Abstract: The specification provides a method and a system for collaboratively updating a model by multiple parties for implementing privacy protection. A server can deliver an aggregation result of a t-th round of common samples to each participant i. Each participant i performs first update on a local ith model according to the t-th round of common samples and the aggregation result. Each participant i performs second update on the ith model obtained after the first update based on a first private sample fixed in a local sample set and a sample label thereof. Each participant i inputs a (t+1)th round of common samples that are used for a next round of iteration into the ith model obtained after the second update, and sends an output second prediction result to the server, so the server aggregates n second prediction results corresponding to n participants for a next round of iteration.

Abstract: The present specification discloses a model training method and apparatus, a service processing method and apparatus, a storage medium, and a device.

Abstract: Implementations of the present specification provide a data processing method, apparatus, and device. The method includes: obtaining to-be-detected target data, and obtaining a target probability that the target data corresponds to each candidate user intention, where the target data includes input data of a user in a human-computer interaction process; dividing the target data to obtain a plurality of pieces of subdata, and obtaining, based on a predetermined gradient integration algorithm, a contribution of each piece of subdata to a correspondence between the target data and each candidate user intention; and determining a target user intention corresponding to the target data based on the target probability that the target data corresponds to each candidate user intention and the contribution of each piece of subdata to the correspondence between the target data and each candidate user intention.

Abstract: Implementations of the present specification describe an intention identification model training method and apparatus, and an intention identification method and apparatus. According to the methods in the implementations, training of a target question can be weakened in the first several rounds of model training, and then an intention identification model obtained in the first several rounds of training can be used to identify intentions corresponding to answers that are to be distinguished. Further, the intention identification model is trained again after labels of these intentions are reset, so that the intention identification model obtained through training can also have a good identification effect on an answer to the target question, thereby improving accuracy of intention identification.

Abstract: A handcart, belonging to the technical field of handcart technology, includes a handcart body, a pull rod and a folding box; the pull rod is provided with a mounting component, the mounting component is rotatably connected to the handcart body; a bottom of the folding box is provided with an insertion component, a top of the folding box is configured with a first groove configured to be inserted in by the insertion component, the handcart body is configured with a second groove configured to be inserted in by the insertion component; the handcart body is rotatably connected to a first snap component, and the folding box is configured with a first snapping groove configured to be snap-fitted with the first snap component

Abstract: Implementations of the present specification disclose an event risk detection method, apparatus, and device.

Abstract: Intent recognition is described. Obtained data to be recognized is preprocessed to obtain target data. Feature extraction processing is performed on the target data based on an intent recognition model to obtain a linear feature and a non-linear feature of the target data, where the intent recognition model is obtained by optimizing and training a bidirectional encoder representations from transformers (BERT) model. Intent recognition processing is performed based on the linear feature, the non-linear feature, and the intent recognition model to obtain an intent recognition result of the obtained data to be recognized.

Abstract: Methods and apparatus, including computer programs encoded on computer storage media, for monitoring resource transfer are provided.

Abstract: A feature-selection system obtains a training data set and associated features; divides the training data set into a first number of training data subsets; and forms a plurality of feature-selecting data sets. A feature-selecting data set comprises a second number of training data subsets. The system processes, in parallel, each feature-selecting data set, which comprises: computing a first evaluation index for the features based on the feature-selecting data set; obtaining a group of index ranks corresponding to the features based on the first evaluation index; and obtaining a group of importance ranks corresponding to the features based on the feature-selecting data set and a machine-learning model. The system further obtains a group of total ranks by fusing groups of index ranks and groups importance ranks obtained from processing the plurality of feature-selecting data sets; and selecting target features from the features based on the group of total ranks.

Abstract: A downhole device configured to be inserted into a borehole includes a device body having an outer surface and a recess formed in the outer surface, a cover covering the recess to form a first cavity, and a shock-absorber configured to support an electrical module within the first cavity, the shock-absorber disposed between a base of the first cavity and the cover opposite the base. The downhole device also includes a vibration-damping layer disposed between the base of the first cavity and the cover, the vibration-damping layer configured to dampen vibration of the electrical module.

Abstract: A method for preparing a lithium-ion battery separator is disclosed. The method comprises: cooling and shaping a liquid-phase stabilization system containing polyethylene, stretching to enlarge pores, extracting with a solvent, and heat-setting to obtain a lithium-ion battery separator, wherein the stretching includes pre-stretching and synchronous bidirectional stretching, and the pre-stretching is completed before the synchronous bidirectional stretching.

