1019 results about "Learning based" patented technology

The invention discloses a personalized learning path recommendation system based on artificial intelligence, and relates to the technical field of path recommendation, firstly, the system collects multi-dimensional feature data of a learner, and constructs a personalized feature vector; secondly, in combination with knowledge graph modeling and graph neural network technologies, deeply mining explicit and implicit knowledge point association; then, predicting an optimal learning path by using a sequence recommendation model, and ensuring reasonable sorting of knowledge points; in the learning process, the system combines real-time interaction data, dynamically adjusts a learning path, and continuously optimizes a recommendation strategy through an adaptive optimization algorithm; and finally, based on the learning result and the behavior data, evaluating the effectiveness of the learning path, and updating the knowledge point weight and recommendation strategy through a feedback mechanism, thereby realizing intelligent and self-adaptive personalized learning recommendation, the accuracy and adaptability of learning path recommendation can be effectively improved, learners are helped to master knowledge more efficiently, and the learning recommendation efficiency is improved. And the learning experience and effect are improved.

The invention provides a case quality intelligent evaluation method and system based on a large language model, and the method comprises the steps: building an evaluation standard library covering various cases based on a law normative file and judgment practice; designing a complete evaluation reasoning chain based on the evaluation standard library; constructing a multi-agent system based on the evaluation reasoning chain; based on the multi-agent system, realizing distributed task allocation for unplanned online tasks by applying an imprecise alternating direction multiplier method algorithm; based on the distributed task allocation result, combining planning and reinforcement learning technologies to solve a relational multi-agent case association problem; based on the multi-agent case association processing result, applying a forced zero method sparse graph technology to realize effective learning based on a case graph; and constructing a case quality assessment knowledge graph based on the effective learning result of the case graph. By adopting the technical scheme, the accuracy, comprehensiveness and efficiency of case quality evaluation are improved.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a time series data processing method, device and equipment and a medium. Performing a data enhancement operation on the original time series data to generate enhanced time series data; training a feature encoder based on the enhanced time sequence data in a comparative learning mode to obtain a pre-trained feature encoder; connecting the pre-training feature encoder with a sparse attention mechanism module to construct a time sequence modeling network; target time series data is processed using the time series modeling network to generate a processing result. According to the method, the enhanced view is constructed on the unlabeled data and the contrast learning training feature encoder is introduced, so that the time sequence representation with generalization ability is obtained, effective modeling of the long dependency relationship is realized in combination with a sparse attention mechanism, and the accuracy of time sequence modeling is improved under the condition of not depending on a large amount of labeled data.

A system answers natural language questions using retrieval-augmented generation. The system stores a set of domain specific documents in a vector database. The system receives a natural language question. The system retrieves a subset of documents relevant to the natural language question from the vector database. The system determines prior knowledge information required in addition to the subset of documents retrieved from the vector database for answering the natural language question. The system generates a prompt for a machine learning based language model including instructions to the machine learning based language model to refrain from using prior knowledge obtained by the machine learning based language model during training of the machine learning based language model. The receives a response generated by executing the machine learning based language model based on the prompt. The system performs an action based on the response.

Technology embodied in a method that includes receiving, as an input to a machine-learning model, data indicative of user-interaction of a particular user with a generative artificial intelligence (AI) system. The machine learning model is trained to identify one or more topics associated with inputs provided to the machine-learning model. The method also includes identifying a first context associated with the data indicative of the user-interaction with the generative AI system, and parsing a file system to determine that one or more folders within the file system correspond to the first context. The file system includes multiple folders each corresponding to a separate topic as identified from historical interactions of the particular user with the generative AI system. The data indicative of the interaction is augmented and provided to the generative AI system for generation of a response to the interaction.

