Malicious code is an ever-growing security threats to computer systems and networks, while malware detection provides effective defense against malicious codes. In this paper, a brief overview is presented on currently prevalent methods to detect malicious codes, including signature-based methods, behavioral-based detection and machine learning (ML) based ones. More specifically, the potentially effective malicious features are summarized and the novel methods using ML are deeply discussed. Furthermore, an ensemble interpretable framework is explored for automatic and efficient malicious code detection. Based on the knowledge graph of malware, the novel framework inclines to achieve robust malware detection even confronted with unseen malicious codes. Finally, both advantages and disadvantages are discussed and experimental results are outlined to verify the effectiveness of the novel methods.

Due to the progressive complexity of traffic networks, the traffic pressure increases sharply and traffic accidents occur frequently, which is largely posed by imprecise traffic information service provided in existing Intelligent traffic service systems (ITSSs). To relieve traffic information trek encountered in complex traffic networks, the personalized traffic information recommendation based on click through rate (CTR) prediction has attracted extensive attention. However, the data sparsity and cold start problems in traditional recommendations hinder their real-time applications in ITSSs. In this paper, the multisource data with different types of context information are utilized to construct a personalized fine-grained recommendation method for intelligent traffic services (PF-ITS), which includes three components: a road condition optimization strategy (RCOS) to capture users' behavior preferences and traffic patterns, an encoder–decoder long short-term memory (LSTM) model to address the data sparsity and cold start problems with rich context information, and an improved DeepFM model based on attention mechanism (DeepAFM) to exert personalized fine-grained traffic information recommendation with embeddings of multisource data. More specifically, the RCOS is proposed to perform coarse-grained route recommendation based on path planning theory, in which the traffic data and drivers' preferences are fully utilized for comprehensive modeling. The encoder–decoder LSTM model is employed for representation learning, in which the traffic sequences and driving behaviors are mapped into dense distributed representations with rich semantic information. The DeepAFM is utilized to achieve effective driving safety guarantee according to an individual's requirements, in which the weighted low-order feature combinations and high-order feature interactions are incorporated for personalized fine-grained recommendation. We also conduct extensive experiments on public data sets and in real-world scenarios. The experimental results demonstrate that the PF-ITS method based on RCOS and DeepAFM outperforms the state-of-the-art baseline models in effectiveness and efficiency.

Emotion recognition has become a research focus in the brain–computer interface and cognitive neuroscience. Electroencephalogram (EEG) is employed for its advantages as accurate, objective, and noninvasive nature. However, many existing research only focus on extracting the time and frequency domain features of the EEG signals while failing to utilize the dynamic temporal changes and the positional relationships between different electrode channels. To fill this gap, we develop the dynamic differential entropy and brain connectivity features based EEG emotion recognition using linear graph convolutional network named DDELGCN. First, the dynamic differential entropy feature which represents the frequency domain feature as well as time domain feature is extracted based on the traditional differential entropy feature. Second, brain connectivity matrices are constructed by calculating the Pearson correlation coefficient, phase-locked value and transfer entropy, and then are used to denote the connectivity features of all electrode combinations. Finally, a linear graph convolutional network is customized and applied to aggregate the features from total electrode combinations and then classifies the emotional states, which consists of five layers, namely, an input layer, two linear graph convolutional layers, a fully connected layer, and a softmax layer. Extensive experiments show that the accuracies in the valence and arousal dimensions reach 90.88% and 91.13%, and the precision reaches 96.66% and 97.02% on the DEAP dataset, respectively. On the SEED dataset, the accuracy and precision reach 91.56% and 97.38%, respectively.

