With the exponential increase in digital threats, the traditional cybercrime reporting process remains largely unstructured and inaccessible to common users.This article proposes an AI-powered cybercrime reporting system that takes advantage of natural language processing and machine learning to offer an intelligent, guided interface for victims to report incidents.The system employs a fine-tuned RoBERTa-base model to classify cybercrimes into 23 predefined categories based on user descriptions and dynamically adjusts the reporting flow to collect appropriate data. Additionally, it enables secure digital evidence handling and automated PDF report generation for law enforcement. The proposed system improves reporting accuracy, user confidence, and evidence completeness, representing a transformative change in digital law enforcement support tools.
Introduction
Cybercrime reporting faces significant challenges due to unclear traditional systems, inconsistent definitions, and limited technical familiarity among users. Victims often struggle to file organized reports, and law enforcement lacks adequate training, making investigations difficult.
AI technologies, especially advanced models like RoBERTa-base, offer solutions by enabling victims to report crimes in natural language while guiding them through relevant questions. This improves crime classification, automates evidence collection, structures reports, and secures sensitive data with role-based access. Such AI-driven systems enhance communication between victims and authorities, streamline investigations, and improve response efficiency.
The literature highlights AI’s role in cybercrime detection, anomaly detection, natural language processing for threat intelligence, and AI-powered security management. However, challenges remain, including AI misuse by attackers, data quality issues, and the need for explainable AI. Current reporting platforms rely heavily on manual input and lack guidance or real-time feedback, limiting effectiveness.
The proposed AI-powered cybercrime reporting system uses a fine-tuned RoBERTa-base model trained on 23 cybercrime types. It features dynamic questioning to engage users, an evidence management system that complies with forensic standards, automated legal report generation, and user safety advice. The system architecture combines modern web technologies and secure backend frameworks, supported by an administrator dashboard to assist law enforcement in managing and analyzing reports effectively.
Conclusion
The study introduces a new system for reporting cybercrime that uses AI technology to make the process better. Traditional methods have several issues, but this system addresses them by combining smart classification, dynamic data collection, and secure evidence handling into one easy-to-use platform. The system uses a specially trained RoBERTa-base model to understand users\' natural language and accurately classify different cybercrimes. It includes a smart questioning tool and can generate structured PDF reports, ensuring that every report contains all necessary details and is immediately helpful for police. Tests show that this system greatly improves accuracy, speed, and user satisfaction. It is designed to grow and adapt, making it an excellent solution for reporting digital crimes in today\'s world. This system effectively bridges the gap between victims and investigators, helping them work together more efficiently in our evolving digital landscape.
References
[1] M. Morina, F. Azemi, M. Eren, I. Zejneli, & E. Papajorgj?, \"Crime scene in cybercrime criminal offenses: evidence management and processing\", Academic Journal of Interdisciplinary Studies, vol. 12, no. 2, p. 179, 2023. https://doi.org/10.36941/ajis-2023-0041
[2] E. Shukurov and U. Jafarov, \"Legal professionals\' perspectives on the challenges of cybercrime legislation enforcement\", ISSLP, vol. 2, no. 4, p. 25-31, 2023. https://doi.org/10.61838/kman.isslp.2.4.5
