Abstract: Disclosed is a method for recommending a Web service based on a graph attention convolutional network and graph pooling. The method includes: constructing a service network graph based on a dual-tower model, and differentiating a focus service; providing a dual-level aware self-attention mechanism in a graph convolution process, controlling an amount of information aggregated from a source node to a target node in a cluster-aware self-attention mechanism, and controlling an amount of information received by the target node from the source node in a query-aware self-attention mechanism; then, further extracting service information according to a graph pooling method taking a heterogeneous node coarsening strategy as a core; and splicing a dynamic representation, a graph-level representation, and a target service of the focus service, learning a composite embedding representation, and predicting a probability of interaction between the target service and a user at a next moment in a prediction layer.

Abstract: Disclosed is a cloud-native application programming interface (API) recommendation method fusing data augmentation and contrastive learning. Service information is included on the basis of a service information double-graph structure, and a mutual attention mechanism is designed to compute an importance degree of each layer of information. A data optimization method for sequence information based on functional similarity and a computation method for similarity between services based on two parts of information are provided; on this basis, data of a service invocation sequence is augmented with the idea of contrastive learning to form an augmented sequence pair; a computational contrastive loss function is combined with a pair-wise recommendation loss function to optimize an overall model, thereby improving the effect of a service recommendation model; and according to a feature embedding representation result of a service, pair-wise recommendation scores are computed to complete service recommendation.

Abstract: The disclosed system may include a display; a lens; and a diffractive optical element, where the diffractive optical element includes at least one Pancharatnam-Berry phase grating and is configured to increase a fill factor of the display when the lens is used to magnify the display. Various other apparatuses, systems, and methods are also disclosed.

Abstract: A method for active control of variable speed limits on highways includes: (a) a mixed traffic flow scenario considering driver individuality is constructed, and scenario parameters are set; (b) traffic state data for a merging road scenario is acquired; (c) an agent is trained using the traffic state data as state variables to obtain a speed limit control optimization model; diverse training scenarios are generated by Poisson distribution during training, an optimal speed limit value and an appropriate speed control location are selected within each control cycle T, and a moving bottleneck is generated using a moving bottleneck variable speed limit constructed based on connected and autonomous vehicles. A system for active control of variable speed limits on highways is further provided.

Abstract: The present application provides a smart floor cleaning device and a control method thereof. The smart floor cleaning device includes a body, a chassis, a processor, a negative pressure source, and a sewage tank on the body. The negative pressure source is connected to the sewage tank and applies a negative pressure to draw sewage during cleaning into the sewage tank. The smart floor cleaning device further includes an angle detection device on the body and/or the chassis. The processor is electrically connected to the angle detection device and obtains a rotation angle of the body around the first axis and/or the second axis based on a detection signal from the angle detection device. The first axis and second axis are perpendicular to each other. The processor is electrically connected to the negative pressure source and controls the output power of the negative pressure source based on the rotation angle.

Abstract: A denoising method based on a multiscale distribution score for a point cloud includes: constructing a two-layer network model based on multiscale perturbation and point cloud distribution, where the two-layer network model includes a feature extraction module for extracting a feature of the point cloud and a displacement prediction module for predicting a displacement of a noise point; constructing a point cloud noise model for improving a denoising effect and retaining a sharp feature and avoiding reducing quality of point cloud data; extracting a global feature h by inputting the point cloud data into the feature extraction module; iteratively learning the displacement of the noise point by the displacement prediction module according to a feature obtained by the feature extraction unit; and defining a loss function of network training, and completing convergence under the condition that the loss function reaches a set threshold or a maximum number of iterations.

Abstract: The present application discloses a self-calibration elevator door lock contact making device, which comprises a stationary base plate, a movable base plate, contacts, adjusting components and a controller, wherein the stationary base plate and the movable base plate are oppositely arranged and are respectively provided with contacts, and the two contacts can move relatively laterally or longitudinally through the adjusting components respectively; and the controller is electrically connected with two groups of adjusting components and provided with a contact detection loop capable of monitoring the contact state of the two contacts. The automatic real-time monitoring and automatic calibration of the door lock can be realized, and the door lock can be locked smoothly when the landing door and the car door are closed, thereby reducing the delay caused by inaccurate door operation and the cost of manual maintenance and adjustment, and improving the safety and reliability of the elevator.

