Title: Anomaly detection method of social media user information based on data mining

Authors: Xiaoyan Wan

Addresses: Qingdao Vocation and Technical College of Hotel Management, Qingdao, 266100, China

Abstract: Aiming at the problems of low detection accuracy, recall and F1 value of traditional social media user information anomaly detection methods, a social media user information anomaly detection method based on data mining is proposed. Firstly, we clean the social media data and eliminate the invalid and missing values in the data. Then, we filter the abnormal user information in the social media data through the unsupervised k-means algorithm in data mining. Finally, according to the screening results, we calculate the text word segmentation of user information, obtain the similarity of word frequency, and complete the detection of abnormal user information. The method provided by social media has the highest accuracy of 97.5%, which is the highest exception detection rate of 97.5% and the highest exception detection rate of social media.

Keywords: data mining; social media; user information; K-means; Weibo.

DOI: 10.1504/IJWBC.2024.136674

International Journal of Web Based Communities, 2024 Vol.20 No.1/2, pp.38 - 50

Received: 14 Feb 2022
Accepted: 09 Jun 2022
Published online: 15 Feb 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article