Abstract: An embodiment includes responsive to receiving a dataset, determining by a system a relationship of a datastore by training a machine learning model wherein the machine learning model comprises an embedding model based on an attribute of the datastore. The embodiment includes determining by the system, a rule of the dataset using the machine learning model and the relationship of the datastore. The embodiment also includes computing using the machine learning model, a result for the rule of the dataset based on the datastore where the result comprises a description of the rule of the dataset in the datastore.

Abstract: An example operation includes one or more of receiving a natural language input via a software application, executing a machine learning model on the natural language input to determine a plurality of sub-queries corresponding to the natural language input, searching a tree structure for field values that correspond to the plurality of sub-queries, relationships between the field values, and computations to be performed, generating a structured query language (SQL) command based on the field values, relationships between the field values, and the computations to be performed, and executing the SQL command on a database to generate query results and return the query results to the software application.

Abstract: Provided are a method, system, and computer program product in which entities from a database are defined and extracted, based on semantic information, via a large language model (LLM). Data for structured query language (SQL) generation is generated to incorporate into a multi-layer semantic routable graph. Nodes related to user queries are determined based on the multi-layer semantic routable graph.

Abstract: An embodiment includes responsive to receiving a dataset, determining by a system a relationship of a datastore by training a machine learning model wherein the machine learning model comprises an embedding model based on an attribute of the datastore. The embodiment includes determining by the system, a rule of the dataset using the machine learning model and the relationship of the datastore. The embodiment also includes computing using the machine learning model, a result for the rule of the dataset based on the datastore where the result comprises a description of the rule of the dataset in the datastore.

Abstract: Generating transportation network solutions using graph neural networks, includes: generating, based on input data describing a product transportation network, a graph neural network representing the product transportation network, wherein the graph neural network comprises multiple edges, multiple source nodes, multiple destination nodes, and multiple intermediary nodes; performing a graph neural network prediction based on the graph neural network, wherein performing the graph neural network prediction comprises predicting inclusion in the transportation network solution for each edge and intermediary node; and generating a transportation network solution by performing a model optimization using an initial state based on the graph neural network prediction.

Abstract: A spiking neural network acceleration method includes constructing an approximate computation model according to a spiking neuron model and a spiking coding mode. The approximate computation model utilizes the characteristic that spiking frequency domain coding ignores the time semantics of a spiking sequence, compresses the distribution of spiking signals on a time step, and greatly reduces a spiking routing process and a subsequent neural computation process. The time-driven accelerator replans the computation process of the spiking neural network, sets a deduplication queue and a bitmap to solve the problem of spiking jitter, and realizes efficient support on the inhibitory spiking neural network.

Abstract: The present invention relates to an auxiliary method and facility for fish resource recovery in an open water diversion channel, which includes: an inlet of the attraction channel arranged on a side wall of the open water diversion channel, a slope provided inside the attraction channel, wherein a lower end of the slope is located inside the open water diversion channel, and a higher end of the slope is located outside the open water diversion channel, an attracted water flow is arranged at the higher end of the slope to guide fish into the attraction channel and an outlet of the attraction channel is connected to a still water zone, wherein the still water zone is provided with a release device. The present invention effectively guides a significant number of fish that have mistakenly entered the open water diversion channel out of the channel through attracted attraction.

Abstract: An example operation may include one or more of receiving a natural language input, executing a generative machine learning (ML) model on the natural language input to generate a structured query language (SQL) statement, executing the SQL statement on fake mockup data to generate a first query result on the fake mockup data, executing the generative ML model on the natural language input and the fake mockup data to generate a second query result on the fake mockup data, and determining whether the SQL statement is valid based on a comparison of the first query result and the second query result.

Abstract: A Monte Carlo tree structure of related software packages for a software project is constructed, wherein one or more vulnerable nodes of the Monte Carlo tree structure indicate a software vulnerability. A Monte Carlo tree search algorithm is applied on the Monte Carlo tree structure, the applying comprising expanding the Monte Carlo tree structure with multiple respective candidate nodes corresponding to alternative software packages for replacement of the one or more vulnerable nodes and performing iterative simulation and backpropagation through different tree updates using the multiple respective candidate nodes to identify at least one best replacement node from the multiple candidate nodes. A repaired software set for the software project is exported, the repaired software set including additional software corresponding to the at least one identified best replacement node, the additional software replacing one or more software packages that are related to the one or more vulnerable nodes.

Abstract: Generating structured query language queries, by extracting database schema from a target database, inputting the database schema into a generative model, training the generative model using a Monte-Carlo method and input data, training a discriminator model to distinguish generated and real input data, determining a loss value for generated input data, further training the generative model using the loss value, receiving, over a network, input text from a user, automatically generating, using the generative model, an output using the input text and providing, over the network, the output to the user.

Abstract: An displaying device includes a backlight module, a display panel located on one side of the backlight module, at least two first driving-chip groups, at least two first circuit boards and a system chip, each of the first driving-chip groups includes at least two first driving chips, and all of the first driving chips are electrically connected to the display panel. Each of the first circuit boards is electrically connected to one first driving-chip group, the first circuit boards are cascaded, and at least some of the signals transmitted in the first circuit boards are synchronous. The first circuit boards are electrically connected to the backlight module. The system chip is electrically connected to the first circuit boards, and configured for providing at least two channels of the same video signal to the at least two first circuit boards, and receiving a touch-controlling signal from the display panel.