The subject matter described herein includes methods, systems, and computer readable media for early detection of a neurodevelopmental or psychiatric disorder using scalable computational behavioral phenotyping. According to one method for early detection of a neurodevelopmental or psychiatric disorder using scalable computational behavioral phenotyping includes obtaining user related information, wherein the user related information includes metrics derived from a user interacting with one or more applications executing on at least one user device; generating, using the user related information and a machine learning based model, a user assessment report including a prediction value indicating a likelihood that the user has a neurodevelopmental or psychiatric (neurodevelopmental / psychiatric) disorder and a prediction confidence value computed using relative contributions of the metrics to the prediction value generated using the machine learning based model; and providing the user assessment report to a display or a data store.

The invention discloses a teaching information processing system and method based on electrical automation control, and relates to the technical field of intelligent teaching, and the method comprises the steps: collecting multi-modal teaching information, and carrying out the preprocessing; extracting and fusing features by using the preprocessed multi-modal teaching information to generate learning state features; constructing and dynamically adjusting a knowledge graph by using the learning state features to obtain a dynamic knowledge graph; generating a personalized learning path based on the dynamic knowledge graph, recommending learning resources, and recording the utilization rate of the learning resources; generating a comprehensive learning state of the student based on the learning state feature and the learning resource utilization rate; and generating teaching feedback according to the comprehensive learning state. According to the invention, by constructing the dynamic knowledge graph and combining the learning state characteristics and the learning resource utilization rate of the students, the knowledge point mastering condition of the students is dynamically evaluated, the personalized learning path is generated, and the adaptive learning resources are pushed.

The invention discloses a learning achievement prediction method based on learning behavior data mining and a computer device. The method comprises the steps of collecting multi-dimensional learning behavior data of students, performing preprocessing and feature extraction, constructing a structured learning behavior feature set, then constructing a target machine learning model suitable for different prediction tasks, inputting structured learning behavior features of the students to be predicted into the model, and obtaining a continuous cumulative risk prediction curve; converting the cumulative risk prediction curve into discrete classification labels, and generating a classification prediction result with reserved time sequence characteristics; and expanding the structured learning behavior feature set by adopting a pseudo-label semi-supervised learning algorithm, iteratively training a target machine learning model, inputting the structured learning behavior features of a to-be-predicted student into the trained model to obtain a comprehensive prediction result, and generating a visual learning achievement prediction scheme according to a classification prediction result. The method can provide a comprehensive and accurate prediction result.

An apparatus comprises at least one processing device configured to generate a first data structure comprising a numerical representation of content of a given system log associated with an information technology asset, to utilize the first data structure to determine a system log cluster to which the given system log belongs, to select non-anomalous system logs which are part of the system log cluster, and to perform contextual contrastive tuning of a machine learning model utilizing the selected non-anomalous system logs. The at least one processing device is further configured to generate a second data structure utilizing the tuned machine learning model which takes as input the first data structure, the second data structure characterizing (i) detected anomalies and (ii) causes of the detected anomalies. The at least one processing device is further configured to perform remediation actions, selected based on the second data structure, for the information technology asset.

The invention relates to the technical field of test question generation, in particular to an intelligent test question generation method and system based on learning behavior analysis. The method comprises the following steps: acquiring interactive behavior data and score data of learners in a user online learning platform in real time; inputting the interactive behavior data and the score data into a pre-trained knowledge state analysis model to generate a user knowledge state matrix; based on the knowledge state matrix and in combination with a preset teaching target library, identifying a target knowledge point set which needs to be strengthened currently and a corresponding cognitive training type; according to the target knowledge point set needing to be strengthened and the corresponding cognitive training type, a test question element combination algorithm is called, question stems, interference items and question solving path prompts are dynamically assembled, and personalized test questions are generated. The method has the advantages that full-closed-loop intelligent teaching from behavior analysis of the user to targeted training is achieved, and personalized test questions adaptive to individual cognitive vulnerabilities are dynamically generated.

