Attackers upload webshell into a web server to achieve the purpose of stealing data, launching a DDoS attack, modifying files with malicious intentions, etc. Once these objects are accomplished, it will bring huge losses to website managers. With the gradual development of encryption and confusion technology, the most common detection approach using taint analysis and feature matching might become less useful. Instead of applying source file codes, POST contents, or all received traffic, this paper demonstrated an intelligent and efficient framework that employs precise sessions derived from the web logs to detect webshell communication. Features were extracted from the raw sequence data in web logs while a statistical method based on time interval was proposed to identify sessions specifically. Besides, the paper leveraged long short-term memory and hidden Markov model to constitute the framework, respectively. Finally, the framework was evaluated with real data. The experiment shows that the LSTM-based model can achieve a higher accuracy rate of 95.97% with a recall rate of 96.15%, which has a much better performance than the HMM-based model. Moreover, the experiment demonstrated the high efficiency of the proposed approach in terms of the quick detection without source code, especially when it only considers detecting for a period of time, as it takes 98.5% less time than the cited related approach to get the result. As long as the webshell behavior is detected, we can pinpoint the anomaly session and utilize the statistical method to find the webshell file accurately.

With the proliferation of smartphones, participatory sensing using smartphones provides unprecedented opportunities for collecting enormous sensing data. There are two crucial requirements in participatory sensing, fair task allocation and energy efficiency, which are particularly challenging given high combinatorial complexity, trade-off between energy efficiency and fairness, and dynamic and unpredictable task arrivals. In this paper, we present a novel fair energy-efficient allocation framework whose objective is characterized by min–max aggregate sensing time. We rigorously prove that optimizing the min–max aggregate sensing time is NP hard even when the tasks are assumed as a priori. We consider two allocation models: offline allocation and online allocation. For the offline allocation model, we design an efficient approximation algorithm with the approximation ratio of $2−1m$, where m is the number of member smartphones in the system. For the online allocation model, we propose two algorithms: greedy algorithm and Robin-Hood algorithm, which achieve the competitive ratio of at most m and $m+1$, respectively. The results demonstrate that the approximation algorithm reduces over 81% total sensing time, the online greedy algorithm and Robin-Hood algorithms reduce the total sensing time 73% and 37.5%, respectively. The offline approximation algorithm and online greedy algorithm achieve much better min–max fairness compared to other algorithms.

With the continuous development of deep learning, more and more domains use deep learning technique to solve key problems. The security issues of deep learning models have also received more and more attention. Nowadays, malware has become a huge security threat in cyberspace. Traditional signature-based malware detection methods are not adaptable to the current large-scale malware detection. Thus many deep learning-based malware detection models are widely used in real malware detection scenarios. Therefore, we need to secure the deep learning-based malware detection models. However, model testing currently focuses on image and natural language processing models. There is no related work to test deep learning-based malware detection models specifically. Therefore, to fill this gap, we propose MalFuzz. MalFuzz uses the idea of coverage-guided fuzzing to test deep learning-based malware detection models. To solve the model state representation problem, MalFuzz uses the first and last layer neuron values to approximately represent the model state. To solve the new coverage calculation problem, MalFuzz uses the fast approximate nearest neighbor algorithm to compute the new coverage. The mutation strategy and seed selection strategy in image model or natural language processing model testing is not appropriate in deep learning-based malware detection model testing. Hence MalFuzz designs the seed selection strategy and seed mutation strategy for malware detection model testing. We performed extensive experiments to demonstrate the effectiveness of MalFuzz. Based on MalConv, Convnet, and CNN 2-d, we compared the modified TensorFuzz and MAB-malware with MalFuzz. Experiment results show that MalFuzz can detect more model classification errors. Likewise, the mutation operation of MalFuzz can retain the original functionality of malware with high probability. Moreover, the seed selection strategy of MalFuzz can help us explore the model state space quickly.

With the development of internet technology, email has become the formal communication method in modern society. Email often contains a large amount of personal privacy information, possible business agreements, and sensitive attachments, which make emails a good target for hackers. One of the most common attack method used by hackers is email XSS (Cross-site scripting). Through exploiting XSS vulnerabilities, hackers can steal identities, logging into the victim’s mailbox and stealing content directly. Therefore, this paper proposes an email XSS detection model based on deep learning technology, which can identify whether the XSS payload is carried in the email or not. Firstly, the model could extract the Sender, Receiver, Subject, Content, Attachment field information from the original email. Secondly, the email XSS corpus is formed after data processing. The Word2Vec algorithm is introduced to train the corpus and extract features for each email sample. Finally, the model uses the Bidirectional-RNN algorithm and Attention mechanism to train the email XSS detection model. In the experiment, the AUC (area under curve) value of the Bidirectional-RNN model reached 0.9979. When the Attention mechanism was added, the accuracy upper limit of the Bidirectional-RNN model was raised to 0.9936, and the loss value was reduced to 0.03.