DIGITAL DNA FILE THREAT ANALYSIS SYSTEM

Nowadays, digital file sharing has increased rapidly through email, websites, and cloud platforms. Along with this
growth, malicious files such as executable scripts, infected archives, and hidden threats have also increased.
Traditional antivirus systems mainly depend on signature-based detection, which is not always effective for detecting
new or unknown threats. In this paper, we propose a Digital DNA-based file threat analysis system that analyzes the
internal characteristics of files instead of executing them. The system extracts features such as file type, size, and hash
signature and generates a confidence score. Based on this score, the file is classified into three categories: GOOD,
SUSPICIOUS, and VERY POOR. The system is implemented as a web application using FastAPI for backend
processing and a modern frontend interface. Results show that the system can effectively classify files while
maintaining safety and simplicity