南京师范大学学报（工程技术版）2025，Vol.25Issue(2)：96-101,6.DOI:10.3969/j.issn.1672-1292.2025.02.009
基于数据挖掘和情感分析的欺诈软件检测方法
Fraud Software Detection Method Based on Data Mining and Sentiment Analysis
摘要
Abstract
With the popularity of smartphones and the widespread use of mobile device applications,cybersecurity is facing unprecedented challenges,with fraud software emerging as a primary threat.Fraud software deceives users through false advertisements,deceptive content or fraudulent behaviours,leading to issues such as data breaches and financial losses,posing a significant threat to the information security of individuals and organisations.Therefore,this paper proposes a fraud detection method based on data mining and sentiment analysis,which comprehensively considers the visual characteristics of applications and user feedback to detect fraud software more efficiently and comprehensively.Through systematic data cleaning and manual review of applications on the Google Play Store,we constructed a sample dataset of 9 092 applications and conducted a comprehensive evaluation of our method.Experimental results show that our approach achieves a remarkable accuracy rate of 99.24%on real-world datasets,with lower false positive and false negative rates compared to alternative methods.This highlights the significant effectiveness and potential of our approach in the field of fraud software detection.关键词
安卓/移动应用/欺诈软件/数据挖掘/情感分析Key words
Android/mobile applications/fraud software/data mining/sentiment analysis分类
计算机与自动化引用本文复制引用
刘雨晨,王闪闪,赵振宇,欧阳谨,陈贞翔..基于数据挖掘和情感分析的欺诈软件检测方法[J].南京师范大学学报（工程技术版）,2025,25(2):96-101,6.基金项目
泰山学者工程项目(tsqnz20221146)、山东省自然科学基金青年项目(ZR2023QF096)、泉城省实验室重点项目(QCLZD202303). （tsqnz20221146）
