Article Preview
TopIntroduction
Data mining (DM) is performed to find rules from data and mine valuable data from these data, so it is of great significance to study the DM algorithm. The traditional DM algorithm is not does not deal with a lot of data and information, the process is not as complicated as it is now, and all data are processed by a single machine. However, after entering the network era, it has become essential to improve the efficiency of large DM in the face of massive information and data (Guan et al., 2022). Network log mining refers to the use of DM through an in-depth analysis of a large quantity of user access records in Web database logs. where one can find information such as user access patterns and hobbies, and provide data support for website management (Yang & Jin, 2020). DM refers to finding correlations from a large quantity of data to be mined, and extracting adequate information by using the correlation between data (Li & Cheng, 2018). Mobile e-business groups have many visits, and the network needs to process a large amount of data. Some lawless elements use mobile e-business to cheat and steal user information (Su, 2017). If the user access control model in mobile e-business is not efficient enough, it will lead to the disadvantages of low classification accuracy and insufficient security. Suppose one can get the utmost out of network log files to mine the changes of users' access patterns to websites and the columns they care about. In that case, it can support the structural design of websites, the setting of search engines, and also lay the foundation for developing personalized services for specific user groups (Li & Miao, 2017).
It is difficult for enterprise decision-makers to make crucial decisions using the massive data information in the database, but only by their intuition (Matsumura et al., 2016). Taking the Web database logs as an example, some Web hotspots' log data are increasing at tens of megabytes every day (Ren & Song, 2017). Finding useful and important knowledge from this massive data is another necessary research and application field of DM and knowledge discovery (Jiang L, 2017). A great deal of interactive information between users and merchants is stored in the Web database, which includes users' browsing information, users' registration information, etc (Bauer & Jannach, 2017). This interactive information is stored in the database in the form of logs, and merchants need to mine this information from the database to discover the behavior regularity of users (Zhang et al., 2018). Driven by AI technology and personalized consumption demand of users, this article puts forward a dynamic interactive and visualization algorithm of network database information based on an improved CF algorithm to help enterprises tap users' needs and make more personalized marketing plans.
With the rapid growth of computer software, hardware, and the Internet, data collection, storage, processing, and analysis in various fields have greatly progressed (Feng et al., 2019). Millions of users in different regions are constantly consuming, producing, and disseminating massive and highly diversified heterogeneous data, ranging from virtual social media data on a global scale to residents' travel data in a single city (Songet al., 2019). This article mainly studies the dynamic interaction and visualization design of database information based on AI: