Nowadays, the issue of information overload is gradually gaining exposure in the Internet of Things (IoT), calling for more research on recommender system in advance for industrial IoT scenarios. ...With the ever-increasing prevalence of various social networks, social recommendations (SoR) will certainly become an integral application that provides more feasibly personalized information service for future IoT users. However, almost all of the existing research managed to explore and quantify correlations between user preferences and social relationships, while neglecting the correlations among item features which could further influence the topologies of some social groups. To tackle with this challenge, in this article, a deep graph neural network-based social recommendation framework (GNN-SoR) is proposed for future IoTs. First, user and item feature spaces are abstracted as two graph networks and respectively encoded via the graph neural network method. Next, two encoded spaces are embedded into two latent factors of matrix factorization to complete missing rating values in a user-item rating matrix. Finally, a large amount of experiments are conducted on three real-world data sets to verify the efficiency and stability of the proposed GNN-SoR.
Nowadays, online spamming has already been a remarkable threat to contents security of Internet of Things. Due to constant technical progress, online spamming activities have been more and more ...concealed. This brings much fuzziness to spammer detection scenarios, yielding the issue of fuzzy detection of spammers. Although existing detection techniques for spammers utilized idea of deep learning, they still ignore to release power of label spaces. As real nature about a user may be usually fuzzy, but the label annotated for a user is always certain. To remedy such gap, this article proposes a label smoothing-based fuzzy detection method for spammers (Fuz-Spam). First of all, deep representation is still utilized to deeply fuse features, which acts as the foundation of neural computing. On this basis, generative adversarial learning is introduced to transform previous label spaces into distributed forms. In addition, two groups of experiments are carried out on two real-world datasets for evaluation. The results demonstrate that the Fuz-Spam improves identification efficiency about 10% to 20% than previous ones, and that the Fuz-Spam is endowed with proper stability.
The fabrication of mechanically robust polymeric materials capable of self-healing and recycling remains challenging because the mobility of polymer chains in such polymers is very limited. In this ...work, mechanically robust supramolecular thermosets capable of healing physical damages and recycling under mild conditions are fabricated by trimerization of bi-(ortho-aminomethyl-phenylboronic acid)- and tri-(ortho-aminomethyl-phenylboronic acid)-terminated poly(propylene glycol) oligomers (denoted as Bi-PBA-PPG and Tri-PBA-PPG, respectively). The resultant supramolecular thermosets are cross-linked by dynamic covalent bonds of nitrogen-coordinated boroxines. The mechanical properties of the supramolecular thermosets can be systematically tailored by varying the ratios between Tri-PBA-PPG and Bi-PBA-PPG, which changes the cross-linking density of nitrogen-coordinated boroxines and the topology of the supramolecular thermosets. The mechanically strongest supramolecular thermosets with a molar ratio of Tri-PBA-PPG to Bi-PBA-PPG being 1:2 have a glass transition temperature of ∼36 °C, a tensile strength of ∼31.96 MPa, and a Young’s modulus of ∼298.5 MPa. The high reversibility of nitrogen-coordinated boroxines and the flexibility of poly(propylene glycol) chains enable the supramolecular thermosets with the strongest mechanical strength to be highly efficiently healed at 55 °C and recycled under a pressure of 4 MPa at 60 °C to regain their original mechanical strength and integrity.
Nowadays, rumor spreading has gradually evolved into a kind of organized behaviors, accompanied with strong uncertainty and fuzziness. However, existing fuzzy detection techniques for rumors focused ...their attention on supervised scenarios that require expert samples with labels for training. Thus, they are not able to well handle the unsupervised scenarios where labels are unavailable. To bridge such gap, this article proposed a fuzzy detection system for rumors through explainable adaptive learning. Specifically, its core is a graph embedding-based generative adversarial network (Graph-GAN) model. First of all, it constructs fine-grained feature spaces via graph-level encoding. Furthermore, it introduces continuous adversarial training between a generator and a discriminator for unsupervised decoding. The two-stage scheme not only solves the fuzzy rumor detection under unsupervised scenarios, but also improves robustness of the unsupervised training. Empirically, a set of experiments are carried out based on three real-world datasets. Compared with seven benchmark methods in terms of four metrics, the results of the Graph-GAN reveal a proper performance, which averagely exceeds baselines by 5-10%.
