Testing the Forensic Interestingness of Image Files Based on Size and Type

Abstract

In this thesis, we investigate the relationship between the size and type of a file and its forensic usefulness. We investigate GIF, MP3, MP4, PNG, and JPEG files found in a large collection called the Real Drive Corpus, and the files classification as software-based, entertainment-based, or personal. Results of these experiments were compared to prior work to find interesting files. Results show that the previous experiments were effective at marking interesting files as interesting, but there were still a lot of uninteresting files that were marked as interesting. Also, the results do not show a correlation between the interestingness of a file, its type, and its size.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 2017
Accession Number
AD1046839

Entities

People

  • Raymond M. Goldberg

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Audio Files
  • California
  • Classification
  • Computational Forensics
  • Computers
  • Crime
  • Criminal Investigations
  • Data Mining
  • Data Transmission
  • Databases
  • Graphical User Interface
  • Information Science
  • Operating Systems
  • Security
  • United States
  • User Interface
  • Web Browsers

Fields of Study

  • Computer science

Readers

  • Canadian European Scientific Immigration and Epilepsy Clearance Studies
  • Database Systems and Applications
  • Regression Analysis.