34 results about "Scenario based" patented technology

There are provided systems and methods for scenario-based decisioning by machine learning models for data processing retry success. A service provider, such as an electronic transaction processor for digital transactions, may detect a failure of data processing for a transaction or other request when processed with a data processing system. In order to minimize cost and wasted resources for retrying transactions that are likely to further fail, machine learning models may be implemented that generates a predictive score for whether a failed transaction is likely to be successful if retried with the data processing system. This may be done when the transactions are received or detected, which may occur prior to failures, for different scenario-based decisioning. Scenarios for failures may be processed by the models with different retry strategies and a predictive score may be used to predict a probability of success of the retry.

The invention discloses a virtual reality interactive music teaching system based on a scene teaching method. The method comprises the following steps: dynamically adjusting the content pushing sequence and opportunity of a teaching scene based on real-time feedback data of a student; creating a virtual reality environment to simulate different music scenes and introducing interactive elements; a multi-level teaching resource library is provided to match the requirements of different learning stages; and a personalized guide prompt is triggered to enhance the understanding and participation sense of the trainee. Through the scheme, the problem of how to intelligently regulate and control the content pushing sequence and time of the teaching scene so as to solve the problem that the course progress is not matched with the understanding degree of students can be solved.

The invention discloses a decision planning system-oriented simulation test scene automatic generation method. The method comprises the steps of scene modeling: constructing a scene model by using a deep learning model; constraint generation: based on an output result of the scene model, utilizing a reinforcement learning model to generate a constraint condition; performing optimization search: based on the scene model and the constraint conditions, generating scene parameters meeting constraints by adopting a Bayesian optimization algorithm; and scene evaluation: evaluating the system performance by using the deep reinforcement learning model. According to the simulation test scene automatic generation method facing the decision planning system, manual intervention is reduced through an automatic method, and the scene generation efficiency is improved; diversified and high-coverage test scenes are generated, the scene diversity is enhanced, and the robustness of the system is comprehensively tested; constraint processing is optimized by efficiently processing complex constraint conditions, and it is ensured that a generated scene meets system requirements; and comprehensive system performance evaluation is provided, multi-dimensional evaluation is realized, and the performance of the system in different scenes is reflected.

Techniques for assigning scenario-based tests to test assets are described. In an example, a scenario-based test operable to test a key performance indicator (KPI) of a System Under Test (SUT), a component behavior exhibited by a first component of the SUT, and a scenario characteristic are received. Based on the component behavior and the scenario characteristic, a first plurality of behavior models associated with the component behavior are identified. Based on the scenario characteristic a characteristic value is extracted from the scenario-based test. Each behavior model of the first plurality of behavior models is executed using the characteristic value to generate a first plurality of predicted behavior outcomes. Based on the first plurality of predicted behavior outcomes, a first test asset type from a plurality of test asset types is selected and the scenario-based test is transmitted to a test asset of the first test asset type.

A method, computer program product, and computer system for receiving, by a computing device, input data from a user participating in a simulation scenario. Input data from the simulation scenario may be received. The simulation scenario may be monitored for one of a predicted performance of the user, the input data from the user participating in the simulation scenario, and the input data from the simulation scenario. One of the predicted performance of the user, the input data from the user participating in the simulation scenario, and the input data from the simulation scenario may be matched to a rule of a plurality of rules. An intervention event may be triggered in the simulation scenario based upon, at least in part, matching the one of the predicted performance of the user, the input data from the user participating in the simulation scenario, and the input data from the simulation scenario to the rule of the plurality of rules.

The present invention provides a method for optimizing a scenario-based code generation strategy, which relates to the technical field of software development. The method includes determining a target code template based on the analysis of project requirements; analyzing the coupling degree between different code segments to generate a reference code generation order; dynamically generating corresponding actual code parts by combining the reference code generation order with code key factors; testing the actual code parts and optimizing according to the test results. By determining a target code template based on the analysis of project requirements and providing an optimized code generation strategy for specific scenarios; analyzing the coupling degree between different code segments to generate a reference code generation order; generating corresponding actual code parts by combining the reference code generation order with code key factors; testing the actual code parts and optimizing according to the test results, code that meets the requirements can be accurately generated according to different project requirements and specific scenarios, improving the software development efficiency and quality.

