DEFENDING MOBILE FINANCIAL USERS: SMISHING ATTACK PREVENTION AND CLASSIFICATION THROUGH MACHINE LEARNING

  • Unique Paper ID: 165122
  • Volume: 11
  • Issue: 1
  • PageNo: 213-218
  • Abstract:
  • Half of all daily transactions take place on the African continent, and by the end of 2021, it is expected that the value of all transactions worldwide will have reached $3 billion. Phishing incurs losses in the millions for both people and organizations. Smishing is a tactic used to fool mobile money system owners into sending virtual currency to their phones. The context and subtleties involved in gathering the attack’s specifics are the only distinctions between phishing and smishing. This research introduces a machine learning-driven approach for categorizing smishing messages aimed at mobile money users. The findings demonstrate that a combination of Extratree classifier feature selection and Random Forest, employing TFIDF vectorization, produces the most effective model, achieving an accuracy score of 99.86%. Smishing gets harder to spot as a consequence. By leveraging minimal resources vocabulary, several models and concepts for detecting smishing crimes were constructed. A machine-learning paradigm can be used to classify texts that have been smuggled.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 1
  • PageNo: 213-218

DEFENDING MOBILE FINANCIAL USERS: SMISHING ATTACK PREVENTION AND CLASSIFICATION THROUGH MACHINE LEARNING

Related Articles