Abstract: The present application discloses a steel belt guide wheel assembly comprising guide wheels and bearings, wherein the guide wheels are made of nylon, the left and right ends of the wheel surface are respectively provided with a left annular convex wall and a right annular convex wall; a belt groove is formed between the left annular convex wall and the right annular convex wall, the bearings are coaxially installed in a wheel body channel of the guide wheel, and the guide wheels can be detachably connected in pairs by a connecting mechanism or glued together by glue; the assembly has good wear resistance, low noise, light weight and high strength, which prolongs the service life, reduces the manufacturing cost and maintenance cost, helps to reduce the load of elevator system, reduces the load of driving parts, improves the riding comfort of passengers, and can be freely combined as required.

Abstract: Disclosed is a secondary recommendation method based on a service complementary relation learning model for a RESTful service.

Abstract: A test method for a variable speed limit (VSL) strategy in a mixed traffic flow considering driver compliance is provided. A road scenario is built, and multiple mixed traffic flows formed by human-driven vehicles (HDVs) and connected and automated vehicles (CAVs) are set, which vary in CAV penetration rate. The VSL strategy is established, and a speed limit of an upstream of the bottleneck area is calculated based on the VSL strategy. A HDV compliance is established, and in combination with the speed limit, the HDV drivers are classified, and a proportion of the HDV drivers in individual types in the road scenario is determined. Based on a traffic flow scenario in which the HDV drivers and CAV drivers are located, indicator data reflecting traffic capacity is calculated. A test system for performing the test method is further provided.

Abstract: Systems and methods for image data management are provided. The methods may include by running a first process, storing raw data collected in a scan of a subject into first storage space of the at least one storage device, and generating first information relating to the raw data. The methods may include by running a second process, obtaining the first information from the first process and storing the first information into second storage space of the at least one storage device. The methods may further include reconstructing the raw data into a reconstruction image of the subject based on the first information.

Abstract: Disclosed is a cloud-native application programming interface (API) recommendation method fusing data augmentation and contrastive learning. Service information is included on the basis of a service information double-graph structure, and a mutual attention mechanism is designed to compute an importance degree of each layer of information. A data optimization method for sequence information based on functional similarity and a computation method for similarity between services based on two parts of information are provided; on this basis, data of a service invocation sequence is augmented with the idea of contrastive learning to form an augmented sequence pair; a computational contrastive loss function is combined with a pair-wise recommendation loss function to optimize an overall model, thereby improving the effect of a service recommendation model; and according to a feature embedding representation result of a service, pair-wise recommendation scores are computed to complete service recommendation.

Abstract: Disclosed in the present invention is a health state assessment method for equipment based on a knowledge graph attention network, includes: steps: 1) constructing a graph data model which can comprehensively reflect change of a health state of the equipment by deeply integrating association relationships of equipment components, monitoring data dependence relationships and priori information, etc. by means of a knowledge graph and by combining with domain priori knowledge; 2) extracting feature information of the health state knowledge graph by using a graph attention network, and obtaining a target node vector representation which accurately reflects the health state of the equipment by means of learning; and 3) making a health state representation vector of the equipment pass through a fully connected layer to obtain a health state classification prediction probability, and performing training to reducing a loss value relative to a true label, thereby obtaining a health state assessment result.

Abstract: A denoising method based on a multiscale distribution score for a point cloud includes: constructing a two-layer network model based on multiscale perturbation and point cloud distribution, where the two-layer network model includes a feature extraction module for extracting a feature of the point cloud and a displacement prediction module for predicting a displacement of a noise point; constructing a point cloud noise model for improving a denoising effect and retaining a sharp feature and avoiding reducing quality of point cloud data; extracting a global feature h by inputting the point cloud data into the feature extraction module; iteratively learning the displacement of the noise point by the displacement prediction module according to a feature obtained by the feature extraction unit; and defining a loss function of network training, and completing convergence under the condition that the loss function reaches a set threshold or a maximum number of iterations.