Abstract: A method for data risk control comprises categorizing batch transaction data in a preset time period according to a first attribute, generating a graph indicator of a corresponding graph for each category of the batch transaction data according to a second attribute, the corresponding graph configured to identify risks in the batch transaction data, inputting the graph indicators corresponding to different categories of the batch transaction data into corresponding risk identification models, and determining whether the batch transaction data corresponding to the input graph indicators has a risk based on results output by the models. This scheme can quickly and accurately identify risks in batch transaction data.

Abstract: Disclosed in present invention are a high-wettability separator and a preparation method therefor. The separator comprises an ethylene copolymer, a grafting polyolefin, an ultrahigh molecular weight polyethylene having a molecular weight ranging from 1.0×106 to 10.0×106, and a high-density polyethylene having a density ranging from 0.940 g/cm3 to 0.976 g/cm3, the content of the ethylene copolymer is 1-5 parts by weight, and the content of the grafting polyolefin is 0-5 parts by weight, on the basis that the total weight of the ultrahigh molecular weight polyethylene and the high-density polyethylene is 100 parts. The separator has a contact angle with lithium ion battery electrolyte of 20° to 40°.

Abstract: An electrolyte for preparing a porous coating having a hierarchical structure on the surface of a titanium alloy by means of micro-arc oxidation is disclosed. A TiO2 coating having high porosity and a hierarchical structure can be prepared on the surface of the titanium alloy by means of a micro-arc oxidation method based on the electrolyte of the present invention. The electrolyte consists of 0.07-0.15 mol/L tetraborate and 0.10-0.40 mol/L strong base. The tetraborate includes one of Li2B4O7, Na2B4O7 and K2B4O7. The strong base is selected from one of KOH and NaOH. The coating prepared from the electrolyte has a hierarchical structure consisting of interconnected micron grooves or micronpores and randomly distributed nanopores, a layered structure consisting of inner pores and outer slots, high porosity and super-hydrophilicity, and can be used for surface modification in the field of orthopedic materials, dental implants, etc.

Abstract: The invention discloses a separator with a wide temperature range and a low heat shrinkage and a method for preparing the same. The invention belongs to the field of electrochemistry. The separator of the invention includes: an irradiation crosslinked fluoropolymer A with a melting point above 150° C. and/or a polymer B containing a benzene ring in its main chain; an ultrahigh molecular weight polyethylene having a molecular weight of 1.0×106-10.0×106, and a high density polyethylene having a density in the range of 0.940-0.976 g/cm3; the temperature difference between pore closing temperature and film breaking temperature of the separator is 80-90° C., preferably 85-90° C., the heat shrinkage of the separator is 2.0% or less.

Abstract: A feature-selection system obtains a training data set and associated features; divides the training data set into a first number of training data subsets; and forms a plurality of feature-selecting data sets. A feature-selecting data set comprises a second number of training data subsets. The system processes, in parallel, each feature-selecting data set, which comprises: computing a first evaluation index for the features based on the feature-selecting data set; obtaining a group of index ranks corresponding to the features based on the first evaluation index; and obtaining a group of importance ranks corresponding to the features based on the feature-selecting data set and a machine-learning model. The system further obtains a group of total ranks by fusing groups of index ranks and groups importance ranks obtained from processing the plurality of feature-selecting data sets; and selecting target features from the features based on the group of total ranks.

Abstract: The invention discloses a polymer electrolyte membrane and a method for preparing the same. The method comprises the steps of: (1) mixing a molecular sieve material, a polymer and a solvent to obtain a slurry; (2) coating the slurry on a base membrane to form a wet membrane; (3) drying the wet membrane to obtain a dry membrane; and (4) immersing the dry membrane in a lithium salt electrolyte solution, and taking out to obtain a polymer electrolyte membrane.

Abstract: A settlement channel recommendation computer-implemented method, medium, and system are disclosed. In one computer-implemented method, to-be-selected settlement channels and historical capital transaction information of a target user group is obtained. Respective historical indicator data of each settlement channel of the to-be-selected settlement channels is calculated based on the historical capital transaction information. A respective percentage of each settlement channel of the to-be-selected settlement channels is calculated based on the respective historical indicator data of each settlement channel of the to-be-selected settlement channels and target indicator data. An optimal settlement channel to the target user group is recommended based on the respective percentage of each settlement channel of the to-be-selected settlement channels.

Abstract: A computer-implemented method includes: receiving, by a computing device, a text report request from a user device associated with a user; obtaining a behavior history and personal information of the user; inputting the behavior history and the personal information of the user into a model, to obtain a plurality of personalized evaluation results, each personalized evaluation result corresponding to a respective text report category of a plurality of text report categories, in which each personalized evaluation result indicates a predicted relevance of the corresponding text report category to a problem faced by the user, and in which the model includes a classification model trained using one or more supervised learning techniques on a plurality of user behavior history samples and a plurality of personal information samples; and determining an order in which the plurality of text report categories are to be presented to the user.