Spam E-mail Recognition Method Based on Neural Network Model
Yingchun Huang, Sen Xing · 2024
In order to improve the accuracy of spam E-mail identification, a neural network model based spam E-mail identification method (ILC, Inception-LSTM-CNN) is proposed. In this method, first of all, a convolutional neural network with improved convolutional kernel Inception module is designed to extract features from the mail text, and then the extracted features are passed into the LSTM layer to obtain context information representation, and finally, the binary output is carried out through the CNN layer. On two different data sets trec06p and uci's spambase, the proposed method was evaluated by using the traditional accuracy, accuracy rate, recall rate and F1-score evaluation indexes. The results show that: The proposed method is superior to SVM, CNN, LSTM-CNN and Incep-LC in each evaluation index.