The invention discloses a machine learning-based weld defect intelligent identification system, relates to the technical field of weld defect intelligent identification, solves the technical problems of multi-modal data fusion precision and robustness optimization and defect shielding or overlapping feature deficiency, and provides a machine learning-based weld defect intelligent identification method based on PSNR dynamic parameter adjustment and gradient weight optimization. The limitation of existing fixed parameter denoising is solved, the edge feature retention rate of cracks, air holes and other defects is improved, the omission ratio is reduced, improved DeepLabv3 + segmentation semantic masks are introduced and mapped to point cloud voxels, geometric + semantic double-attribute enhanced point clouds are formed, the defect area positioning accuracy is improved, and through a cross-modal attention module, the defect area positioning accuracy is improved. Weights are dynamically distributed according to illumination intensity and workpiece materials, feature waste caused by fixed weights is avoided, depth mutation and a shielding area with semantic defects are positioned by utilizing depth information of enhanced point cloud, real overlapping and projection overlapping can be effectively distinguished by combining an improved Poisson fusion algorithm, and the overlapping defect recognition accuracy is improved.

The invention belongs to the technical field of sea transportation port congestion prediction, and relates to a port congestion prediction deep learning method based on ship AIS data. The core of the method is that two neural networks of Transform and LSTM are combined, a global sea transportation liner transportation network is integrated, and the constructed model can capture the temporal and spatial change characteristics of the port congestion state from a space angle and a time angle. According to the method, the prediction precision of the port congestion state can be effectively improved by combining the dependency relationship of space and time. The method provided by the invention is obviously superior to an existing learning-based method, can learn and efficiently capture the complex and potential space-time correlation of the port congestion state, and is good in effect and high in practicability. And a certain enlightenment is provided for a port operator to monitor the congestion state of the port in real time so as to improve the operation efficiency of the port.

The invention discloses a medical data structured extraction method based on machine learning, and the method comprises the following steps: carrying out the standardization processing of multi-source heterogeneous data in different medical scenes, constructing a time and condition two-dimensional filtering rule, and extracting preliminary data; and a modular index structure is formed according to medical process and technical attribute division. And generating analysis limiting conditions by fusing the medical knowledge graph and the knowledge base, guiding an analysis engine to perform semantic routing and reasoning, and outputting a structured result. Finally, disease identification and quality judgment are achieved, and structured information meeting or not meeting the standard is output. The method aims at efficiently extracting the structured information from various types of medical documents.

Some embodiments of a method may include: decoding a motion feature by accessing a motion bitstream; predicting a predicted feature based on the motion feature and one or more reference point cloud frames; decoding a first feature representing an occupancy status of a child level voxel; predicting a second feature based on the first feature and the predicted feature; and decoding a tree voxel occupancy status of the child level voxel via the second feature.

The invention provides a multi-modal data alignment method and device, electronic equipment and a storage medium, and the method comprises the steps: inputting a multi-modal training session into a to-be-trained feature extraction model, and obtaining a modal feature of each modal data in the multi-modal training session; performing comparative learning according to the modal features of each modal data in the multi-modal training session to obtain a first loss; according to the modal feature of each modal data in the multi-modal training session, constructing a fusion feature, and inputting the fusion feature into a preset question and answer model to obtain a question answer corresponding to the multi-modal training session; semantic features of the answers to the questions are extracted, the semantic features are compared with the fusion features, and second losses are obtained; and according to the first loss and the second loss, adjusting training parameters of a feature extraction model to obtain a trained feature extraction model, and extracting a plurality of mutually aligned modal features of the multi-modal real-time session through the feature extraction model. According to the invention, the feature alignment effect of the multi-modal data can be improved.

