become more intense. Although modern technology can collect a large amount of data, the collected original data is often useless and contains little information. How to extract useful information from massive amounts of data has become an urgent problem. Driven by artificial intelligence (AI) technology and personalized consumption demand of users, this article puts forward a dynamic interactive and visualization algorithm of e-business database information based on an improved collaborative filtering (CF) algorithm to help enterprises more efficiently mine the required potential customer groups from massive customer data and log data. Experiment results prove the effectiveness of the model and algorithm. Data mining (DM) technology is applied to the user access control model in this model. First, the maximum forward reference sequence of mobile e-business groups is mined by data technology. Then a user access control model is established according to this sequence to control user access so enterprises can formulate reasonable marketing strategies based on this knowledge.