Abstract: Systems and methods for training models to improve fairness.

Abstract: A computer vision learning system and corresponding computer-implemented method extract meaning from image content. The computer vision learning system comprises at least one image sensor that transforms light sensed from an environment of the computer vision learning system into image data representing a scene of the environment. The computer vision learning system further comprises a digital computational learning system that includes a network of actor perceiver predictor (APP) nodes and a library of visual methods available to the APP nodes for applying to the image data. The digital computational learning system employs the network in combination with the library to determine a response to a query and outputs the response determined. The query is associated with the scene. The computer vision learning system is capable of answering queries not just about what is happening in the scene, but what would happen based on the scene in view of hypothetical conditions and/or actions.

Abstract: A plant monitoring device (20) is provided with: a detection value acquisition unit (211) that acquires a bundle of detection values; a first Mahalanobis distance calculation unit (212) that calculates a first Mahalanobis distance by using as a reference a unit space generated on the basis of a bundle of past detection values; a first SN ratio calculation unit (214) that calculates a first SN ratio for each of a plurality of evaluation items; a second Mahalanobis distance calculation unit (215) that calculates a second Mahalanobis distance by increasing or decreasing each of the detection values; a second SN ratio acquisition unit (216) that converts the first SN ratio for each of the evaluation items and acquires a second SN ratio on the basis of the first and second Mahalanobis distances; and an addition unit (217) that calculates an added value of a plurality of the second SN ratios acquired within a prescribed period for each of the evaluation items.

Abstract: Embodiments disclosed include methods and apparatus for visualization of data and models (e.g., machine learning models) used to monitor and/or detect malware to ensure data integrity and/or to prevent or detect potential attacks. Embodiments disclosed include receiving information associated with artifacts scored by one or more sources of classification (e.g., models, databases, repositories). The method includes receiving inputs indicating threshold values or criteria associated with a classification of maliciousness of an artifact and for selecting sample artifacts. The method further includes classifying and selecting the artifacts, based on the criteria, to define a sample set, and based on the sample set, generating a ground truth indication of classification of maliciousness for each sample artifact in the sample set.

Abstract: A system and method detecting an artificial intelligence (AI) pipeline in a cloud computing environment. The method includes: inspecting a cloud computing environment for an AI pipeline component; detecting a connection between a first AI pipeline component and a second AI pipeline component; generating a representation of each of: the first AI pipeline component, the second AI pipeline component, and the connection, in a security database; and generating an AI pipeline based on the generated representations.

Abstract: A system and method of constructing a machine learning workflow by using machine learning suggestions derived from determining path lengths in a plurality of existing workflows, assigning a frequency threshold for each path and determining a probability for each path. This information is utilized to determine transpositions and deletions between paths that can be used as training for a machine learning algorithm that will suggest to the user which operators to put in a new machine learning workflow.

Abstract: In some implementations, a device may obtain a first document set associated with a first code repository. The device may generate a first embedding set of one or more embeddings for respective documents included in the first document set. The device may obtain a second embedding set of one or more embeddings for respective documents included in a second document set associated with a second code repository. The device may compare the first embedding set to the second embedding set. The device may generate a code repository similarity score that indicates a similarity between the first code repository and the second code repository. The device may perform, based on the code repository similarity score satisfying a threshold, an action associated with the first code repository and/or the second code repository.

Abstract: An apparatus comprising neural processors, a command processor, and a shared memory is provided. The command processor, in response to receiving a context start signal indicating a start of a context of a neural network model from a host system, directly accesses a memory in the host system to read command stream data for the neural network model. The command processor selects a current command and determines whether the current command is a branch command or a command describing neural network model tasks. The command processor selects a command among commands in the command stream data as a next command to be executed, based on a determination on whether the current command is the branch command or the command describing neural network model tasks.

Abstract: One embodiment provides a method, comprising: training, using deep imitation learning, a neural network associated with a predetermined ghosting model to predict player movements for at least one player during at least one sequence in a game; receiving, at an information handling device, tracking data associated with a player movement path for at least one player during the at least one sequence; analyzing, using a processor, the tracking data to determine at least one feature associated with the at least one player at a plurality of predetermined time points during the at least one sequence; and determining, using the predetermined ghosting model and the at least one feature, a ghosted movement path for the at least one player beginning from one of the plurality of predetermined time points. Other aspects are described and claimed.

