Enhancing Data Integrity in Mobile Crowdsensing Environment with Machine Learning and Cost-Benefit Analysis

Abstract

A model has been proposed to achieve better data integrity by filtering out fake reviews from real-time data sets using the machine learning approach. Our model uses data fuzzification over a mathematical model that categorizes users or customer feedback using ratings provided by Customers or reviewers in Mobile Crowdsensing Environment. In this model, users can provides feedback for the desired location through various electronics gadgets using the specifically developed Android App or web-based applications. This feedback will be stored in a cloud platform. The said dataset can be analyzed through fuzzy logic to detect genuine reviews for maintaining data integrity, which can be used in various real-time applications, such as medical, tourism, education, etc. It also categorizes into three categories such as honest, suspicious, and malicious. Further accuracy of the proposed model has been judged using various machine learning (ML) algorithms such as Naive Bayes (NB), Bayes Net(BN), Support Vector Machine(SVM), Decision Tree(J48), and Random Forest(RF) in Cross-Validation modes. Initially, it achieves 99.79% of accuracy using the Random Forest algorithm that has been enhanced to 100% using cost-benefit analysis in cross-validation mode.