The invention relates to a method for constructing a multi-scene credit risk control model and a risk control processing method and device, and the method comprises the steps: obtaining training data, which comprises user credit related historical data and credit states in a plurality of target scenes; performing feature extraction and feature preprocessing on the historical data related to the user credit to obtain training sample features; the training sample features are input into a to-be-trained multi-task learning model for training, default prediction probabilities in multiple target scenes are output, and the training labels are credit states in the corresponding target scenes; the trained multi-task learning model is used as a multi-scene credit risk control model; wherein the multi-task learning model is used for performing feature cross processing on input to obtain output features fused with global context information; and carrying out default prediction processing in the plurality of target scenes based on the output features to obtain corresponding default prediction probabilities in the target scenes. The model can give consideration to multi-scene prediction and guarantee the prediction accuracy.

The present invention discloses a multi-scenario virtual language dialogue training method and system, which relates to the technical field of language training. It includes a scenario topology management module for constructing a dynamic scenario relationship graph, which includes: a scenario node library for storing semantic vectors, dialogue strategy templates and trigger rules of each scenario; a dynamic edge weight calculation unit for real-time updating the connection weights between scenarios based on semantic similarity and user historical behavior data. The present invention realizes cross-domain dialogue intelligent migration through a dynamic scenario relationship graph, reduces deployment costs; multi-modal perception captures emotional fluctuations in real time and optimizes strategies; efficient data transmission supports immersive interaction; adversarial training enhances logical and emotional processing capabilities.

The application discloses a kind of based on cabin high delay scene volume control method, system and vehicle, method steps include: real-time monitoring user in preset time volume adjustment frequency, when volume exceeds the safety volume threshold, start volume anti-explosion strategy, detect audio signal, determine the normal fluctuation range of audio signal, the abnormal data points are removed to audio signal is intelligently analyzed and filtered, and verification is carried out using filter or machine learning model;Based on adaptive adjustment algorithm, volume setting is adjusted according to user usage habit and environmental noise, control volume smooth reduction, the adaptive adjustment algorithm has user volume preference model built-in.The application can effectively prevent the problem of too large or too small volume caused by improper operation by implementing anti-explosion strategy in the process of optimizing volume setting, so as to ensure the accuracy and stability of volume setting and improve the volume display effect.

To provide players with an engaging and enjoyable game. [Solution] The game system 1 includes a scenario acquisition unit 212 that acquires scenarios generated by a large-scale language model using creation instruction information, and a game provision unit 211 that provides a game progressing using the scenario to the player's terminal device. The creation instruction information includes outline information that shows the outline of the scenario, and a part of the outline information differs depending on the scenario rank. The scenario rank is determined by the player's actions, and the game progress result based on a scenario generated when the scenario rank is a first value is more advantageous to the player than the game progress result based on a scenario generated when the scenario rank is a second value lower than the first value.

The invention relates to the technical field of spacecraft orbit control and game decision, in particular to a reinforcement learning-based information waiting strategy forming method in an orbit pursuit scene, which comprises the following steps of: establishing a pulse round system game environment for describing relative movement of an orbit based on an orbit mechanical equation; based on the relative distance between the chasing party and the target, calculating the real-time observation confidence of the chasing party for observing the authenticity of the target, and updating the observation belief of the chasing party in real time by adopting Beta distribution; designing an expert strategy, and generating a state action database containing states and expert actions in a simulation environment; constructing a strategy network with an observation belief feature embedding layer, and pre-training the strategy network by using a state action database; and defining a reward function which aims at minimizing the capture time and contains distance penalty when the capture is not carried out, refining by maximizing a soft Q value of the sum of the reward and the strategy entropy, and finally outputting an optimal chasing party information waiting strategy with the high-entropy characteristic.

