Abstract: The present invention belongs to the technical field of data security of industrial Internet, and provides a method and system for detecting abnormal nodes in industrial Internet, a medium and a device. In the industrial Internet, different data holders firstly transform their own local node data into graphic data. Before local model training, the data holders firstly use a spectral clustering algorithm to perform certain clustering operations on local data, cluster the node data of the same category into the same cluster, and then perform local model training on a clustered result to obtain partial aggregation features. The trained partial features are uploaded to a trusted third-party server for global feature aggregation. Through an attention mechanism, different weights are assigned for partial features uploaded by different data holders, and the aggregated global features are delivered to each data holder for a new round of training.

Abstract: The present invention belongs to the technical field of data security of industrial Internet, and provides a method and system for detecting abnormal nodes in industrial Internet, a medium and a device. In the industrial Internet, different data holders firstly transform their own local node data into graphic data. Before local model training, the data holders firstly use a spectral clustering algorithm to perform certain clustering operations on local data, cluster the node data of the same category into the same cluster, and then perform local model training on a clustered result to obtain partial aggregation features. The trained partial features are uploaded to a trusted third-party server for global feature aggregation. Through an attention mechanism, different weights are assigned for partial features uploaded by different data holders, and the aggregated global features are delivered to each data holder for a new round of training.

Abstract: The present invention relates to the technical field of expression recognition, and in particular, to an expression recognition method and system based on multi-scale features and spatial attention. The method includes: performing feature extraction on acquired facial image data by using an HNFER neural network model to obtain an original input feature map; performing pooling and concatenation on extracted features based on a CoordAtt attention mechanism to obtain a feature map; performing deep convolution processing on the feature map to obtain an attention map, and then performing element-by-element multiplication to obtain a final feature map; and performing feature transformation and normalization on the final feature map to obtain an expression category probability and output the expression category probability.

Abstract: A hydrogen-bonded organic framework nanosheet, a preparation method and application thereof are provided. The preparation method includes in a small heat-resistant glass container, mixing 1,2,4,5-tetrakis (4-carboxyphenyl) benzene (H4TCPB) with N,N-dimethylformamide (DMF), heating until the solid is dissolved, placing the uncapped small glass container in a large heat-resistant glass container filled with some water, sealing the large glass container, standing and heating the large glass container in a constant-temperature oven to obtain colorless crystals, suction filtering to obtain a solid material, drying, grinding, and then dispersing the solid material in a solvent, performing ultrasonication in an ice-water bath, centrifuging to discard supernatant, and drying to obtain the hydrogen-bonded organic framework nanosheets.

Abstract: A suspended bed electro-membrane bioreactor system is provided, including a cathode chamber with a tubular structure with a sealed bottom; an anode chamber with a tubular structure located in the cathode chamber, where there is a gap between an outer wall of the anode chamber and an inner wall of the cathode chamber, and the anode chamber is provided with a proton channel region in a middle, is fixedly connected with an anode cover plate in a top and is filled with an anode assembly; a stainless steel membrane module located in the gap and electrically connected with the anode assembly; a polymethyl methacrylate partition plate located in the gap, forming a closed cylindrical space with the stainless steel membrane module and communicating the cathode chamber and the anode chamber; and a suspended bed filled between the closed cylindrical space and the anode chamber.

Abstract: The present invention discloses a double cantilever beam-encoding lead screw combined sensing tensile test method and machine. The double cantilever beam-encoding lead screw combined sensing tensile test machine comprises a main frame, a standard, a test piece and a microcomputer numerical control unit. The main frame is a force-deformation combined sensing mechanism composed of a double cantilever beam sensor, an encoding lead screw and a drive device. The double cantilever beam sensor is composed of a fixed cantilever beam sensor and a movable cantilever beam sensor. The encoding lead screw is composed of a drive lead screw and a tristate encoder. The double cantilever beam sensor matches with the encoding lead screw to achieve three functions, namely, test piece clamping, force sensing and deformation sensing, as well as to measure the size of the test piece.

Abstract: The present invention relates to a benzothiazole derivative of formula 1 or a pharmaceutically acceptable salt thereof and a process for preparation thereof. The present invention also relates to a pharmaceutical composition comprising the compounds and the use of the compounds in the preparation of an anti-tumor medicament.

Abstract: The present invention relates to a pyrazolo[1,5-a]pyrimidine derivative of general formula 1 and a pharmaceutically acceptable salt thereof. The present invention also relates to use of compounds of formula 1 in the preparation of an anti-tumor medicament.