Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68694
Volume:
10 |
11 |
Nov 2026
Pages:
1-4
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

f Artificial Intelligence & Data Science (AI&DS)
Shri Swami Samarth Institute Of Management And Technology Malwadi, Bota

Mechanical
Shri Swami Samarth Institute Of Management And Technology Malwadi, Bota

E&TC
Shri Swami Samarth Institute Of Management And Technology Malwadi, Bota
Scam messages increasingly use urgency, impersonation, requests for credentials or payments, shortened links, and social engineering to persuade recipients to act quickly. Keyword-only filters can miss new wording and can also incorrectly flag legitimate messages. This proposal presents AI for Cyberscam Prevention, a student-scale system that classifies potentially scam-like messages and provides evidence-based explanations in plain language. The system combines a transparent baseline using TF-IDF features and logistic regression with a stronger comparison model if resources permit. It extracts message-level signals such as urgency phrases, credential or payment requests, suspicious URL characteristics, impersonation cues, and linguistic patterns. Rather than presenting a model score as proof, the interface displays the relevant signals, uncertainty, and a safe next step. Evaluation will use public, appropriately licensed datasets and/or synthetic examples, with careful separation of training and test data. Metrics include precision, recall, F1-score, false-positive rate, calibration, and performance across message categories. A privacy-by-design approach avoids retaining unnecessary personal messages and excludes live inbox access from the first prototype. The proposal’s contribution is a reproducible comparison of model performance and explanation usefulness for digital-safety warnings; all performance claims must be measured during implementation.
KEYWORDS
The proposed framework presents a practical AI&DS research project focused on safer digital communication. Its distinguishing emphasis is not only message classification, but also evidence-linked explanations, privacy-minimizing design, and systematic analysis of false alarms. A careful baseline, transparent evaluation, and clear limitations can make the project academically defensible and useful in a portfolio. The next step is to review recent related work, confirm the dataset license, implement the pipeline, and report actual experimental results.
[1] National Institute of Standards and Technology. (2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1. https://doi.org/10.6028/NIST.AI.100-1 [2] National Institute of Standards and Technology. (2020). Security and Privacy Controls for Information Systems and Organizations, NIST SP 800-53 Rev. 5. https://doi.org/10.6028/ NIST.SP.800-53r5 [3] OWASP Foundation. (2025). Phishing and related awareness resources. Verify resource via https://owasp.org/ [4] Scikit-learn developers. (2024). Working with text data and model evaluation documentation. https://scikit-learn.org/stable/ [5] National Cyber Security Centre (UK). (2024). Phishing attacks: defending your organisation. https://www.ncsc.gov.uk/ [6] Verma, A., & Shah, R. (2023). Explainable AI for Phishing Detection: A Survey of Feature-Level Interpretability. IEEE Transactions on Information Forensics and Security, 18, 1245-1258. [7] Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). "Why Should I Trust You?": Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135-1144. [8] Zhang, X., & Yuan, Y. (2024). Deep Learning vs. Shallow Classifiers in Cyberscam Detection: A Performance and Compute Trade-off Analysis. Computers & Security, 136, 103560. [9] Kumar, S., & Joshi, P. (2025). Privacy-by-Design in Natural Language Processing: Methods for Minimizing Data Retention. Journal of Cybersecurity and Privacy, 5(2), 180-195. [10] Lundberg, S. M., & Lee, S.-I. (2017). A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems (NeurIPS 2017), 30, 4765-4774.
Publication Date
11/5/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
5
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...