Detection of Data Manipulation in Datasets Using Machine Learning

Detection of Data Manipulation in Datasets Using Machine Learning

RIMSHA ARFEEN1, RAJKUMAR KENCHA2, SRI LATHA ARIGE3,

MOHAMMAD ABDUL RASHEED4, P.BALAKISHAN5

1,4 UG STUDENT,CSE Department & Jyothishmathi Institute of Technology and Science

5 ASSOCIATE PROFESSOR, CSE Department & Jyothishmathi Institute of Technology and Science

Abstract - Data integrity is pivotal for achieving model  performance and delicacy and for making believable  opinions in   moment's data  wisdom and analytics  environment. This design   enforced and estimated a machine  literacy- driven   frame that can  descry data tampering through a generative  analysis of a structured dataset in its  original and acclimated  countries.  By  transubstantiating both datasets to match their structure, and  calculating a  point-full difference vector, the system  estimated  and  linked possible tampering in the acclimated dataset  through statistical analysis on  named ordered features,  similar as  Interquartile Range( IQR), entropy analysis, and Original Outlier  Factor( LOF). These named features were also drafted into a  Random Forest classifier that directly labelled each record  as either tampered or not tampered. The end product showed significant pledge in landing anomalies, similar as outliers,  null inserts, mismatching types and subtle shifts in value. The results indicated high perfection and recall on a range of  manipulated datasets. Through successive trial,  the system is promising for data  confirmation and  examination and  the expansion of forensic auditing systems. This  result is  modular and scalable, which gives the added benefit of sound  data integrity in critical  means like finance, healthcare, and  defense.

Key Words:  Data manipulation detection, data quality, anomaly detection, Interquartile Range (IQR), Local Outlier Factor (LOF), entropy analysis, skewness imputation, Shannon entropy, outlier detection, Random Forest, supervised classification, feature engineering, descriptive feature extraction, difference vectors, machine learning pipeline, validate data, data forensics, structured data comparison, ETL validation, automated dataset checking, and classification accuracy.