Selfish mining attacks sabotage the blockchain systems by utilizing the vulnerabilities of consensus mechanism. The attackers' main target is to obtain higher revenues compared with honest parties. More specifically, the essence of selfish mining is to waste the power of honest parties by generating a private chain. However, these attacks are not practical due to high forking rate. The honest parties may quit the blockchain system once they detect the abnormal forking rate, which impairs their revenues. While selfish mining attacks make no sense anymore with the honest parties' departure. Therefore, selfish miners need to restrain when launch selfish mining attacks such that the forking rate is not preposterously higher than normal level. The crux is how to illustrate the attacks toward the view of honest parties, who are blind to the private chain. Generally, previous works, especially those using Markov decision processes, stress on the increment of attackers' revenues, while overlooking the detection on forking rate. In this paper, we propose, to maintain the benefit from selfish mining, an improved selfish mining based on hidden Markov decision processes (SMHMDP). To reduce the forking rate, we also relax the behaviors of selfish miners (also known as semi-selfish miners), who mine on the private chain, to mine on public chain with a small probability ρ. Simulation results show that SMHMDP can trade off between revenues and forking rate. Put differently, selfish miners benefit from attacking within an acceptable forking rate toward the view of honest parties, without leading selfish mining attacks to be an armchair strategist.

vMixing coins strategy can realize the anonymity of user information, thereby protecting the user's privacy. Ideally, the blacklist is public information and all bad coins are recorded in it. However, due to the failure of some bad coins to be registered in the blacklist in time, users can only obtain part of the blacklist information, which allows illegal criminals to take advantage of it. How to prevent illegal activities under the partial information blacklist and how to design coins' quality updating rule rationally have become open issues in mixing coins. The updating rule of coins' quality in mixing is addressed since illegal criminals may carry out illegal activities, for example, money laundering. ImpSuic, an improved suicide strategy, is proposed as a new quality updating rule. The intuition is: all coins of the one who has the highest bad coins according to the blacklist, are recorded as bad coins. On the other hand, the coins' quality of others remain unchanged. Besides, linear programming is introduced into ImpSuic strategy to predict the maximum utility after mixing coins, which facilitates users to make reasonable decisions before mixing coins. Simulation results show that the quality updating rule in ImpSuic strategy can preserve users' privacy and antimoney launder.

Emotion recognition has become an important component of human–computer interaction systems. Research on emotion recognition based on electroencephalogram (EEG) signals are mostly conducted by the analysis of all channels' EEG signals. Although some progresses are achieved, there are still several challenges such as high dimensions, correlation between different features and feature redundancy in the realistic experimental process. These challenges have hindered the applications of emotion recognition to portable human–computer interaction systems (or devices). This paper explores how to find out the most effective EEG features and channels for emotion recognition so as to only collect data as less as possible. First, discriminative features of EEG signals from different dimensionalities are extracted for emotion classification, including the first difference, multiscale permutation entropy, Higuchi fractal dimension, and discrete wavelet transform. Second, relief algorithm and floating generalized sequential backward selection algorithm are integrated as a novel channel selection method. Then, support vector machine is employed to classify the emotions for verifying the performance of the channel selection method and extracted features. At last, experimental results demonstrate that the optimal channel set, which are mostly located at the frontal, has extremely high similarity on the self-collected data set and the public data set and the average classification accuracy is achieved up to 91.31% with the selected 10-channel EEG signals. The findings are valuable for the practical EEG-based emotion recognition systems.

In recent years, automatic emotion recognition renders human–computer interaction systems intelligent and friendly. Emotion recognition based on electroencephalogram (EEG) has received widespread attention and many research results have emerged, but how to establish an integrated temporal and spatial feature fusion and classification method with improved convolutional neural networks (CNNs) and how to utilize the spatial information of different electrode channels to improve the accuracy of emotion recognition in the deep learning are two important challenges. This paper proposes an emotion recognition method based on three-dimensional (3D) feature maps and CNNs. First, EEG data are calibrated with 3 s baseline data and divided into segments with 6 s time window, and then the wavelet energy ratio, wavelet entropy of five rhythms, and approximate entropy are extracted from each segment. Second, the extracted features are arranged according to EEG channel mapping positions, and then each segment is converted into a 3D feature map, which is used to simulate the relative position of electrode channels on the scalp and provides spatial information for emotion recognition. Finally, a CNN framework is designed to learn local connections among electrode channels from 3D feature maps and to improve the accuracy of emotion recognition. The experiments on data set for emotion analysis using physiological signals data set were conducted and the average classification accuracy of 93.61% and 94.04% for valence and arousal was attained in subject-dependent experiments while 83.83% and 84.53% in subject-independent experiments. The experimental results demonstrate that the proposed method has better classification accuracy than the state-of-the-art methods.