Abstract: Devices and processing systems are configured to enable physiological event prediction based on blepharometric data analysis. For example, some embodiments provide methods and associated technology that enable retrospective analysis of blepharometric data driving subsequent hardware/software configuration, thereby to provide for personalized and/or generalized biomarker identification.

Abstract: Systems and methods for an accelerated tuning of hyperparameters of a model supported with prior learnings data include assessing subject models associated with a plurality of distinct sources of transfer tuning data, wherein the assessing includes implementing of: [1] a model relatedness assessment for each of a plurality of distinct pairwise subject models, and [2] a model coherence assessment for each of the plurality of distinct pairwise subject models; constructing a plurality of distinct prior mixture models based on the relatedness metric value and the coherence metric value for each of the plurality of distinct pairwise subject models, identifying sources of transfer tuning data based on identifying a distinct prior mixture model having a satisfactory model evidence fraction; and accelerating a tuning of hyperparameters of the target model based on transfer tuning data associated with the distinct prior mixture model having the satisfactory model evidence fraction.

Abstract: A characteristic included in data is determined with relatively high precision. An inspection system according to one aspect of the present invention acquires a plurality of learning data sets respectively including a combination of image data and correct answer data, and sets a difficulty level of determination for each of the learning data sets in accordance with a degree to which a result which is obtained by determining the acceptability of a product in the image data of each of the learning data sets by a first discriminator conforms to a correct answer indicated by the correct answer data. Besides, the inspection system constructs a second discriminator that determines the acceptability of the product by executing stepwise machine learning in which the learning data sets are utilized in ascending order of the set difficulty level.

Abstract: A machine learning framework is described for performing generation of candidate molecules for, e.g., drug discovery or other applications. The framework utilizes a pre-trained encoder-decoder model to interface between representations of molecules and embeddings for those molecules in a latent space. A fusion module is located between the encoder and decoder and is used to fuse an embedding for an input molecule with embeddings for one or more exemplary molecules selected from a database that is constructed according to a design criteria. The fused embedding is decoded using the decoder to generate a candidate molecule. The fusion module is trained to reconstruct a nearest neighbor to the input molecule from the database based on the sample of exemplary molecules. An iterative approach may be used during inference to dynamically update the database to include newly generated candidate molecules.

Abstract: An approach to identifying architectures of machine learning models meeting a user defined constraint. The approach can receive input associated with evaluating machine learning models from a user. The approach can determine acceptable architectural templates to evaluate the machine learning models based on the input and determine a list of architectures and metrics based on a calculation of maximum neural network sizes of the acceptable architectural templates not exceeding the constraint. The approach can send the list of architectures and metrics to the user for selection.

Abstract: A computer system is provided that is programmed to select feature sets from a large number of features. Features for a set are selected based on metagradient information returned from a machine learning process that has been performed on an earlier selected feature set. The process can iterate until a selected feature set converges or otherwise meets or exceeds a given threshold.

Abstract: Embodiments disclosed include systems, apparatus, and/or methods to receive a target health status of and/or quality of bioproduct produced by a managed livestock and indications of health status and quality of bioproduct. The systems, apparatus, and/or methods generate a set of input vectors based on the target health status or bioproduct quality, and the indications of bioproduct quality, and health status, and provide the set of input vectors to a machine learning model trained to generate an output indicating a feed selection. The feed selection can be included in a feed blend and administered to the managed livestock, such that, upon consumption, it increases a likelihood of collectively improving the health status of the managed livestock and the bioproduct quality of the managed livestock.

Abstract: Disclosed is a method for warning abnormal gas transmission loss, comprising obtaining gas flow data, gas pressure data, and ambient temperature data of a plurality of time points respectively based on the gas metering devices, the pressure detection devices, and the temperature monitoring devices; determining a gas metering error by processing the ambient temperature data, gas metering device information, and gas information using an error model; determining whether gas loss is abnormal loss based on the gas flow data, the gas pressure data, and the gas metering error; in response to the gas loss being the abnormal loss, determining a warning level based on a position and a size of the abnormal loss; and determining, based on the warning level, a warning notice corresponding to the warning level, and sending the warning notice to the gas user platform through the gas service platform to send a warning to a user.