Abstract: A multi-gas sensor to detect food spoilage and a method of forming the same are disclosed. The multi-gas sensor includes a silicon substrate and a plurality of chemical sensitive field effect transistor (CSFET) sensors formed on a surface of the silicon substrate, wherein each one of the plurality of CSFET sensors are decorated with a different material to detect a different gas associated with food spoilage.

Abstract: A message processing method can help protect confidential information, or other information that one desires not to disclose, from being disclosed, spread, or leaked, and improve the security of such protected information. A chat interface comprising at least one chat message is displayed. A target operation instruction is received that indicates a target operation that is to be executed on a target chat message. The target operation is executed in response to the target operation instruction. Protection processing on the protected content is performed when the target chat message includes a protected content, thereby protecting the information from being further disclosed and improving security of the information.

Abstract: Pfu DNA polymerase mutants with reverse transcriptase activity and their applications are provided. The Pfu DNA polymerase mutants include K467R/F588L/W769R, R382H/Y385H/V390I, I38L/R97M, E665K/E735K, K118I/N713V, with K118I/N713V showing the strongest reverse transcription activity. These Pfu DNA polymerase mutants still retain the polymerase activity of wild-type enzymes, thus providing higher assurance compared to traditional reverse transcriptase. Meanwhile, these Pfu DNA polymerase mutants exhibit high heat tolerance. Therefore, these Pfu DNA polymerase mutants can efficiently and high fidelity generate cDNA using RNA as a substrate, and further amplify the cDNA under conventional PCR reaction conditions without the need for additional reverse transcriptase, achieving a “one-step” RT-PCR process.

Abstract: A backlight module includes a light guide plate having a bottom surface, a light exit surface, and a light incident surface a substrate located on a side of the bottom surface away from the light exit surface; first light-emitting devices located on a side of the substrate proximate to the bottom surface second light-emitting devices disposed opposite to the light incident surface; and a prism sheet located on a side of the light guide plate away from the substrate. The first light-emitting devices are configured to emit first light, so that the first light passes through the light guide plate and the prism sheet to become light of a first viewing angle. The prism sheet is configured to cooperate with the light guide plate to modulate second light emitted by the second light-emitting devices into light of a second viewing angle which is smaller than the first viewing angle.

Abstract: A data processing method based on augmented reality is provided. The method includes: displaying a captured image of a scene in a human-computer interaction interface, the captured image comprising at least one object; identifying a subject object from the at least one object, the subject object being an only object or an object with a largest size in the captured image; and displaying at least one piece of information associated with the subject object in the human-computer interaction interface.

Abstract: A backlight module, and a display device and a driving method thereof, which relate to the technical field of displaying. The backlight module includes a substrate, and a plurality of light sources and a plurality of reflector cups arranged at one side of the substrate; each reflector cup includes a lower port, an upper port, a cup body connected between the lower port and the upper port, and a plurality of micro structures arranged on an inner wall surface of the cup body; the plurality of light sources are separately located at the lower ports of the different reflector cups; and light-emitting surfaces of the light sources face the upper ports; the micro structures have a plurality of reflecting surfaces: the reflecting surfaces are configured for reflecting light emitted by the light sources.

Abstract: A data processing method is performed by a receiving device. The method includes: receiving, through at least two network channels, data elements transmitted by a transmitting device, the data elements comprising delivery serial numbers; obtaining subsequent data elements of the received nth data element on the basis of the delivery serial numbers; and determining, when the subsequent data elements comprise a target data element, that a delivery result of the nth data element is discarded, a state of the target data element being a complete state, the target data element having a same data type as the data type of the nth data element, the target data element being an independent data element, the data type being determined on the basis of an effect of the data element, and a dependence degree of the independent data element on other data elements during post-processing being less than a specified degree.

Abstract: An array substrate, a display device, and a drive-for-heating method for the display device. The array substrate includes: pixel units on a first base substrate; a data line on the first base substrate; a touch signal line on the first base substrate; a heating element on the first base substrate, and configured to generate heat under driving of a drive-for-heating current; a common electrode on the first base substrate and reused as a touch electrode; and a pixel electrode on the first base substrate. The thin film transistor includes an active layer in a semiconductor layer and a gate electrode in a first conductive layer. The data line is in a second conductive layer, one of the pixel electrode and the common electrode is in a third conductive layer, and a layer where the heating element is located is between the semiconductor layer and the third conductive layer.

Abstract: A sediment dating correction method and system are provided, relating to the technical field of sediment dating. By utilizing Isotope characteristics of 210Pb, a sediment layer before 210Pb reaches equilibrium is taken as a modern substance (namely, modern sediment horizon), AMS14C dating data of the modern sediment horizon and AMS14C dating data of sediment to be dated obtained by an AMS14C dating method are interpolated to obtain a correction value of an old carbon, so as to correct all 14C dating data to obtain a chronological framework of the sediment to be dated.