Methods, systems, and non-transitory computer-readable media are configured to perform operations comprising determining a symmetric scenario for a scenario; training a first machine learning model for the scenario based on first training data generated from second training data for the symmetric scenario; and generating a prediction for the scenario based on the first machine learning model.

A system (100) for performance enhancement of professionals and participants using artificial intelligence is provided An automated transcription module (130) records and transcribes video and audio data. An analysis module (135) performs real-time analysis using a large language model and a graph retrieval-augmented generation training model to identify performance gaps and areas for improvement. A performance metrics evaluation module (140) extracts a plurality of metrics to evaluate the performance. The personalized training module (145) generates customized training programs with rule-based prompts. A collaboration module (150) aggregates best practices and fosters knowledge sharing among professionals, while the performance tracking module (155) monitors their participation in development activities and tracks progress. A machine learning and adaptive learning module (160) predicts potential challenges and adapts training programs. A rewards module (165) enhances engagement through badges, leaderboards, and progress tracking, and a simulations and scenario-based learning module (170) for interactive practice in virtual environments.

This invention discloses a kill net generation method and system based on offline reinforcement learning, relating to the fields of simulation and reinforcement learning. To address the problems of slow kill net generation speed, poor adaptability, reliance on manual intervention, and lack of end-to-end coordination in existing technologies, this invention performs scene modeling based on the tasks, resources, and dynamic factors involved in the simulation scenario, including state space, action space, reward function, and state transition rules. Based on the scene modeling, reinforcement learning interactive simulation is performed to obtain offline data containing simulation trajectories. The offline data is divided into multi-task segments to obtain a multi-task offline dataset, and a meta-reinforcement learning training is performed on the generative trajectory decision network. The trained generative trajectory decision network is then used to generate the target kill net. This invention can efficiently generate highly adaptable and automated kill nets, quickly respond to changing battlefield environments, reduce manual intervention, and improve generation efficiency.

This invention relates to the interdisciplinary field of artificial intelligence and intelligent education, and provides a method for adaptively constructing virtual simulation training scenarios based on generative AI. It aims to address the problems of insufficient personalization and poor dynamic adaptability in existing training technologies due to their reliance on static pre-built scenario libraries. The method dynamically constructs and updates the cognitive state graph of trainees to achieve accurate state perception, and generates optimal scenario strategies through multi-agent collaborative inference and multi-objective game evaluation based on this. The optimal scenario strategies are then integrated with a static domain knowledge graph to generate structured instructions that drive a multimodal generative AI model to synthesize specific training scenario content. Finally, the virtual simulation training engine is driven to update the training scenario and collect feedback data, thereby constructing a closed loop of "perception-decision-generation-execution." This allows for the dynamic generation of personalized training content based on the trainees' real-time state, significantly improving the timeliness, adaptability, relevance, and personalization of the training.

The invention discloses a central control screen instruction set processing method and system based on a scene mode, and relates to the technical field of instruction set processing, and the method comprises the steps: carrying out the real-time fusion of multi-dimensional information flows from voice, a driver monitoring system, a vehicle bus and an external environment, and forming the depth perception of the current state and abstract intention of a driver; based on this, a large-scale language model is introduced as a decision core, the extracted high-order intention is taken as a target, and a preset vehicle function atomic operation library is taken as an available resource to carry out dynamic logic arrangement and combination. And finally, the initial instruction sequence generated by the model is combined with the real-time vehicle state to carry out instant verification and safety optimization, and an accurate and executable composite macro instruction is output. Therefore, a decision closed loop from multi-modal scene perception to adaptive task execution can be realized, so that the active service capability and seamless interaction experience of the intelligent cabin are improved.