Abstract: A content retrieval system may receive a query associated with a plurality of content items in a repository. For each content item of the plurality of content items: a respective first and second similarity score may be generated based on a similarity between embeddings indicative of a first and second data type generated from the query and for the content item; and a respective normalized similarity score may be generated based on a combination of the respective first and second similarity scores. A set of content items with respective normalized similarity scores that satisfy a similarity score threshold may be identified. An exact-match (lexical) search may yield respective mapping scores for content items that may also be ranked. An output indicative of content items that are identified in the set of content items with high-ranking similarity scores and identified in the set of content items with high-ranking mapping scores.

Abstract: A sound data processing method of a sound data processing device, the sound data processing device including a processing unit configured to acquire sound data of a target by input and to process the sound data, the sound data processing method including: a step of generating, by using acquired normal sound data of the target, simulated abnormal sound data that becomes a simulated abnormal sound of the target; and a step of performing machine learning by using the acquired normal sound data and the generated simulated abnormal sound data as learning sound data, and generating a learning model for determining an abnormal sound of the sound data of the target to perform abnormal sound detection.

Abstract: Providing custom machine learning models to client computer systems. Multiple machine learning models are accessed. Client-specific data for multiple client computer systems are also accessed. For each of at least some of the client computer systems, performing the following actions: First, using the corresponding client-specific data for the corresponding client computer system to determine which subset of the multiple machine learning models is applicable to the corresponding client computer system. The subset of the multiple machine learning models includes more than one of the multiple machine learning models. Then, aggregating the determined subset of the multiple machine learning models to generate an aggregated subset of machine learning models that is customized to the corresponding client computer system.

Abstract: Systems and methods for content selection and presentation are disclosed. Training data including indicating one or more interactions with one or more content elements and associated with one of a plurality of individual contexts is received. A selection model is trained by applying a reinforcement learning mechanism and an individual explore-exploit mechanism. A context for a user is selected by applying the selection model, which is configured to determine an expected future reward value of at least one of the plurality of individual contexts, determine an expected future reward value of a global context based on a past click-through rate, a reward value, and a future click-through rate, and, select the global context or one of the one or more individual contexts based on a comparison of the expected future reward values.

Abstract: Disclosed is a split computing device operating in a serverless edge computing environment. The split computing device includes a transceiver configured to receive resource information of a terminal from the terminal and to measure a data transmission rate between the terminal and the split computing device in a process of receiving the resource information of the terminal; and a splitting point deriver configured to determine a splitting point of a deep neural network (DNN) model for split computing and an activation status of a container instance for each of tail models of a DNN corresponding to the respective splitting points using the resource information of the terminal, the data transmission rate, and resource information of the split computing device.

Abstract: Examples include a method for providing multi-site orchestration in a public network for factory automation. The public network provides communication and computing functionality to a plurality of sites which are configured to communicate with each other by means of slice of the public network. The method includes building a multi-site orchestration model based on initial performance of communication between different sites; determining choreography opportunities between the different sites by using the multi-site orchestration model; triggering choreography between the different sites; and evaluating performance of the choreography between the different sites and updating the multi-site orchestration model.

Abstract: A local productivity prediction and management system including a weather monitoring device and a productivity prediction device. The weather monitoring device 10 including at least one of the following sensors adapted to take weather measurements of local weather conditions. The sensors include a temperature sensor 12, a humidity sensor 13, a rainfall sensor 14 and a sunlight and/or ultraviolet light sensor 15. Wherein, the productivity prediction device is adapted to over time collect local actual livestock production values. The productivity prediction device is also adapted to apply a productivity prediction model which uses one or more correlating patterns between weather measurements and actual livestock production values, whether either are local and/or offsite to provide a set of one or more predicted livestock production values.

Abstract: A computer-implemented method stratified elusion includes selecting hypothetical cutoff ranks when a stopping point is reached, calculating for each respective cutoff rank a recall value, elusion value, and remaining count; and displaying each respective cutoff rank, recall value, elusion value, and remaining count. A stratified elusion system includes a processor and a memory storing instructions that, when executed, cause the system to select cutoff ranks when a stopping point is reached, calculate for each respective cutoff rank a recall value, elusion value, and remaining count; and display each respective cutoff rank, recall value, elusion value, and remaining count.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for storing and accessing data in a cloud system. One of the methods includes receiving an identification of log data that records occurrences of events; receiving a specification of a plurality of different event types to be indexed; indexing the log data according to the specification and group identifiers; receiving a query specifying a reference parameter and requesting one or more predicted events; searching the indexed groups to identify a plurality of groups having events associated with the reference parameter; computing one or more predicted events, from the identified plurality of groups, that are most likely to co-occur in the indexed groups with events associated with the reference parameter; and providing the computed one or more predicted events.

