Improving intelligent IR effectiveness in forensic analysis

Abstract

In the era of Information technology textual evidence is important to the vast majority of digital investigations. Important text-based evidence include Email, Internet browsing history instant messaging, system logs and so on. The investigator is flooded with data and has to spend valuable investigative time scanning through noisy search results and reviewing irrelevant search results. Current digital forensic text string search tools use match and/or indexing algorithms to search digital evidence at the physical level to locate specific text strings. The text string search tools fail to group and/or order search hits. This research uses text data mining principles and technologies for design and implementation which improves IIR (Intelligent Information Retrieval) effectiveness in digital forensics. The proposed system can analyze the corpus of mail data or SMS data with domain specific keywords. The searching and ranking of the mails in the proposed system is based on the weight of keywords of forensic interest. © 2012 ICST Institute for Computer Science, Social Informatics and Telecommunications Engineering.