A system is provided.SOLUTION: A system, comprising: means for collecting sales data, market data, store data, and salesperson data; means for pre-processing the collected data; means for analyzing the pre-processed data and extracting features; means for generating a role play scenario based on the extracted features; means for distributing the generated scenario to a terminal of each salesperson; means for performing role play by the salesperson based on the scenario and collecting feedback; and means for updating a machine learning model based on the collected feedback.SELECTED DRAWING: Figure 1

The present disclosure is a system and method to provide an interactive AI-driven soft skills training system that simulates realistic scenarios based on a five-factor personality model. It offers personalized coaching, real-time feedback, and a rewind and retry feature for iterative learning. Additionally, the system includes a meta-prompt functionality to generate industry-specific simulations by leveraging large language models to search and integrate relevant data, providing a comprehensive and adaptive training environment.

The present disclosure is a system and method to provide an interactive AI-driven soft skills training system that simulates realistic scenarios based on a five-factor personality model. It offers personalized coaching, real-time feedback, and a rewind and retry feature for iterative learning. Additionally, the system includes a meta-prompt functionality to generate industry-specific simulations by leveraging large language models to search and integrate relevant data, providing a comprehensive and adaptive training environment.

Methods, systems, and non-transitory computer-readable media are configured to perform operations comprising determining a symmetric scenario for a scenario; training a first machine learning model for the scenario based on first training data generated from second training data for the symmetric scenario; and generating a prediction for the scenario based on the first machine learning model.

This invention discloses an interactive tactical analysis method for team sports based on a large language model, comprising: multimodal alignment of a visually edited tactical board and tactical description text, and extraction of tactical knowledge from the aligned tactical image-text pair; customization of scenarios based on provided visual elements and extraction of scenario knowledge from the scenarios, and fusion of scenario knowledge and tactical knowledge to form domain knowledge; construction of a prompting engineering template for tactical analysis using the large language model; input of natural language based on provided structured tags, forming input prompts based on domain knowledge and the input natural language, combined with the prompting engineering template; tactical analysis and reasoning based on the input prompts by the large language model, and visualization of the tactical analysis results. This method allows users to fully express their tactical intentions to the large language model through visualization, including complex spatiotemporal context information, and enables iterative reasoning through thought chains, outputting easily understandable tactical analysis results.

The present application relates to the technical field of safety training and data analysis, and specifically discloses a safety training scenario construction method based on behavior tracing, which comprises the following steps: firstly, collecting the behavior time series data of the individual and the collaborative instruction time series data of the central control system in parallel; secondly, performing time stamp alignment and cross-modal fusion coding on the two types of data to generate joint representation data; thirdly, identifying and labeling the system-induced behavior nodes induced by the collaborative instruction logic through multi-scale sliding window analysis and bidirectional guidance intensity calculation; fourthly, synchronously reconstructing the behavior and instruction processes in a virtual simulation environment and constructing an interactive training scenario through high-correlation presentation of dynamic visual connectors; and finally, setting branch choices in the scenario, driving multi-path deduction and comparison, and generating a traceability analysis report containing individual decision evaluation and collaborative instruction quantitative optimization suggestions; the present application improves the accuracy and systematic improvement value of training.

An object of the system according to the embodiment is to create a comic that effectively conveys service content without expert knowledge.SOLUTION: A system according to an embodiment includes a service content acquisition section, a scenario generation section, and a comic creation section. The service content acquisition unit acquires service content from a user. The scenario generation unit generates a scenario based on the service content acquired by the service content acquisition unit. The comic creation section creates a comic on the basis of the scenario generated by the scenario generation section.SELECTED DRAWING: Figure 1

The system according to this embodiment aims to allow users to view past, present, and future events at a glance and to provide dialogue scenarios tailored to specific events. [Solution] The system according to the embodiment comprises a timeline view unit, a diary unit, a future prediction scenario unit, a special event scenario unit, and an emotion analysis and feedback unit. The timeline view unit allows users to view past, present, and future events at a glance. The diary unit records daily events. The future prediction scenario unit generates future scenarios based on the events recorded by the diary unit. The special event scenario unit provides special dialogue scenarios tailored to specific events. The emotion analysis and feedback unit analyzes the interaction with the user, understands their emotions, and provides appropriate feedback.

