Open AccessDetection of Visual Similarity Snooping Attacks in Emails using an Extended Client Based Technique
Author(s) -
Mwangi Muhindi George,
Geoffrey Mariga Wambugu,
Aaron Mogeni Oirere
Publication year - 2021
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.d2296.0410421
Subject(s) - flagging , computer science , python (programming language) , cosine similarity , similarity (geometry) , replica , classifier (uml) , artificial intelligence , data mining , pattern recognition (psychology) , machine learning , image (mathematics) , operating system , art , visual arts , archaeology , history
This paper provides an Extended Client BasedTechnique (ECBT) that performs classification on emails usingthe Bayessian classifier that attain in-depth defense by performingtextual analysis on email messages and attachment extensions todetect and flag snooping emails. The technique was implementedusing python 3.6 in a jupyter notebook. An experimental researchmethod on a personal computer was used to validate the developedtechnique using different metrics. The validation results produceda high acceptable percentage rate based on the four calculatedvalidation metrics indicating that the technique was valid. Thecosine of similarity showed a high percentage rate of similaritybetween the validation labels indicating that there is a high rate ofsimilarity between the known and output message labels. Thedirection for further study on this paper is to conduct a replicaexperiments, which enhances the classification and flagging ofthe snooped emails using an advanced classification method.