Recent Papers

A rise in email triage as a result of a high volume of spam emails results in losses of USD 355 million annually. Sorting spam emails into categories like fraudulent or promotions from unwelcome parties is one approach to minimize this loss. Simple techniques, including word filters, served as the foundation for the early stages of the development of spam message classification. More sophisticated techniques are now being used, like machine learning-based language modeling. Networks using Recurrent Neural Units are among the most popular approaches to the text classication problem (GRU). GRU approaches were utilized because the purpose of this research is on the categorization of phishing emails. The findings of this investigation demonstrate that GRU achieved a high precision rate in the dropout-free situation. The constrained mailbox capacity is impacted by the sizeable volume of SPAM mail that is generated globally from numerous botnets. They have an impact on communication space loss as well as the safety of personal mail. They have an impact on the amount of time needed to recognize and respond to spam emails. Email spam identification is still regarded as a difficult task nowadays. This is due to the continued prevalence of email spam. It's because there is still much room for improvement in the identification. In order to detect spam emails, the author of this research creates a GRU-RNN. Using the Spam basis dataset, the new technique was evaluated. The method displays a 98.7% accuracy rate. The researcher comes to the conclusion that the suggested approach demonstrates an exceptional ability to recognize spam emails after completing extensive testing.

Copyright © 2023 S.Mani. This is an open access article distributed under the Creative Commons Attribution License.