Machine Learning for Identifying and Validating Document Authenticity

Abstract

The aim of the study is to identify and validate information authenticity using machine learning (ML). The primary objective is to develop a robust model capable of accurately distinguishing between authentic and fraudulent documents. To achieve this, the study employs advanced techniques such as Res-Net 50, a deep learning architecture renowned for its image classification prowess, and SHA-256 cryptography, a secure hashing algorithm. The impressive outcomes of the model are showcased by its remarkable achievements, boasting a 99.26% accuracy level. This accomplishment is clearly reflected in both the confusion matrix and the classification report. This study underscores the potential of combining Res-Net 50 and SHA-256 cryptography in crafting a potent solution for identifying and validating document authenticity, with far-reaching implications for fraud detection and document verification processes.