Abstract: A method (100) for identifying time series data using a time series retrieval system (800), comprising: receiving (120) a plurality of time series, each time series comprising a plurality of datapoints, wherein a least some of the plurality of times series comprise datapoints obtained at irregular time intervals within the time period; storing (130) the received plurality of time series in a database; generate (140) a context vector for each of the plurality of time series; receiving (150) a request for identification of one or more of the plurality of time series based on similarity to a time series query; identifying (160), based on similarity to the query time series context vector, one or more of the stored generated context vectors; retrieving (170) each time series associated with the identified one or more stored generated context vectors; and providing (180) the retrieved time series.

Abstract: Disclosed herein is a computing device that includes a memory and a processor, which is coupled to the memory. The memory stores processor executable instructions for a robotic process engine. In operation, the robotic process engine generates a robot tray comprising a canvas and dynamically configures the canvas based on inputs. The dynamic configuring includes adding a widget onto the canvas.

Abstract: A method utilizes a framework for transaction categorization personalization. A transaction record is received. a baseline model is selected from a plurality of machine learning models. An account identifier, corresponding to the transaction record using the baseline model, is selected. The account identifier for the transaction record is presented.

Abstract: Techniques for verifying the identity of users transferring between devices are disclosed. An example method includes receiving device usage data by monitoring user-device interactions of a first set of users interacting with a first device and a second set of users interacting with a second device. The method also includes extracting features from the data and aggregating feature samples from the user-device interactions for first set of users and the second set of users. The method also includes generating a score for each feature based on an analysis the feature samples. The feature score indicates a degree to which the first set of feature samples and the second set of feature samples differentiate between the devices. The method also includes identifying features as transferrable if the score is below a specified threshold, and generating a new behavior model by modifying existing user behavior data according to the transferrable features.

Abstract: This application provides a model training-based communication method and apparatus, and a system, to effectively decrease a data amount of a parameter transmitted between the communication device and the central server. The method includes: The communication device determines a change amount of a first model parameter value. If the communication device determines, based on the change amount of the first model parameter value, that a first model parameter is stable, the communication device stops sending an update amount of the first model parameter value to the central server in a preset time period. The update amount of the first model parameter value is determined by the communication device based on user data in a process of performing model training. The communication device receives a second model parameter value sent by the central server.

Abstract: The invention provides an information processing apparatus comprising a first acquiring unit which acquires, from an external apparatus which can communicate with the information processing apparatus via a network, a first learning model which outputs recognition information in response to input of image information; a learning unit which causes the first model to learn a result of control by the information processing apparatus using recognition information that is output from a second learning model in response to input of the image information in an execution environment; and an output unit which causes the first learning model learned by the learning unit to output recognition information by inputting image information in the execution environment to the first learning model.

Abstract: A device may receive packet data identifying packets exchanged between client devices via a network of network devices. The device may classify the packet data based on timestamps and protocols associated with the packets and to generate classified packet data. The device may group the classified packet data into packet data sets corresponding to packet flows between pairs of the client devices. The device may select a packet data set, from the packet data sets, based on one or more filtering criteria. The device may process the packet data set to determine whether the packet data set is associated with one or more problem packets. The device may perform one or more actions based on determining whether the packet data set is associated with one or more problem packets.

Abstract: Techniques are disclosed by which a coding parameter is determined to encode video data resulting in encoded video data possessing a highest possible video quality. Features may be extracted from an input video sequence. The extracted features may be compared to features described in a model of coding parameters generated by a machine learning algorithm from reviews of previously-coded videos, extracted features of the previously-coded videos, and coding parameters of the previously-coded videos. When a match is detected between the extracted features of the input video sequence and extracted features represented in the model, a determination may be made as to whether coding parameters that correspond to the matching extracted feature correspond to a tier of service to which the input video sequence is to be coded.

