A Big Data-based Approach to Web User Behavior Analysis

Abstract

In China, the Internet has developed to a relatively mature scale, and Internet applications have gradually transitioned from being singular to diversified. The Internet is changing people's ways of learning, working, and living, and even influencing the progress of the entire society. Against the backdrop of rapid Internet development, we have gradually entered the "big data era." Faced with such a vast amount of data, single-machine statistics have become inadequate. This paper first introduces the background and significance of the research, providing a preliminary description of network traffic from the perspective of network services. It then elaborates on the concepts and classifications of network user behavior, along with key data, mainly introducing analysis methods. The primary method used in this paper is cluster analysis, including distance and similarity coefficients in cluster analysis, with a focus on the main steps and algorithm process of k-means clustering. Finally, the paper conducts user behavior analysis based on the clustering results.