ABLE: Meta-Path Prediction in Heterogeneous Information Networks

Given a heterogeneous information network (HIN) H, a head node h , a meta-path P, and a tail node t , the meta-path prediction aims at predicting whether h can be linked to t by an instance of P. Most existing solutions either require predefined meta-paths, which limits their scalability to schema-rich HINs and long meta-paths, or do not aim at predicting the existence of an instance of P. To address these issues, in this article, we propose a novel prediction model, called ABLE, by exploiting the A ttention mechanism and B i L STM for E mbedding. Particularly, we present a concatenation node embedding method by considering the node types and a dynamic meta-path embedding method that carefully considers the importance and positions of edge types in the meta-paths by the Attention mechanism and BiLSTM model, respectively. A triplet embedding is then derived to complete the prediction. We conduct extensive experiments on four real datasets. The empirical results show that ABLE outperforms the state-of-the-art methods by up to 20% and 22% of improvement of AUC and AP scores, respectively.

SPEX: A Generic Framework for Enhancing Neural Social Recommendation

Social Recommender Systems (SRS) have attracted considerable attention since its accompanying service, social networks, helps increase user satisfaction and provides auxiliary information to improve recommendations. However, most existing SRS focus on social influence and ignore another essential social phenomenon, i.e., social homophily. Social homophily, which is the premise of social influence, indicates that people tend to build social relations with similar people and form influence propagation paths. In this article, we propose a generic framework Social PathExplorer (SPEX) to enhance neural SRS. SPEX treats the neural recommendation model as a black box and improves the quality of recommendations by modeling the social recommendation task, the formation of social homophily, and their mutual effect in the manner of multi-task learning. We design a Graph Neural Network based component for influence propagation path prediction to help SPEX capture the rich information conveyed by the formation of social homophily. We further propose an uncertainty based task balancing method to set appropriate task weights for the recommendation task and the path prediction task during the joint optimization. Extensive experiments have validated that SPEX can be easily plugged into various state-of-the-art neural recommendation models and help improve their performance. The source code of our work is available at: https://github.com/XMUDM/SPEX.

LSTM-Based Path Prediction for Effective Sensor Filtering in Sensor Registry System

The Internet of Things (IoT) is expected to provide intelligent services by receiving heterogeneous data from ambient sensors. A mobile device employs a sensor registry system (SRS) to present metadata from ambient sensors, then connects directly for meaningful data. The SRS should provide metadata for sensors that may be successfully connected. This process is location-based and is also known as sensor filtering. In reality, GPS sometimes shows the wrong position and thus leads to a failed connection. We propose a dual collaboration strategy that simultaneously collects GPS readings and predictions from historical trajectories to improve the probability of successful requests between mobile devices and ambient sensors. We also update the evaluation approach of sensor filtering in SRS by introducing a Monte Carlo-based simulation flow to measure the service provision rate. The empirical study shows that the LSTM-based path prediction can compensate for the loss of location abnormalities and is an effective sensor filtering model.

Optimal Security Protection Selection Strategy Based on Markov Model Attack Graph

Intrusion intent and path prediction are important for security administrators to gain insight into the possible threat behavior of attackers. Existing research has mainly focused on path prediction in ideal attack scenarios, yet the ideal attack path is not always the real path taken by an intruder. In order to accurately and comprehensively predict the path information of network intrusion, a multi-step attack path prediction method based on absorbing Markov chains is proposed. Firstly, the node state transfer probability normalization algorithm is designed by using the nil posteriority and absorption of state transfer in absorbing Markov chain, and it is proved that the complete attack graph can correspond to absorbing Markov chain, and the economic indexes of protection cost and attack benefit and the index quantification method are constructed, and the optimal security protection policy selection algorithm based on particle swarm algorithm is proposed, and finally the experimental verification of the model in protection Finally, we experimentally verify the feasibility and effectiveness of the model in protection policy decision-making, which can effectively reduce network security risks and provide more security protection guidance for timely response to network attack threats.

Fine-Tuning Word Embeddings for Hierarchical Representation of Data Using a Corpus and a Knowledge Base for Various Machine Learning Applications

Word embedding models have recently shown some capability to encode hierarchical information that exists in textual data. However, such models do not explicitly encode the hierarchical structure that exists among words. In this work, we propose a method to learn hierarchical word embeddings (HWEs) in a specific order to encode the hierarchical information of a knowledge base (KB) in a vector space. To learn the word embeddings, our proposed method considers not only the hypernym relations that exist between words in a KB but also contextual information in a text corpus. The experimental results on various applications, such as supervised and unsupervised hypernymy detection, graded lexical entailment prediction, hierarchical path prediction, and word reconstruction tasks, show the ability of the proposed method to encode the hierarchy. Moreover, the proposed method outperforms previously proposed methods for learning nonspecialised, hypernym-specific, and hierarchical word embeddings on multiple benchmarks.

Generating Bird’s Eye View from Egocentric RGB Videos

In this paper, we present a method for generating bird’s eye video from egocentric RGB videos. Working with egocentric views is tricky since such the view is highly warped and prone to occlusions. On the other hand, a bird’s eye view has a consistent scaling in at least the two dimensions it shows. Moreover, most of the state-of-the-art systems for tasks such as path prediction are built for bird’s eye views of the subjects. We present a deep learning-based approach that transfers the egocentric RGB images captured from a dashcam of a car to bird’s eye view. This is a task of view translation, and we perform two experiments. The first one uses an image-to-image translation method, and the other uses a video-to-video translation. We compare the results of our work with homographic transformation, and our SSIM values are better by a margin of 77% and 14.4%, and the RMSE errors are lower by 40% and 14.6% for image-to-image translation and video-to-video translation, respectively. We also visually show the efficacy and limitations of each method with helpful insights for future research. Compared to previous works that use homography and LIDAR for 3D point clouds, our work is more generalizable and does not require any expensive equipment.

A Multiphase Dynamic Deployment Mechanism of Virtualized Honeypots Based on Intelligent Attack Path Prediction

As an important deception defense method, a honeypot can be used to enhance the network’s active defense capability effectively. However, the existing rigid deployment method makes it difficult to deal with the uncertain strategic attack behaviors of the attackers. To solve such a problem, we propose a multiphase dynamic deployment mechanism of virtualized honeypots (MD2VH) based on the intelligent attack path prediction method. MD2VH depicts the attack and defense characteristics of both attackers and defenders through the Bayesian state attack graph, establishes a multiphase dynamic deployment optimization model of the virtualized honeypots based on the extended Markov’s decision-making process, and generates the deployment strategies dynamically by combining the online and offline reinforcement learning methods. Besides, we also implement a prototype system based on software-defined network and virtualization container, so as to evaluate the effectiveness of MD2VH. Experiments results show that the capture rate of MD2VH is maintained at about 90% in the case of both simple topology and complex topology. Compared with the simple intelligent deployment strategy, such a metric is increased by 20% to 60%, and the result is more stable under different types of the attacker’s strategy.