Abstract: A computer system obtains information describing a geographical area segmented into multiple first clusters serviced by a telecommunications network. Multiple test locations are identified within the first clusters. Each test location is located within a grid of the geographical area. Each first cluster is recursively segmented into multiple second clusters until a difference between a number of test locations within each second cluster and a target number of test locations is less than a threshold number of test locations. A route is generated connecting test locations within each second cluster, using a routing application programming interface for performing drive testing of the telecommunications network. The computer system sends the route to one or more computer devices for performing the drive testing at the test locations in a sequence corresponding to the route.

Abstract: Systems and methods for modelling prediction errors in path-learning of an autonomous learning agent are provided. The traditional systems and methods provide for machine learning techniques, wherein estimation of errors in prediction is reduced with an increase in the number of path-iterations of the autonomous learning agent. Embodiments of the present disclosure provide for a two-stage modelling technique to model the prediction errors in the path-learning of the autonomous learning agent, wherein the two-stage modelling technique comprises extracting a plurality of fitted error values corresponding to a plurality of predicted actions and actual actions by implementing an Autoregressive moving average (ARMA) technique on a set of prediction error values; and estimating, by implementing a linear regression technique on the plurality of fitted error values, a probable deviation of the autonomous learning agent from each of an actual action amongst a plurality of predicted and actual actions.

Abstract: The present invention may include a computing device that receives root data, where the root data comprises a plurality of datasets. The computing device determines a context of each dataset within the plurality of datasets. The computing device identifies one or more users based on monitoring behavior of the one or more users related to each dataset. The computing device appends metadata associated with each dataset within the plurality of datasets with the determined context and the one or more identified users and setts access restrictions to each dataset in the metadata based on the context and the one or more identified users.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for collaboratively training an interaction prediction machine learning model using a plurality of user devices in a manner that respects user privacy. In one aspect, the machine learning model is configured to process an input comprising: (i) a search query, and (ii) a data element, to generate an output which characterizes a likelihood that a given user would interact with the data element if the data element were presented to the given user on a webpage identified by a search result responsive to the search query.

Abstract: Methods and systems are provided for key predictors and machine learning for configuring cell performance. One or more parameters relating to the cell may be measured, via a measurement apparatus, with the cell including a cathode, a separator, and a silicon-dominant anode, and the cell may be managed, based on the one or more parameters, with the managing including predetermining cycle life of the cell based on the one or more parameters using a machine learning model. The cell may be within a battery pack that includes a plurality of cells. The battery pack may be in an electric vehicle. At least one parameter may be measured before a formation process of the cell. At least one parameter may be measured during the formation process. At least one parameter may be measured during cycling of the cell.

Abstract: The present teaching relates to method, system, medium, and implementations for machine learning. A bid is received, from an expert during training, for a training sample with an amount within a level of available bidding currency associated with the expert. The training sample is used for training a model associated with the expert. It is determined whether the expert is among at least one winner selected based on bids from one or more experts. If the expert is among the at least one winner, the training sample is sent to the expert. The at least one winner is selected based on one or more criteria aiming at expert diversification.

Abstract: Methods are provided for inference processing of a decision tree model in processing apparatus which executes vector instructions to perform inference computations on vectors of operands stored in vector registers of the apparatus. Such a method includes, for each decision tree of the model, indexing nodes of the tree by consecutive node indexes which are assigned to nodes in a breadth-first order and increase with node-depth in the tree. During the inference processing, a vector of N node indexes, corresponding to a set of nodes for which N inference computations will be performed in parallel, is stored in a vector register of the apparatus. The method further includes adaptively selecting the granularity N of the vector of node indexes, in dependence on (at least) the node-depth of nodes in the set, to accelerate inference processing of the model.

Abstract: Disclosed is a method for generating a prediction model. The model can be used in processing machine event data to predict behavior of a plurality of industrial machines under supervision. The prediction model can be configured to determine current and future states of the industrial machines. The method can include extracting event features from event codes and structuring the event features into feature vectors. A first dimension of a first feature vector corresponds to a first event feature, and a second dimension of the first feature vector corresponds to a second event feature. The method can also include generating the prediction model by clustering the feature vectors into a plurality of vector clusters, the vector clusters assigned to respective machine states. The prediction model can be constructed based on event data from a first industrial machine and be applied to control an operating state of a second industrial machine.

