• MAHENDIHASAN HEERA

This project proposes a deep learning approach Bidirectional Encoder Representations from Transformers (BERT) for email spam detection that automatically learns and extracts relevant features from email data. The proposed deep learning model utilizes a recurrent neural network (RNN) architecture to capture sequential dependencies and patterns within email content. In this work we will use traditional machine learning algorithms, including Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Multinomial Naive Bayes Classifier (MNB), Decision Tree Classifier (DT), Logistic Regression (LR) and Random Forest Classifier (RF), to learn from a labeled dataset of spam and non-spam emails. These algorithms extract relevant features from the email content, such as the presence of specific keywords, structural patterns, and metadata, and use these features to train classification models. To enhance the model's performance, various pre-processing techniques are employed, including tokenization, stop-word removal, and word embedding. These techniques enable the model to handle different email formats and reduce the dimensionality of the input, improving computational efficiency. The project work gives the accuracy of higher accuracy model shows that the model can predict spam and non-spam emails.

Email Spam detection, Machine, Learning, Deep Learning, Dataset

"CORRESPONDENCE JUNK DISCOVERY EXPENDING APPLIANCE CULTURE AND DEEPLEARNING BOLDNESS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 7, page no.251-255, July-2018, Available :http://www.jetir.org/papers/JETIR1807A37.pdf

"CORRESPONDENCE JUNK DISCOVERY EXPENDING APPLIANCE CULTURE AND DEEPLEARNING BOLDNESS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 7, page no. pp251-255, July-2018, Available at : http://www.jetir.org/papers/JETIR1807A37.pdf