| Title |
Ontology-driven digital profiling for identification and linking evidence across social media platform / |
| Authors |
Grigaliūnas, Šarūnas ; Brūzgienė, Rasa ; Venčkauskas, Algimantas |
| DOI |
10.1109/ACCESS.2023.3322162 |
| Full Text |
|
| Is Part of |
IEEE Access.. Piscataway, NJ : IEEE. 2023, vol. 11, p. 111672-111691.. ISSN 2169-3536 |
| Keywords [eng] |
social media ; digital evidence ; profiling ; digital forensics ; digital artefacts |
| Abstract [eng] |
In an era in which social media platforms are proliferating and becoming primary communication channels, the identification of evidence for crimes from such platforms is crucial for digital forensics and legal proceedings. This paper presents a novel approach for systematically structuring and categorising digital attributes that are interlinked across social media platforms using digital ontologies, as well as a method for user profiling using domain-specific digital artefacts. The ontology models consist of classes with subclass distinctions for text, image, and video types of evidence. These models are flexible and can be expanded to include various social media platforms and evidence categories. Simultaneously, the user profiling method employs mathematical formulas and visual representations to develop comprehensive profiles of individuals based on extracted social media data. This methodology evaluates the relevance of a set of digital artefacts and related attributes, such as interests, location, and activities, using their weights. Additionally, the research addresses the legal and ethical considerations pertinent to the collection of data from social media. Despite the approaches’ immense potential for expediting evidence collection and developing insightful profiles, obstacles such as scalability, legal complexities, and data noise are identified. This work makes a substantial contribution to the development of digital forensics and cybercrime investigations involving social media platforms. |
| Published |
Piscataway, NJ : IEEE |
| Type |
Journal article |
| Language |
English |
| Publication date |
2023 |
| CC license |
|
