Aim: The study aims to explore the synergistic effects between different herbal pairs and the original traditional Chinese medicine prescriptions by developing a model for analyzing compatibility correlation and synergy in prescriptions. Method: We have applied TCMNPAS, a comprehensive analysis platform that integrates network formularology and network pharmacology. This platform is designed to investigate in-depth the compatibility characteristics of TCM formulas and their potential molecular mechanisms. Liuwei Dihuang prescription(LDP), formulated following the principle of "Three Tonics and Three Purgatives," was used as an example. An analysis model built via association network was applied to evaluate the relationship and synergistic impacts between the original prescription and three distinct drug pairs related to specific diseases. Results: Node similarities measured by the nearest-neighbor expansion network and shortest path expansion network between LDP and kidney yin deficiency disease network were 0.873 and 0.685, respectively. The average random walk correlation score between LDP and diseases showed a significant difference, suggesting a strong correlation with diseases. Among these, the herbal pair of Corni FructusMoutan Cortex(SZY-MDP) demonstrated node similarities of 0.742 and 0.433 with kidney yin deficiency on the nearest-neighbor expansion network and shortest path expansion network, respectively. Also, a significant difference was noted in the average random walk correlation score of this pair with diseases, signifying that this herb pair is most closely related to the original prescription and diseases. Simultaneously, within LDP, the SZY-MDP herbal pair showed the closest association with the original prescription concerning shared targets, active compounds, and disease targets. Notably, the most prominent KEGG signaling pathway for LDP intervention in hyperthyroidism was the regulation of adipocyte lipolysis, indicating its regulatory effect on body energy metabolism and lipid metabolism. Conclusion: Based on the analysis model focused on compatibility correlation and synergy, the "tonifying and purgative" drugs in LDP exhibit specific "correlation-synergy" effects on kidney yin deficiency.

Abstract With the rapid development of social networks, academic social networks have attracted increasing attention. In particular, providing personalized recommendations for learners considering data sparseness and cold-start scenarios is a challenging task. An important research topic is to accurately discover potential friends of learners to build implicit learning groups and obtain personalized collaborative recommendations of similar learners according to the learning content. This paper proposes a personalized explainable learner implicit friend recommendation method (PELIRM). Methodologically, PELIRM utilizes the learner's multidimensional interaction behavior in social networks to calculate the degrees of trust between learners and applies the three-degree influence theory to mine the implicit friends of learners. The similarity of research interests between learners is calculated by cosine and term frequency–inverse document frequency. To solve the recommendation problem for cold-start learners, the learner's common check-in IP is used to obtain the learner's location information. Finally, the degree of trust, similarity of research interests, and geographic distance between learners are combined as ranking indicators to recommend potential friends for learners and give multiple interpretations of the recommendation results. By verifying and evaluating the proposed method on real data from Scholar.com, the experimental results show that the proposed method is reliable and effective in terms of personalized recommendation and explainability.

Purpose: We explored the mechanism effects of “The Same Treatment for Different metabolic Diseases” concept in traditional Chinese medicine (TCM).Methods: Network pharmacology, molecular docking technology and KATZ score were employed to predict the targets and signaling pathways affected by Shugan Quzhi Capsules (SGQZ), which are used to treat obesity, hyperlipidemia, and non-alcoholic fatty liver disease.Results: A total of 492 active ingredients in SGQZ were found to act on 666 potential targets. They were involved in GO biological processes related to lipid metabolism, oxidative stress, anti-inflammatory effects, and calcium ion channel effects; the three most significantly enriched KEGG pathways were endocrine resistance, advanced glycation end-product (AGE)/receptor for AGEs signaling pathway in diabetic complications, and peroxisome proliferator-activated receptor (PPAR) signaling. The correlation between SGQZ and each metabolic disease was ranked as NAFLD>hyperlipidemia>obesity.Conclusions: Our results clarify the mechanisms of action of SGQZ on three conditions, demonstrating the utility of “The Same Treatment for metabolic Different Diseases” from both the target and pathway perspectives. Among them, leptin, apolipoprotein B, and PPARG and PPAR-/AMP-activated protein kinase-related signal pathways are the key targets and pathways for SGQZ. In TCM terminology, the spleen and liver meridians are the key routes by which quercetin and apigenin in SGQZ exert their effects.

Despite recent progress in depth sensing and 3D reconstruction, mirror surfaces are a significant source of errors. To address this problem, we create the Mirror3D dataset: a 3D mirror plane dataset based on three RGBD datasets (Matterpot3D, NYUv2 and ScanNet) containing 7,011 mirror instance masks and 3D planes. We then develop Mirror3DNet: a module that refines raw sensor depth or estimated depth to correct errors on mirror surfaces. Our key idea is to estimate the 3D mirror plane based on RGB input and surrounding depth context, and use this estimate to directly regress mirror surface depth. Our experiments show that Mirror3DNet significantly mitigates errors from a variety of input depth data, including raw sensor depth and depth estimation or completion methods.

This paper presents a multi-agent (MAS) solution to the energy management in microgrid. An introduction and analysis of the system, including the structure of microgrid, characteristics of MAS and the potential game theory, are presented firstly. Then, micro models of agents for each component are proposed. The MAS is shown of having attributes of distribution, heterogeneity, autonomous, dynamism and openness. And then, the overall model of MAS and the process of optimization are presented in the next part. Potential game theory is introduced which applying as the distributed optimization method embedded in each micro agent model. Finally, the MAS for microgrid is formulated and implemented by the JADE platform. A scenario case study on the MAS and its testing process are included in the paper.

A control scheme for dynamic tracking of mobile robots is presented, which integrates a velocity controller based on backstepping techniques and a torque controller based on improved RBF neural networks. Because the torque control strategy derived from sliding modes depends on the dynamics of mobile robots, the robustness of the system cannot be guaranteed due to the uncertainties of robot dynamics. In order to decrease the impact of the uncertainties and improve the robustness of the system, improved RBF neural networks are designed online to model the dynamics of mobile robot. Thus the torque controller based on sliding mode is composed of a neural network controller and a robust compensator. Simulations demonstrate the efficacy of the proposed system for robust tracking of mobile robots