Big Data in Forensic Linguistics: Improving Authorship Attribution and Threat Detection in Cyber Contexts

Abstract

Forensic linguistics, the study of language in legal and investigative contexts, has gained increasing relevance in the digital era. The proliferation of online communication has created both challenges and opportunities for authorship attribution and threat detection.  This study explores how Big Data enhances forensic linguistic practices by enabling large-scale analysis of digital texts, such as emails, chat messages, and social media posts.  Using natural language processing (NLP), stylometry, and machine learning techniques, we analyze millions of documents to identify linguistic fingerprints, detect threatening language, and attribute authorship in cybercrime cases.  Results demonstrate that Big Data improves accuracy in identifying authorship through stylistic markers and enhances the detection of threats by analyzing lexical, syntactic, and pragmatic patterns. However, ethical concerns—including privacy, consent, and the risk of algorithmic bias—pose significant challenges. This article argues that Big Data-driven forensic linguistics represents a powerful tool for law enforcement and legal proceedings, but its application must be guided by strict ethical frameworks.  By combining linguistic theory, computational models, and Big Data analytics, forensic linguistics can significantly contribute to cybercrime prevention and the protection of digital communities.