[3] G. Anis, A. Aboutabl, & A. Galal, \"Machine learning for detecting cybercrime in the banking sector\", Journal of Southwest Jiaotong University, vol. 58, no. 5, 2023. https://doi.org/10.35741/issn.0258-2724.58.5.60
[4] \"Digital forensics in cybercrime investigation\", International Journal of Science and Engineering Applications, 2024. https://doi.org/10.7753/ijcatr1310.1010
[5] M. Bruce, J. Lusthaus, R. Kashyap, N. Phair, & F. Varese, \"Mapping the global geography of cybercrime with the world cybercrime index\", Plos One, vol. 19, no. 4, p. e0297312, 2024. https://doi.org/10.1371/journal.pone.0297312
[6] A. Yeboah-Ofori and F. Opoku-Boateng, \"Mitigating cybercrimes in an evolving organizational landscape\", Continuity & Resilience Review, vol. 5, no. 1, p. 53-78, 2023. https://doi.org/10.1108/crr-09-2022-0017
[7] D. Dunsin, M. Ghanem, K. Ouazzane, & V. Vassilev, \"A comprehensive analysis of the role of artificial intelligence and machine learning in modern digital forensics and incident response\",, 2023. https://doi.org/10.2139/ssrn.4554035
[8] S. Hassan and A. Ibrahim, \"The role of artificial intelligence in cyber security and incident response\", International Journal for Electronic Crime Investigation, vol. 7, no. 2, 2023. https://doi.org/10.54692/ijeci.2023.0702154
[9] Y. Mohammed, M. Badara, & H. Dan’azumi, \"An intelligence-based cybersecurity approach: a review\", Journal of Intelligent Communication, vol. 4, no. 1, p. 32-43, 2024. https://doi.org/10.54963/jic.v4i1.232
[10] S. A and R. Kavitha, \"Unraveling cyber threats: the role of forensic investigation in cyber security\", IJMRSET, vol. 7, no. 05, p. 10258-10260, 2024. https://doi.org/10.15680/ijmrset.2024.0705108
[11] I. Yudhianto, \"Simple, fast, and accurate cybercrime detection on e-government with elastic stack siem\", JurnalEdukasi Dan PenelitianInformatika (Jepin), vol. 9, no. 2, p. 263, 2023. https://doi.org/10.26418/jp.v9i2.64213
[12] A. Hakim, K. Ramli, T. Gunawan, & S. Windarta, \"A novel digital forensic framework for data breach investigation\", Ieee Access, vol. 11, p. 42644-42659, 2023. https://doi.org/10.1109/access.2023.3270619
[13] K. Parti, T. Dearden, W. Foriest, J. E.Hawdon, P. Räsänen, Á. Szigetiet al., \"Cross-country comparison analysis of individual and institutional factors of anomie and online offending\", European Journal of Criminology, vol. 22, no. 2, p. 230-255, 2024. https://doi.org/10.1177/14773708241276944
[14] O. Lunhol and P. Torhalo, \"Artificial intelligence in law enforcement: current state and development prospects\",, 2024. https://doi.org/10.55295/psl.2024.ii12
[15] H. Hakim, C. Praja, & S. Ming-Hsi, \"Ai in law: urgency of the implementation of artificial intelligence on law enforcement in indonesia\", Jurnal Hukum Novelty, vol. 14, no. 1, p. 122, 2023. https://doi.org/10.26555/novelty.v14i1.a25943
[16] F. Sufi, \"A new ai-based semantic cyber intelligence agent\", Future Internet, vol. 15, no. 7, p. 231, 2023. https://doi.org/10.3390/fi15070231
[17] B. Sjølin, W. Hansen, A. Morin-Martinez, M. Petersen, L. Rieger, T. Veggeet al., \"Perqueue: managing complex and dynamic workflows\",, 2024. https://doi.org/10.26434/chemrxiv-2024-368wd
[18] M. Mahdi and S. Omeleze, \"Proof of concept of a digital forensic readiness cybercrime language as a service\", International Conference on Cyber Warfare and Security, vol. 19, no. 1, p. 191-199, 2024. https://doi.org/10.34190/iccws.19.1.2059
[19] S. Omeleze, R. Ikuesan, & H. Venter, \"Functional architectural design of a digital forensic readiness cybercrime language as a service\", European Conference on Cyber Warfare and Security, vol. 22, no. 1, p. 73-82, 2023. https://doi.org/10.34190/eccws.22.1.1240
[20] Y. Mohammed, M. Badara, & H. Dan’azumi, \"An intelligence-based cybersecurity approach: a review\", Journal of Intelligent Communication, vol. 4, no. 1, p. 32-43, 2024. https://doi.org/10.54963/jic.v4i1.232
[21] N. Rakha, \"Cybercrime and the law: addressing the challenges of digital forensics in criminal investigations\", Mexican Law Review, p. 23-54, 2024. https://doi.org/10.22201/iij.24485306e.2024.2.18892