Abstract: A system for maintaining consistency of a data value using a probability includes an interface and a processor. The interface is configured to receive a data value for storing. The processor is configured to store the data value in a data element of a data structure and determine, using an adaptive filter, a probability of certainty associated with the data value. The adaptive filter receives a previously stored data value in a previously stored data element of a previously stored data structure as input to determine the probability of certainty associated with the data value. The adaptive filter provides as output the probability of certainty. The processor is further configured to store the probability of certainty associated with the data value in the data structure.

Abstract: A device includes a memory, a network interface, and a processor. The memory is configured to store an aircraft performance model. The aircraft performance model is based on historical flight data of one or more aircraft. The aircraft performance model includes a recurrent neural network layer. The network interface is configured to receive real-time time-series flight data from a data bus of a first aircraft. The processor is configured to receive, via the network interface, the real-time time-series flight data. The processor is also configured to generate, based on the real-time time-series flight data and the aircraft performance model, one or more aircraft performance parameters. The processor is further configured to provide the aircraft performance parameters to a display device.

Abstract: The present disclosure includes: transforming a time-domain voltage signal collected by an MPI system device to a frequency domain; calculating a square root of a square sum of a real part and an imaginary part at each frequency point of a frequency domain signal; arranging acquired amplitudes in a descending order, and acquiring a screening threshold by an amplitude ratio method; screening an amplitude through the screening threshold and constructing frequency domain signal data; acquiring a row vector of a system matrix corresponding to each frequency point of the data, so as to construct an update system matrix; and solving, based on the frequency domain signal array and the update system matrix, an inverse problem in a form of a least square based on an L2 constraint to obtain a three-dimensional magnetic particle concentration distribution result, so as to achieve a fast reconstruction of the MPI system.

Abstract: A method generates a fence as a barrier between a user and an area of risk conditions within a designated space. One or more processors recognize an area of risk conditions, based a machine learning model trained by pre-determined indicators of the risk conditions and areas within the designated space. Movement patterns of a user are determined, based on machine learning of video images of the user's movements and behaviors. A risk condition is detected, based on the machine learning training of pre-determined areas and the pre-determined indicators of the risk conditions. Responsive to determining the user within or approaching the area of risk conditions, deploying, by the one or more processors, one or more mobile IoT devices as a fence between the user and the area of risk conditions, and generating a distraction diverting the user away from the area of the risk condition.

Abstract: A vehicle transaction support system operates by: generating, based on customer profile data and utilizing a vehicle recommendation engine having a first artificial intelligence model trained via machine learning, vehicle options data that indicates vehicles selected from a vehicle database; generating terms data associated with each of the vehicles selected from the vehicle database; sending the vehicle options data and terms data for display via a graphical user interface of a mobile device in a customer mode of operation of the mobile device; and communicating agent data with the mobile device in an agent mode of the mobile device, wherein the agent data is not displayed via the graphical user interface of the mobile device in the customer mode of operation.

Abstract: A method for detecting a deserialization attack may include identifying, in a byte stream, a class name corresponding to a class, generating, for the class, a feature vector, generating, by applying a benign deserialization model to the feature vector, a benign probability window, generating, by applying a malicious deserialization model to the feature vector, a malicious probability window, comparing the benign probability window and the malicious probability window to obtain a comparison result, and determining, based on the comparison result, that the class is malicious.

Abstract: The disclosure includes a system and method for obtaining, using one or more processors, case management signal data associated with one or more alerts, the case management signal data based on human interaction; training, using the one or more processors, a first model based on the case management signal data associated with the one or more alerts; and applying, using the one or more processors, the first model.

Abstract: The present application provides a method of training a natural language processing model, which relates to a field of artificial intelligence, and in particular to a field of natural language processing. A specific implementation scheme includes: performing a semantic learning for multi-tasks on an input text, so as to obtain a semantic feature for the multi-tasks, wherein the multi-tasks include a plurality of branch tasks; performing a feature learning for each branch task based on the semantic feature, so as to obtain a first output result for each branch task; calculating a loss for each branch task according to the first output result for the branch task; and adjusting a parameter of the natural language processing model according to the loss for each branch task. The present application further provides a method of processing a natural language, an electronic device, and a storage medium.