The running-in of cylinder liner-piston rings (CLPRs) is the most important process that must be performed before a marine diesel engine can be operated. The quality of running-in directly affects ...the reliability of a CLPR. The surface texture of a CLPR has been proven to significantly affect its lubrication performance. In this study, the tribological behavior of a CLPR during running-in is investigated. Three types of surface textures are generated on the CLPR via laser processing: dimple texture on piston rings, groove texture on cylinder liners, and co-texture on both sides. Subsequently, a series of tests are performed on a slice tester. A load of 300 N (1.64 MPa) is applied, and two speeds (50 and 100 rpm) are adopted. The CLPR running-in quality is characterized based on three parameters, i.e., the friction coefficient, contact resistance, and wear topography. Experimental results show that, compared with a non-textured surface, the three types of surface textures mentioned above improved the friction performance during running-in. The lubricant supply capacity of the dimple texture on the piston ring, as a mobile oil reservoir, is stronger than that of the groove texture on the cylinder liner serving as a static oil reservoir. By contrast, the wear resistance of the dimple texture, as a movable debris trap on the piston ring, is weaker than that of the groove texture on the cylinder liner, which serves as a static debris trap. It is demonstrated that the co-texture combines the advantages of dimples and groove textures. Compared with non-textured surfaces, the friction coefficient decreased the most at 100 rpm (44.5%), and the contact resistance improved the most at 50 rpm (352.9%). The coupling effect provides the surface with improved running-in quality by optimizing the tribological performance, particularly at the dead center. This study provides guidance for the tribological design and manufacturing of CLPR in marine diesel engines.
New structures with richer electromagnetic properties are in high demand for developing novel microwave and optic devices aimed at realizing fast light-based information transfer and information ...processing. Here we show theoretically that a topological photonic state exists in a hexagonal LC circuit with short-range textures in the inductance, which is induced by a band inversion between p- and d-like electromagnetic modes carrying orbital angular momentum, and realize this state experimentally in planar microstrip arrays. Measuring both amplitude and phase of the out-of-plane electric field accurately using microwave near-field techniques, we demonstrate directly that topological interfacial electromagnetic waves launched by a linearly polarized dipole source propagate in opposite directions according to the sign of the orbital angular momentum. The open planar structure adopted in the present approach leaves much room for including other elements useful for advanced information processing, such as electric/mechanical resonators, superconducting Josephson junctions and SQUIDs.
Motivated by unique topological semimetals in condensed matter physics, we propose an effective Hamiltonian with four degrees of freedom to describe evolutions of photonic double Weyl nodal line ...semimetals in one-dimensional hyper-crystals, which supports the energy bands translating or rotating independently in the form of Weyl quasiparticles. Especially, owing to the unit cells without inversion symmetry, a pair of reflection-phase singularities carrying opposite topological charges emerge near each nodal line, and result in a unique bilateral drumhead surface state. After reducing radiation leakages and absorption losses, these two singularities gather together gradually, and form a quasi-bound state in the continuum (quasi-BIC) ring at the nodal line ultimately. Our work not only reports the first realization of controllable photonics Weyl nodal line semimetals, establishes a bridge between two independent topological concepts-BICs and Weyl semimetals, but also heralds new possibilities for unconventional device applications, such as dual-mode schemes for highly sensitive sensing and switching.
Finite-element analysis (FEA) combined with experimental observation was conducted on preheated Cu particles deposited on Cu substrate to clarify the deposition behavior of thermally softened ...particles in cold spraying. An explicit FEA code, ABAQUS, was used to predict the deformation features of the thermally softened particles. The experiment was performed by a home-made cold-spray system with a powder preheating device. Considering the possible serious oxidation of the cold-sprayed particles under high-temperature conditions, the preheating temperature was limited to 300°C for each test. Based on the numerical and experimental results, a new concept called the thermal softening zone within which thermal softening occurs is proposed in the present work. It is found that thermally softened particles deform more intensively compared to non-preheated particles, and a more prominent metal jet can be achieved at the rim of the deformed particles with higher initial temperature. Moreover, the results also reveal that increasing the particle preheating temperature can stimulate the occurrence of thermal softening. For non-preheating or low-temperature preheating particles, thermal softening mainly occurs at the interfacial region. If the preheating temperature is sufficiently high, the whole particle can experience thermal softening. In addition, it is also found that preheated particles are more likely to deposit on the substrate surface than non-preheated particles. In addition, particle preheating is also found to facilitate the coating formation process, enabling the coating to be very thick. The coating microhardness decreases with increasing particle preheating temperature due to the elimination of work hardening by thermal softening.