With the rapid development of online courses, the requirements of personalized course recommendation have been increasing. The traditional collaborative filtering algorithm confronts with the challenge of cold start, which is difficult to settle on online course recommendation effectively. In this paper, we propose a novel click through rate (CTR) model for personalized online course recommendation, with discriminative user features, item features and cross features. The feature representation ability of the CTR model is improved and the serious challenge of cold start is alleviated. Furthermore, transfer learning is introduced to deal with the problem of insufficient data in models training. More specially, eye tracking technology is applied to capture the users' cognitive styles, which are visualized with the heat map and fixation point trajectory. Finally, the recommendation interface sent to the learners, according to the user's cognitive style. The experiments show that the novel CTR model improves the performance of the personalized online course recommendation.

Perpetrators leverage the untraceable feature to conduct illegal behaviors leading security issues with respect to mixing coins. Generally, bad coins are blocked based on a common blacklist. However, the blacklist may not be updated in time, which results in that bad coins escape the blocking. Consequently, perpetrators can still conduct illicit behaviors such as money laundering. In this paper, we apply game theory under imperfect information to study how coins' quality restrain these illicit behaviors under the incomplete scenario. More specifically, we propose a strategy for participants to submit deposits if they hope to mix coins with others even if they are not in blacklist at this time. The deposits will not be refunded when participants are included in the blacklist after mixing. Therefore, no participants have incentives to mix with bad coins. At the last part of this paper, we also simulate the incomes for participants, which indicates that deposits strategy is effective to prevent illicit behaviors.

The vulnerabilities in cryptographic currencies facilitate the adversarial attacks. Therefore, the attackers have incentives to increase their rewards by strategic behaviors. Block withholding attacks (BWH) are such behaviors that attackers withhold blocks in the target pools to subvert the blockchain ecosystem. Furthermore, BWH attacks may dwarf the countermeasures by combining with selfish mining attacks or other strategic behaviors, for example, fork after withholding (FAW) attacks and power adaptive withholding (PAW) attacks. That is, the attackers may be intelligent enough such that they can dynamically gear their behaviors to optimal attacking strategies. In this paper, we propose mixed-BWH attacks with respect to intelligent attackers, who leverage reinforcement learning to pin down optimal strategic behaviors to maximize their rewards. More specifically, the intelligent attackers strategically toggle among BWH, FAW, and PAW attacks. Their main target is to fine-tune the optimal behaviors, which incur maximal rewards. The attackers pinpoint the optimal attacking actions with reinforcement learning, which is formalized into a Markov decision process. The simulation results show that the rewards of the mixed strategy are much higher than that of honest strategy for the attackers. Therefore, the attackers have enough incentives to adopt the mixed strategy.

Blockchain is a “decentralized” system, where the security heavily depends on that of the consensus protocols. For instance, attackers gain illegal revenues by leveraging the vulnerabilities of the consensus protocols. Such attacks consist of selfish mining (SM1), optimal selfish mining ( ϵ-optimal), bribery selfish mining (BSM), and so forth. In existing works, the attacks only consider the circumstances, where part of miners are rational. However, miners are hardly nonrational in the blockchain system since they hope to maximize their revenues. Furthermore, attackers prefer intelligent tools to increase their power for more additional revenues. Therefore, new models are urgently needed to formulate the scenarios, where attackers are purely rational and intelligent. In this paper, we propose a new BSM model, where all miners are rational. Moreover, rational attackers are intelligent such that they optimize their strategies by utilizing reinforcement learning to boost their revenues. More specifically, we propose a new selfish mining algorithm: intelligent bribery selfish mining (IPBSM), where attackers choose optimal strategies resorting to reinforcement learning when they interact with the external environment. The external environment can be further modeled as a Markov decision process to facilitate the construction of reinforcement learning. The simulation results manifest that IPBSM, compared with SM1 and ϵ-optimal, has lower power thresholds and higher revenues. Therefore, IPBSM is a threat no to be neglected to the blockchain system.