The invention relates to an Internet of Vehicles edge computing multi-target unloading method and system fusing dynamic environment modeling and improved SARSA, and belongs to the field of intelligent traffic and edge computing fusion. The method and the system comprise MEC environment perception and multi-dimensional state construction, dynamic reward feedback oriented to multi-dimensional performance indexes, intelligent decision model construction and learning based on improved SARSA, and antagonism training oriented to real disturbance. A high-fidelity environment model is constructed through a space-time attention mechanism, an SARSA algorithm is improved to realize hierarchical qualification trace attenuation and collaborative Q table updating, a multi-target hierarchical reward engine is combined to implement differential optimization on an emergency task and a conventional task, and an adversarial training mechanism is introduced to improve robustness. The core problems of high mobility, task diversity, resource limitation and the like in the Internet of Vehicles are effectively solved, the comprehensive performance is optimal in multiple dimensions of delay, energy consumption, resource utilization rate and the like, and the actual landing of the edge computing technology of the Internet of Vehicles is promoted.

The invention provides a federated learning differential privacy method based on Rayleigh divergence and adaptive noise adjustment, and aims to balance data privacy protection and model training performance and improve model accuracy and convergence speed of federated learning on the premise of protecting user data privacy. And the contradiction between privacy protection and model performance in the existing federated learning is solved. The method comprises the following steps: step 1, constructing a privacy loss quantification model based on Rayleigh divergence; 2, deducing a tight upper bound of a Gaussian noise standard deviation; 3, initializing noise parameters of the federated learning system; 4, the client side executes local model training and noise adding; 5, updating the aggregation model of the central server and evaluating the performance; step 6, implementing a self-adaptive noise adjustment decision based on model performance; and step 7, iterating federal learning training until convergence or completion.

A system receives a natural language request for configuring a computing infrastructure using a cloud platform. The system executes a machine learning based language model to generate infrastructure-as-code (IaC) to configure a cloud platform to obtain the desired computing infrastructure. The system may display the IaC generated by the machine learning based language model via a user interface as an example for use by the user. The system may send instructions to the cloud platform to provision computing infrastructure in accordance with the IaC obtained from the machine learning based language model. The system may repeatedly determine whether the desired computing infrastructure is deployed on the cloud platform and if the computing infrastructure currently provisioned on the cloud platform fails to match the desired computing infrastructure according to the natural language request, the system reconfigures the computing infrastructure deployed on the cloud platform.

The invention provides a personalized learning path planning system and method, and belongs to the technical field of emerging software and emerging technical services, and the system comprises a data acquisition module which is used for collecting learning behavior data of a terminal user, obtaining a learner portrait package based on the learning behavior data, and sending the learner portrait package to a server; the path planning module is used for carrying out node matching on the learning ability feature vector and a pre-constructed knowledge and skill map, outputting a to-be-learned content node set, dividing learning advanced levels and generating learning path description containing to-be-learned nodes and the advanced levels, and the path generation module is used for generating a primary path planning scheme, the scheme feasibility verification module is used for carrying out scheme feasibility verification in combination with the learning advanced level and the path adaptation parameters, and then outputting a final executable path scheme, and the scheme execution module is used for executing the final executable path scheme and controlling learning content pushing and progress adjustment. The problems that in the prior art, personalized learning path planning is not high in precision, not high in adaptability, lack of dynamic optimization and the like are solved.

A machine learning based computing method for automatic managing credit risks of first users, is disclosed. The machine learning based computing method includes: receiving inputs from electronic devices associated with second users; retrieving data associated with first users from databases; preprocessing the data to remove noises, outliers, and missing values, from datasets; determining, the credit risks of the entities based on the pre-processed data by machine learning models; generating credit decisions for the entities; generating confidence scores for credit decisions to classify the credit decisions, based on correlation between the data and credit decisions; determining recommended credit values, recommended first credit limits, and recommended second credit limits, based on classification of the credit decisions; and providing an output of the credit decisions, the recommended credit values and the recommended credit limits, to the second users on user interfaces associated with electronic devices.

