Clustering Analysis Based on Short Video Content Producers
Jiaju Wu, Linggang Kong · 2021 China Automation Congress (CAC) · 2021
With the popularity of smart phones and the acceleration of people’s lives, short video apps like Douyin, Kuaishou, and Bilibili have become substitutes for current content and social media. This article conducts research from the perspective of short video content producers, selects the Bilibili platform, crawls the basic information of Bilibili users, and extracts the relevant characteristics of the short video content producers, including the total number of videos played, the number of videos, rank, number of fans, number of likes. k-means++ clustering analysis is performed on short video content producers. In the cluster analysis, the better number of clusters is obtained according to the intra-cluster error variance (SSE) and Silhouette Coefficient k, then The short video content producers of the Bilibili Platform are grouped into k clusters, and finally the potential correlation between the features and the clustering results is found, and feature differences between different groups of short video producers are compared.