No abstract

The phishing email is one of the significant threats in the world today and has caused tremendous financial losses. Although the methods of confrontation are continually being updated, the results of those methods are not very satisfactory at present. Moreover, phishing emails are growing at an alarming rate in recent years. Therefore, more effective phishing detection technology is needed to curb the threat of phishing emails. In this paper, we first analyzed the email structure. Then, based on an improved recurrent convolutional neural networks (RCNN) model with multilevel vectors and attention mechanism, we proposed a new phishing email detection model named THEMIS, which is used to model emails at the email header, the email body, the character level, and the word level simultaneously. To evaluate the effectiveness of THEMIS, we use an unbalanced dataset that has realistic ratios of phishing and legitimate emails. The experimental results show that the overall accuracy of THEMIS reaches 99.848%. Meanwhile, the false positive rate (FPR) is 0.043%. High accuracy and low FPR ensure that the filter can identify phishing emails with high probability and filter out legitimate emails as little as possible. This promising result is superior to the existing detection methods and verifies the effectiveness of THEMIS in detecting phishing emails.

No abstract

In recent years, the number of vulnerabilities discovered and publicly disclosed has shown a sharp upward trend. However, the value of exploitation of vulnerabilities varies for attackers, considering that only a small fraction of vulnerabilities are exploited. Therefore, the realization of quick exclusion of the non-exploitable vulnerabilities and optimal patch prioritization on limited resources has become imperative for organizations. Recent works using machine learning techniques predict exploited vulnerabilities by extracting features from open-source intelligence (OSINT). However, in the face of explosive growth of vulnerability information, there is room for improvement in the application of past methods to multiple threat intelligence. A more general method is needed to deal with various threat intelligence sources. Moreover, in previous methods, traditional text processing methods were used to deal with vulnerability related descriptions, which only grasped the static statistical characteristics but ignored the context and the meaning of the words of the text. To address these challenges, we propose an exploit prediction model, which is based on a combination of fastText and LightGBM algorithm and called fastEmbed. We replicate key portions of the state-of-the-art work of exploit prediction and use them as benchmark models. Our model outperforms the baseline model whether in terms of the generalization ability or the prediction ability without temporal intermixing with an average overall improvement of 6.283% by learning the embedding of vulnerability-related text on extremely imbalanced data sets. Besides, in terms of predicting the exploits in the wild, our model also outperforms the baseline model with an F1 measure of 0.586 on the minority class (33.577% improvement over the work using features from darkweb/deepweb). The results demonstrate that the model can improve the ability to describe the exploitability of vulnerabilities and predict exploits in the wild effectively.

No abstract