Institutional Repository of Vilnius University
Analysis of data from popular messaging applications on android devices /
Abstract
dc:descriptionSmartphones have become essential communication tools in everyday life, offering unmatched connectivity and functionality. However, as these technologies advance, new security and privacy risks continue to emerge. Android devices dominate the mobile market and store enormous amounts of user data, making them valuable sources for forensic analysis. This thesis aims to develop a dynamic tool designed to extract and analyze messaging data from Android devices. The tool applies two machine learning classifiers: Naive Bayes and Linear Support Vector Machine (Linear SVM), to categorize messages and identify conversation topics. The tool was created with the consideration that many existing tools are not compatible with all Android mobile devices. The implementation uses Bash and Python scripts and operates in a Linux environment, extracting data through ADB (Android Debug Bridge) communication. The research focuses on two popular messaging applications: WhatsApp and Viber, with a mechanism that allows adding new applications through separate loaders. The tool helps to extract message databases from the device, processes and formats the data for analysis, and generates visual representations of the results. Both the forensic message timeline and classifier outputs are compiled into PDF reports for easy review. Testing was conducted using Samsung and Xiaomi devices running Android 16. The tool successfully extracted message corpora, performed classification analysis, and produced PDF reports suitable for forensic examination.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Niewulis, Matas,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD277690073&prefLang=en_US