Increasing The Efficiency of Damaged File Detection Tools Based on The Use of Hidden Markov Models

Abstract

This paper reviews a new algorithm developed to improve the performance of file detection based on hidden Markov models (HMM). Hidden Markov models are an effective method for detecting various types of disturbances and damages using time series and probabilities. The algorithm uses learning-based technologies to quickly and accurately detect file corruption. The article is based on recovery of damaged files, detection and analysis of changes in their structure. Monitoring file corruption processes through hidden Markov models increases the possibility of correctly predicting errors. The algorithm's performance is more efficient in more complex structures compared to simple statistics, it quickly detects file integration violations and creates a recovery mechanism. The main part of the article provides a detailed explanation of the mechanisms for explaining and predicting possible file corruption or invalidation using HMM. The new algorithm is designed to improve security and speed up the recovery of damaged files.