The invention provides a data preparation method and device for low-code GIS system use scheme recommendation, and relates to the technical field of geographic information systems. The scheme structured data generated by the method comprises the fields of scheme classification, scheme description, coded data marked with a user operation sequence, a scheme coding structure, data corresponding to a preset scheme execution efficiency index, data corresponding to a preset scheme used condition index and the like. The rich data dimensions enable a preset large model to more accurately understand user requirements when performing multiple rounds of dialogues with the user after learning training, find the most appropriate scheme under the corresponding classification according to the user requirements, and optimize and improve the scheme according to dialogue content.

The invention provides a user guidance method and system based on a teaching scene and a storage medium, and belongs to the technical field of artificial intelligence, and the method comprises the steps: determining the type of the teaching scene according to a teaching guidance request sent by a user; if the category is that guidance takes effect immediately, acquiring first observation information, processing the first observation information, and inputting the processed first observation information into a trained first coach agent to obtain first guidance information for the user; and if the category indicates that the guidance delay takes effect, acquiring second observation information, processing the second observation information, and inputting the processed second observation information into a trained second coach agent to obtain second guidance information for the user. According to the invention, the corresponding coach agents are used for processing different teaching scenes, the guidance information for the user can be automatically obtained, the labor cost is saved, and the response speed of guidance is improved.

A system is provided.SOLUTION: A system comprising: user registration means; means for displaying a plurality of scenario data; generation means for generating a conversation scenario based on scenario data selected by a user; acquisition means for acquiring answer data from the user; analysis means for analyzing the acquired answer data and generating feedback in real time; providing means for providing the generated feedback to the user; and means for storing and analyzing progress data of the user.SELECTED DRAWING: Figure 1

According to various embodiments, the disclosure discloses systems, methods and media for formulating training datasets for learning-based components in an autonomous driving vehicle (ADV). In an embodiment, an exemplary method includes allocating training datasets for training a learning-based model in the ADV, each training dataset being allocated to one of multiple predefined driving scenarios; determining a weight of each training dataset out of the training datasets; and optimizing the weight of each training dataset in one or more iterations according to a predetermined algorithm until a performance of the learning-based model reaches a predetermined threshold. The predetermined algorithm is one of a random search algorithm, a grid search algorithm, or a Bayesian algorithm.

A system is provided.SOLUTION: A system comprising: data input means for customizing based on data corresponding to a user; scenario generation means for generating a role playing scenario based on the input data; interaction means for interacting with the user according to the role playing scenario; response generation means for generating a response based on an interaction of the user; feedback generation means for generating feedback after the interaction; and storage means for storing a result of the interaction and the feedback.SELECTED DRAWING: Figure 1

This invention provides a scenario-based sandplay game-based language teaching method for early childhood, belonging to the field of early childhood language education technology. This method uses sandplay games as a medium and constructs a personalized language teaching system through a four-step core process: "theme scenario generation – age-appropriate design – interactive guided output – closed-loop evaluation and optimization." The core innovation lies in generating thematic sandplay scenarios such as characters, transportation, and animals according to the language development patterns of children aged 3-6, adapting to the age-appropriate goals of "vocabulary accumulation" in lower grades, "sentence structure application" in middle grades, and "story creation" in upper grades. Practical verification shows that this method can increase children's frequency of spontaneous speech by 60%, increase the percentage of children who tell complete stories from 25% to 72%, and reduce social conflicts by 40%. It also enhances teachers' ability to develop school-based curriculum, forming a new "gamified, contextualized, and personalized" early childhood language teaching model, applicable to kindergarten group teaching, group activities, and home-school collaboration scenarios, with significant promotional value.