Details

Autor(en) / Beteiligte

• Liu, Yao
• Liu, Lu
• Liu, Qiao
• Lan, Tian
• Bai, Xiaoyu
• Zhou, Le

Titel

Unsupervised Mobile User Behavior Detection Based on Siamese Neural Networks

Ist Teil von

• 2023 IEEE International Conferences on Internet of Things (iThings) and IEEE Green Computing & Communications (GreenCom) and IEEE Cyber, Physical & Social Computing (CPSCom) and IEEE Smart Data (SmartData) and IEEE Congress on Cybermatics (Cybermatics), 2023, p.518-523

Ort / Verlag

IEEE

Erscheinungsjahr

2023

Quelle

IEEE Xplore

Beschreibungen/Notizen

• In recent years, mobile intelligent terminals such as smartphones and tablets have achieved increasing commercial success and have become indispensable elements in the daily lives of billions of people. People use mobile devices for various activities such as communication, finance, gaming, video conferencing, and shopping, among others. These activities generate a vast amount of data, making it increasingly important to classify and understand the behavior patterns of mobile applications (Apps). The behavior patterns of applications on mobile intelligent devices are complex and dynamic, requiring a deeper understanding of user behavior and App characteristics for classification. Therefore, this paper proposes a mobile App user behavior recognition model based on the siamese neural network framework to study the user behavior patterns of different Apps on mobile devices. This model extracts features from user and App behavior data, performs comparisons and classification, and identifies different types of behavior patterns. Additionally, to address the issue of data imbalance, this model innovatively utilizes silhouette coefficients for data preprocessing. The research presented in this paper will contribute to a better understanding of the relationship between users and Apps, as well as the impact of different behavior patterns on the mobile ecosystem.