E-Commerce Information System Management Based on Data Mining and Neural Network Algorithms

Abstract

The rapid development of artificial intelligence technology has led to rapid development in various fields. It has many hidden related customer behavior information and future development trends in the e-commerce information system. The data mining technology can dig out useful information and promote the development of e-commerce. This research analyzes the significance and advantages of data mining technology in the application of e-commerce management systems and analyzes the related technologies of data mining and future trend prediction. This research has taken the advantages of clustering and naive Bayesian methods in data mining to classify product information and purchase preferences and other information and mine the associated data. Then, the nonlinear data processing advantages of neural networks are used to predict future purchasing power. The results show that data mining technology and neural networks have high accuracy in predicting future consumer purchasing power information. The correlation coefficient between real consumption data and predicted consumption data reached 0.9785, and the maximum relative average error was only 2.32%. It fully shows that data mining technology can obtain some unrecognizable related information and future consumption trends in e-commerce systems, and neural networks can also predict future consumption power and consumption patterns well.

Data Mining Modeling in Clustering Car Products Sales Data in the Automotive Industry in Indonesia

“…In Figure 1, there are numerous procedures or steps that must be performed before initiating the clustering process. Data collection, data preprocessing, data extraction and visualization, model building, and model evaluation are the processes involved [25], [26]. The data mining procedure employing clustering algorithms is carried out using the Microsoft Excel and RapidMiner programs.…”

Application of Gradient Boosting Regression Model in Intelligent Distribution of E-commerce

“…Improper datasets have a significant impact on prediction accuracy when GBRM techniques have high standards for datasets. The processes of data collection and data preprocessing [22] comprise the data preparation process. Selecting appropriate data is important for assignments involving research, including finding the frequency and how much customers purchase.…”

Research on Segmenting E-Commerce Customer through an Improved K-Medoids Clustering Algorithm

“…Third, with the continuous development of data mining technology, the indicator selection methods based on customer behavior are becoming a hot topic. In these literatures, multidimensional features are used to reflect the consumption behaviors and habits of different customer groups [ 17 , 18 ]. As a classic customer value model, the RFM model has been successfully applied to customer segmentation [ 19 , 20 ].…”