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.

Embedding Heterogeneous Information Network in Hyperbolic Spaces

Heterogeneous information network (HIN) embedding, aiming to project HIN into a low-dimensional space, has attracted considerable research attention. Most of the existing HIN embedding methods focus on preserving the inherent network structure and semantic correlations in Euclidean spaces. However, one fundamental problem is whether the Euclidean spaces are the intrinsic spaces of HIN? Recent researches find the complex network with hyperbolic geometry can naturally reflect some properties, e.g., hierarchical and power-law structure. In this article, we make an effort toward embedding HIN in hyperbolic spaces. We analyze the structures of three HINs and discover some properties, e.g., the power-law distribution, also exist in HINs. Therefore, we propose a novel HIN embedding model HHNE. Specifically, to capture the structure and semantic relations between nodes, HHNE employs the meta-path guided random walk to sample the sequences for each node. Then HHNE exploits the hyperbolic distance as the proximity measurement. We also derive an effective optimization strategy to update the hyperbolic embeddings iteratively. Since HHNE optimizes different relations in a single space, we further propose the extended model HHNE++. HHNE++ models different relations in different spaces, which enables it to learn complex interactions in HINs. The optimization strategy of HHNE++ is also derived to update the parameters of HHNE++ in a principle manner. The experimental results demonstrate the effectiveness of our proposed models.

Architectural Holarchy of the Next Generation Manufacturing System

Nowadays, the global energy network can generate and transmit, between any two points belonging to it, high quantity of energy. During recent years, a global information network, able to process, store, and transmit huge amounts of information, has been developed as well. These networks entirely cover the industrial space, already giving the opportunity to make permanently available, in any of its points, at any time, as much as needed, both energy and information. On the other hand, the mass customization trend has led to the pronounced increase of “manufacturing to order” (MTO) production, taking place in a higher and higher number of small & medium enterprises. At this level, a given manufacturing system cannot be quickly and appropriately configured to a given product, due to production high variability in range. As consequence, the manufacturing system is, quite always, more or less unadjusted to the manufactured product, its performance being significantly affected. Starting from here, the challenge is to make a conceptual rebuilding of the manufacturing system, aiming to increase its degree of appropriateness to products, by taking advantage from the opportunities brought by the existence of global energy & information networks. This paper approach is to see the next generation manufacturing system as holonic modular cyber-physical system. System architecture permanently accords to the manufactured product requirements. The function, procedure, topology and holarchy model of the system are presented. The main features of the system are also revealed.

Practice-driven research for statewide scale up: Implementation outcomes of the California Autism Professional Training and Information Network

Scaling up the use of evidence-based practice (EBP) for autism across service sectors and regions has presented a considerable challenge indicating a clear need for continued development. The California Autism Professional Training and Information Network (CAPTAIN) integrates implementation drivers into specific procedures and methodology as an implementation strategy to support statewide scale up. The current study was designed to evaluate the impact of CAPTAIN on provider-level outcomes including attitude toward, and knowledge, fidelity and use of autism EBPs, and overall classroom quality. Overall, results indicated variability across measures, with some significant differences between CAPTAIN-trained and non-CAPTAIN-trained providers. CAPTAIN-trained providers reported more openness to EBP. Significantly more CAPTAIN-trained direct service providers reported collecting fidelity of implementation data (χ2(2, N = 1515) = 10.95, p = 0.004), collecting student data (χ2(2, N = 1509) = 14.19, p = 0.001), and reported using their primary EBP with “most or all students” (χ2(2, N = 1514) = 11.41, p = 0.003) than providers not trained by CAPTAIN. In summary, these preliminary findings show promise for the efficacy of the CAPTAIN model to increase dissemination and implementation of EBP at the classroom level. Lay abstract Supporting use of evidence-based practice in public service programs for autistic individuals is critical. The California Autism Professional Training and Information Network (CAPTAIN) brings together best practices from intervention and implementation research to support scale up of autism services. The current study was designed to evaluate the impact of CAPTAIN on provider-level outcomes including attitude toward, knowledge, fidelity, and use of autism EBPs and overall classroom quality. Overall, results indicated variability across measures, with some significant differences between CAPTAIN-trained and non-CAPTAIN-trained providers. These preliminary findings show promise for the efficacy of the CAPTAIN model to increase dissemination and implementation of EBP at the classroom level.

Network Embedding-Based Approach for Detecting Collusive Spamming Groups on E-Commerce Platforms

Information security is one of the key issues in e-commerce Internet of Things (IoT) platform research. The collusive spamming groups on e-commerce platforms can write a large number of fake reviews over a period of time for the evaluated products, which seriously affect the purchase decision behaviors of consumers and destroy the fair competition environment among merchants. To address this problem, we propose a network embedding based approach to detect collusive spamming groups. First, we use the idea of a meta-graph to construct a heterogeneous information network based on the user review dataset. Second, we exploit the modified DeepWalk algorithm to learn the low-dimensional vector representations of user nodes in the heterogeneous information network and employ the clustering methods to obtain candidate spamming groups. Finally, we leverage an indicator weighting strategy to calculate the spamming score of each candidate group, and the top-k groups with high spamming scores are considered to be the collusive spamming groups. The experimental results on two real-world review datasets show that the overall detection performance of the proposed approach is much better than that of baseline methods.