The invention provides a knowledge graph link reasoning method, and belongs to the field of artificial intelligence and knowledge graphs. The method comprises the steps that a local structure is reserved through a social triple integrity hypergraph, global association is captured through a social relation semantic hypergraph, and complementary social network double hypergraph representation is constructed; based on composite feature fusion of heterogeneous interaction, homogeneous parallel interaction and overall interaction, and in combination with a hypergraph attention network, relationship-guided dynamic semantic propagation is realized; designing multi-channel comparative learning based on social data; the social triple score is optimized based on a marginal sorting loss function, and a knowledge graph link prediction task and a self-supervision comparison task are fused through a joint learning framework. The problems of insufficient local structure modeling, global semantic information splitting, weak representation generalization ability, high dependence on labeled data and the like existing in knowledge graph link prediction in an existing social network are solved, and link reasoning tasks in a dynamic, sparse and high-noise-disturbance social scene are difficult to effectively deal with.

Techniques for applying a generative artificial intelligence (AI) model to identify and correct anomalies in remediation records are disclosed. A system trains and applies a generative AI model to displayed datasets to predict remediation record anomalies. If the system detects the generation of a remediation record in a dataset to reconcile the displayed datasets, the system generates a generative AI prompt that includes the remediation record. The generative AI model generates an output that identifies anomalies in the remediation record and the datasets being reconciled. The generative AI model further generates recommendations for remediating errors in the remediation record.

The invention discloses an enterprise and policy service automatic matching method based on machine learning, and the method comprises the steps: carrying out the processing of enterprise multi-source data, constructing an enterprise portrait graph, and generating an enterprise graph embedding and structuring feature set; performing policy text analysis and condition recognition, constructing a policy condition graph and generating condition graph embedding representation; constructing a condition constraint field based on policy conditions, and generating a cross-graph alignment relationship and fusion features; integrating a multi-source feature input improved model to carry out joint modeling, and outputting a basic matching score; constructing a condition boundary manifold, and generating an anti-fact feature sample and a matching elasticity score; and based on the basic and elastic scores, generating a comprehensive score and outputting matching result information. According to the invention, by introducing graph structure perception alignment, conditional constraint guide interaction and a multi-channel scoring aggregation mechanism, high-precision, high-interpretability and intelligent reachability automatic matching between enterprises and policy services is realized.

A multimodal content management system having a block-based data structure can include an artificial intelligence (AI)-based embeddings generator and indexer. After receiving an item update instruction that includes an object (e.g., a block content, a block property, or a block schema) identifier and an update payload, the system can transform the update payload—for example, by generating a chunk to capture at least a portion of the update payload. The chunk can correspond to a particular content modality included in the update payload. The system can generate and retrievably store a vector comprising a set of embeddings corresponding to the chunk, where the embeddings represent a vectorized portion of block content, block property, or block schema.

The invention relates to the technical field of remote sensing data processing and earth surface parameter inversion, and discloses an earth surface matrix parameter hyperspectral data inversion method and system based on machine learning. Comprising the following steps: constructing a multi-source heterogeneous hyperspectral data set; performing feature screening on the preprocessed hyperspectral data set based on an adaptive band selection algorithm, constructing a dynamic weight matrix by calculating mutual information entropy and inter-class distance measurement between spectral bands to realize intelligent screening of key feature bands, and combining spectral derivative conversion and spectral index calculation to generate an enhanced feature vector; and a multi-task transfer learning neural network model is constructed, and an output layer realizes multi-parameter collaborative inversion based on a multi-task learning architecture. And performing preprocessing and feature enhancement operation which is the same as that of the training data on the hyperspectral image data of the to-be-inverted region, inputting the trained neural network model, and outputting a surface matrix parameter inversion result.

A system and a method are disclosed for AI / ML model LCM. A method performed by a UE includes transmitting, to a base station, a plurality of properties of supported functionalities for each of a plurality of AI / ML use cases, wherein each of the plurality of AI / ML use cases is configured with an individual use case ID; transmitting, to the base station, a maximum number of AI / ML active functionalities that the UE supports across the plurality of AI / ML use cases; receiving, from the based station, a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI / ML use cases and the maximum number of AI / ML active functionalities that the UE supports across the plurality of AI / ML use cases; generating a report based on the report configuration; and transmitting the report to the base station.

