Abstract: A check valve for low pressure operating conditions includes: a valve cover, a valve body, and an upper diaphragm between the body and the cover. A first pressure cavity is provided on a lower surface of the valve cover and used for accommodating a backward direction medium, a second pressure cavity is machined on an upper surface of the valve body, within which an inflow passage and an outflow passage are provided, an inflow hole is machined from an inner wall of the inflow passage to a bottom face of the second pressure cavity, and a medium enters into the second pressure cavity from the inflow passage via the inflow hole, and applies upward pressure on the upper diaphragm. A backflow channel in communication with the first pressure cavity is machined on a side wall of the outflow passage.

Abstract: The present disclosure provides an unbalanced sample classification method and an unbalanced sample classification apparatus. The method includes: obtaining unbalanced sample data; calculating a sample contribution rate based on the sample data and the characteristic data; filtering out a part of the sample data within a preset sample contribution threshold according to the sample contribution rate to determine as target sample data; and inputting the target sample data into a sample classification model to calculate a sample classification result through a classification algorithm. By using two variables of the characteristic value contribution rate and the characteristic contribution rate, the characteristics and samples with low contribution rate for classification are eliminated to effectively reducing the processing of unbalanced sample data, and a machine learning classification algorithm can be used on this basis to adopt the effective characteristics or samples to achieve efficient classification.

Abstract: The present disclosure provides a method, a device and a computer readable storage medium for food risk traceability information classification.

Abstract: The present disclosure provides a method for automatically collecting and matching laboratory data, including: obtaining a creation time of experimental data, determining target experimental data corresponding to a target time in accordance with the creation time, segmenting the target experimental data into a plurality data blocks, generating a data block index table, including at least one data block identifier, according to the data blocks, selecting a target matching mode from a plurality of predetermined matching modes according to the data block index table, obtaining the data block identifier upon determining the target experimental data in a storage node is loaded, and extracting data content in the target experimental data corresponding to the data block identifier by the target matching mode. This method may greatly reduce the number of string matching and may reduce the complexity of the algorithm.

Abstract: A FCN-based MTS data classification method is disclosed, comprising: generating input conditions according to a parameter of a multivariate Gaussian model and the MTS data; establishing correspondence between the input conditions and data categories of the MTS data via learning ability of an artificial intelligence model; determining at least one corresponding current input conditions according to current MTS of a target object; and determining current data categories corresponding to the current input conditions through the correspondence, and determining data categories corresponding to the input conditions identical to the current input conditions in the correspondence as the current data categories. The parameters of the multivariate Gaussian model corresponding to the MTS data are served as the input conditions, so that the accuracy is guaranteed, while a training speed of an artificial intelligence model is greatly improved, and the higher the data set dimension, the more significant the improvement.

Abstract: The present disclosure provides a method for automatically collecting and matching laboratory data, including: obtaining a creation time of experimental data, determining target experimental data corresponding to a target time in accordance with the creation time, segmenting the target experimental data into a plurality data blocks, generating a data block index table, including at least one data block identifier, according to the data blocks, selecting a target matching mode from a plurality of predetermined matching modes according to the data block index table, obtaining the data block identifier upon determining the target experimental data in a storage node is loaded, and extracting data content in the target experimental data corresponding to the data block identifier by the target matching mode. This method may greatly reduce the number of string matching and may reduce the complexity of the algorithm.

Abstract: The present disclosure provides an unbalanced sample classification method and an unbalanced sample classification apparatus. The method includes: obtaining unbalanced sample data; calculating a sample contribution rate based on the sample data and the characteristic data; filtering out a part of the sample data within a preset sample contribution threshold according to the sample contribution rate to determine as target sample data; and inputting the target sample data into a sample classification model to calculate a sample classification result through a classification algorithm. By using two variables of the characteristic value contribution rate and the characteristic contribution rate, the characteristics and samples with low contribution rate for classification are eliminated to effectively reducing the processing of unbalanced sample data, and a machine learning classification algorithm can be used on this basis to adopt the effective characteristics or samples to achieve efficient classification.

Abstract: The present disclosure provides a logistics node tracing method and apparatus for finding a trace node among logistics nodes in a logistics chain network corresponding to a logistics unit. The method includes: obtaining chain network information of a logistics chain network corresponding to a logistics unit, and determining a target analysis domain and confidence node(s) of the logistics unit according to the chain network information; determining fast node(s) according to the chain network information, the target analysis domain, and a timeliness level of each of the logistics nodes in the logistics chain network; determining a predicted logistics route corresponding to the logistics unit according to the chain network information, the target analysis domain, and the confidence node(s); and determining the trace node corresponding to the logistics unit according to the fast node(s) and the predicted logistics route.

Abstract: The present disclosure provides a method, a device and a computer readable storage medium for food risk traceability information classification.

Abstract: The present disclosure provides a logistics route prediction method and apparatus. The method includes: obtaining chain network information of a logistics chain network corresponding to a logistics unit, and determining a target analysis domain and confidence node(s) of the logistics unit according to the chain network information; determining fast node(s) according to the chain network information, the target analysis domain, and a timeliness level of each of the logistics nodes in the logistics chain network; and determining a predicted logistics route corresponding to the logistics unit according to the chain network information, the target analysis domain, and the confidence node(s). In which, the fast nodes at which the logistics unit passing through is determined first, thereby improving the prediction efficiency, and the confidence nodes are used as the basis for determining the predicted logistics route among multiple possible flow routes, thereby improving the reliability of the prediction.