Abstract: The disclosed system may include a display; a lens; and a diffractive optical element, where the diffractive optical element is configured to increase a fill factor of the display when the lens is used to magnify the display. Various other apparatuses, systems, and methods are also disclosed.

Abstract: A digital twin based method for monitoring behavior of a passenger on an escalator includes: first, constructing a digital twin virtual scene of the escalator carrying the passenger; constructing a virtual person model in a physics engine, generating various types of risky behavior in reality, and outputting different types of behavior of a virtual person; second, through a human posture recognition method for monitoring behavior of a passenger, extracting a feature map as an input by means of a visual geometry group 19 (VGG19) pre-training network, to enter part affinity fields (PAFs) and a part confidence map (PCM), so as to complete recognition of human postures; classifying the human postures with a support vector machine; and performing emergency stop control on the escalator under risky behavior of the passenger according to a result of posture classification.

Abstract: Systems and methods for image data management are provided. The methods may include by running a first process, storing raw data collected in a scan of a subject into first storage space of the at least one storage device, and generating first information relating to the raw data. The methods may include by running a second process, obtaining the first information from the first process and storing the first information into second storage space of the at least one storage device. The methods may further include reconstructing the raw data into a reconstruction image of the subject based on the first information.

Abstract: Disclosed is an Aspergillus niger seed continuous culture method, comprising the steps of: (1) at a startup stage, Aspergillus niger spores are inoculated into a seed culture medium to obtain a seed liquid; (2) at a seed continuous culture stage, continuous dispersion treatment is performed on the seed liquid obtained in step (1), continuous culture is performed on the seed liquid obtained by dispersion, and meanwhile, a fresh seed feed medium is replenished; and (3) at a stop stage, the replenishment of the fresh seed feed medium and the dispersion treatment are stopped, continuous culture is performed to obtain a seed liquid, and then the seed liquid is transferred into the fermentation medium for fermentation culture. The method according to the present invention makes breakthrough to solve problems that multi-cellular filamentous bacteria grow slowly and mycelium pellets are easy to lose in continuous culture, thus fully achieving seed continuous culture.

Abstract: The present invention relates to aromatic ring or heteroaromatic ring compounds, a preparation method therefor and a medical use thereof. Particularly, the present invention relates to a compound as shown in general formula (I) and a preparation method therefor, a pharmaceutical composition comprising the compound and a use thereof as an agonist for a farnesoid X receptor (FXR). The compound and the pharmaceutical composition comprising the compound can be used for treating and/or preventing FXR activity-related diseases, for example, cholestatic symptoms, diabetes and complications thereof, nonalcoholic fatty liver disease (NAFLD), nonalcoholic steatosis heptitis (NASH), obesity or metabolic syndrome (diseases associated with dyslipidemia, diabetes, abnormally high body weight index), cardiovascular diseases and so on. The definition of each substituent in the general formula (I) is the same as that in the description.

Abstract: Chiral polymer microspheres have a porous structure of a concentric multi-shell structure. Each layer of the multi-shell structure is optically and structurally anisotropic. The optical axes of adjacent layers have a sequential slight twist. All layers of the multi-shell structure generate a helix configuration and the chiral polymer microspheres are optically active. A method for preparing the chiral polymer microspheres, includes: forming a homogeneous liquid crystal mixture; dispersing the liquid crystal mixture into a continuous phase to form liquid crystal droplets through an emulsification process; polymerizing the reactive liquid crystal to form intermediate microspheres; and removing the one non-reactive liquid crystal and the chiral additive to form the chiral polymer microspheres. The chiral polymer microspheres have a porous structure and a swelling ability, and can be used as the stationary phase in chiral chromatograph, improving separation efficiency.