The invention relates to the technical field of numerically-controlled machine tool diagnosis, and discloses a numerically-controlled machine tool fault diagnosis system based on machine learning. The system comprises a multi-source sensing data acquisition module for acquiring multi-dimensional sensing data such as vibration spectrum, spindle current waveform, temperature distribution, servo motor encoder feedback and the like; the operation feature coding module receives the multi-dimensional sensing data, extracts time domain statistical features and frequency domain energy distribution features, and generates a multi-source feature coding result; the incremental learning analysis module dynamically updates the feature weight through an incremental learning algorithm, and constructs an incremental training data set; the genetic optimization module optimizes the network structure and hyper-parameter configuration of the fault diagnosis model according to the incremental training data set, and generates optimized network structure parameters; and the integrated diagnosis decision module receives the current operation state data and the optimized network structure parameters, fuses diagnosis results of a plurality of base classifiers through an integrated learning algorithm, and outputs fault type classification signals.

The invention belongs to the technical field of information data processing, and particularly relates to a method and device for solving catastrophic memory of learning forgetting based on knowledge distillation. The method comprises the steps that an initial model is selected, a standard data set is used for training, and initial model parameters are optimized by minimizing cross entropy loss; selecting a target forgetting category, adjusting model parameters by applying a category forgetting method, and generating a forgetting model which is low in target forgetting category identification capability and high in other category identification accuracy; and selecting residual category data from the original data set as incremental learning data, adding quantitative target forgetting category data as input data, taking the generated forgetting model as a teacher model, initializing a student model, constraining output distribution by the student model on the input data through knowledge distillation loss, and updating parameters of the student model. The problem that the forgotten model cannot keep the forgetting performance in the incremental learning process, and catastrophic forgetting may occur when new information is absorbed is solved.

The invention discloses a learning-based composite layered anti-interference control system and method suitable for a large-windward-area unmanned aerial vehicle, is used for realizing robust wind resistance control of the large-windward-area unmanned aerial vehicle, and aims at the flight characteristics of large windward area and sensitivity to external wind field change in a vertical take-off and landing stage. A wind speed estimation method without an additional sensor is provided, modeling of external disturbing force is realized in combination with a Gaussian process regression algorithm, and compensation of the disturbing force is realized by using quaternion-based model prediction control, so that the influence of an external change wind field on the dynamics of the large-windward-area unmanned aerial vehicle is reduced, and the trajectory tracking precision is remarkably improved. A backstepping controller based on SO (3) is designed in an attitude control loop, a nonlinear disturbance observer is integrated, and active compensation of external disturbance torque and robust tracking control of attitude are realized. Based on the technical characteristics, a complete dynamic control link capable of estimating and feeding back compensation disturbance in real time is further constructed.

The invention discloses a mining area ecological restoration scheme intelligent decision-making system based on machine learning, and relates to the technical field of mining area ecological restoration, and the system comprises a multi-source data collection module, a preprocessing module, an ecological damage diagnosis module, a scheme generation and optimization module, and a model iteration optimization module. Multi-dimensional data information of a mining area is obtained through the multi-source data acquisition module, multi-source data are fused through a space-time alignment algorithm, a structured data set is formed, a basis is provided for subsequent analysis, specific ecological problem types of the mining area are recognized through a GBDT model in the ecological damage diagnosis module, ecological damage indexes of all areas are calculated, and the mining area ecological damage diagnosis method is applied to the mining area. According to the method, damage grades are divided, the accuracy of ecological damage diagnosis is improved, and based on an ecological damage assessment report, a high-matching-degree scheme is preliminarily screened in combination with historical case data, and an optimal scheme is finally screened out, so that the pertinence and effectiveness of the scheme are ensured.