76 results about "Learning analytics" patented technology

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 relates to the technical field of substation monitoring, and discloses an indoor substation high-voltage live display state identification method and system. The system comprises a data acquisition module for acquiring multi-modal state data of a display and optimizing an acquisition process; the state feature extraction module is used for performing incremental feature learning analysis on the real-time data; the identification model optimization module is used for optimizing model parameters and structures through a multi-target particle swarm; the state reasoning module is used for performing deep reasoning by using taboo search in combination with newly added data and an optimization model and converting the data into identification signals; the control output module is used for adjusting control constraints by using an approximate gradient method and optimizing output by combining a potential field method so as to generate a control instruction; and the state verification module verifies the state of the equipment through adaptive filtering and feeds parameters back to the related module so as to adjust the identification strategy. The system can comprehensively and accurately identify the state of the display, improves the adaptability and operation and maintenance safety, and adapts to the dynamic operation environment of a transformer substation.

The invention discloses a multi-modal learning data conjoint analysis method and system, a medium and a product, and relates to the field of multi-modal learning analysis. Comprising the following steps: acquiring synchronously acquired multi-modal data and extracting a feature vector; and calculating a confidence score of each modal feature vector and a semantic conflict coefficient between the modal feature vectors. When the semantic conflict coefficient is greater than a preset conflict threshold value, determining the modal feature vector with the maximum confidence score as a final state feature vector; and when the semantic conflict coefficient is smaller than or equal to the threshold value, calculating a fusion weight according to the confidence score, the semantic conflict coefficient and the task priority parameter, performing weighted fusion on each modal feature vector to obtain a final state feature vector, and generating an evaluation result according to the final state feature vector. According to the invention, through a decision-making mechanism that the optimal information source is selected in high conflict and context adaptive fusion is carried out in low conflict, the problem of inaccurate evaluation caused by forced fusion of contradictory signals is solved, and the accuracy and robustness of learning state evaluation in a complex scene are improved.

A method for cybersecurity threat actor severity scoring, the method comprising: receiving public data that includes publicly available information obtained via monitoring of a data connection between one or more networks; parsing first data related to a cybersecurity event from the public data; associating the first data with a first threat actor; obtaining second data that includes information regarding one or more previous cybersecurity events associated with the first threat actor; determining a first threat actor score based on the first data and the second data; receiving a second threat actor score for a second threat actor; causing a graphical user interface to display a graphical depiction of a ranking of the first threat actor and the second threat actor based on the first threat actor score and the second threat actor score.

There is a need for solutions that classification solutions in hierarchical prediction domains. This need can be addressed by, for example, performing one or more online machine learning, co-occurrence analysis machine learning, structured fusion machine learning, and unstructured fusion machine learning. In one example, structured predictions inputs are processed in accordance with an online machine learning analysis to generate structurally hierarchical predictions and in accordance with a co-occurrence analysis machine learning analysis to generate structurally non-hierarchical predictions. Then, the structurally hierarchical predictions and the structurally non-hierarchical predictions in accordance with processed by a structured fusion model to generate structure-based predictions. Afterward, the structure-based predictions and non-structure-based predictions are processed in accordance with an unstructured fusion model to generate one or more unstructured-fused predictions.

The invention discloses a learner cognitive diagnosis method oriented to an intelligent education scene, and belongs to the technical field of cognitive diagnosis and education data analysis. According to the method, concept-level multi-dimensional modeling is carried out on the ability and exercise difficulty of a learner by constructing multi-dimensional representation of concept perception so as to describe mastering characteristics of the learner on different cognitive levels; and meanwhile, distinguishing a pre-correction dependency relationship and a semantic approximation relationship, establishing a relationship-perceived concept dependency model, and deducing a potential exercise-concept association structure according to the relationship-perceived concept dependency model. Further, the annotated exercise-concept incidence matrix and the inference incidence matrix are fused in a unified diagnosis layer, and comprehensive evaluation of the knowledge mastering state of the learner is achieved. According to the method, an end-to-end mode is adopted for optimization, the fineness, knowledge coverage and stability of cognitive diagnosis can be effectively improved, better prediction performance and generalization ability are shown on multiple real education data sets, and the method is suitable for intelligent education application scenes such as learning analysis and personalized teaching.

The invention provides a parent terminal applet system for recommending and reporting student learning conditions based on parent portraits. The system comprises a data acquisition module for acquiring a student multi-source learning data set from students after a parent terminal applet logs in; meanwhile, parent portrait information is recommended and determined for parents of the students; the fusion processing module performs data processing on the multi-source learning data set of the student by adopting a data fusion method; the behavior analysis module performs learning behavior analysis according to the student multi-source learning processing data to obtain student learning analysis data; the report generation module is used for generating a learning condition report text based on student learning analysis data in combination with parent portrait information through a natural language processing model to obtain learning condition report text information; and the condition feedback module feeds back the learning condition of the student in the parent-side applet according to the learning condition report text information. According to the invention, learning conditions of students can be reflected more deeply, and parents can be ensured to accurately understand learning condition feedback information of the students.

A robotic insurance platform is disclosed that transforms an autonomous service robot into a tamper-resistant “digital witness” capable of supplying legally probative evidence without human intervention. The robot is equipped with surround video cameras, a spatial microphone array, an on-board processor, a cryptographically isolated secure enclave, a wireless communication module, and a rolling-buffer memory that retains encrypted audio-video data for a configurable period such as forty days. A companion mobile application enables an insured user to register, perform know-your-customer identity verification, and pair the robot with an insurance policy stored on a cloud server that hosts an claim-decision engine and policy database. Following explicit verbal consent from the user, the robot records continuously while performing ordinary tasks. When an incident is detected—either by on-board heuristics or by a user-initiated claim request—the processor extracts a time window surrounding the event, computes a cryptographic hash of the clip inside the secure enclave, and commits that hash as an immutable anchor to a permissioned or public blockchain ledger. Only after blockchain confirmation is the encrypted clip transmitted to the insurance server, where the claim-decision engine verifies integrity, applies machine-learning analytics to determine causation, and issues a coverage determination. Approved claims trigger repair dispatch, replacement shipment, or direct monetary reimbursement, while unclaimed data exceeding the retention interval are securely erased. The platform delivers objective, bias-free evidence, virtually eliminates false claims, and reduces end-to-end settlement time from weeks to minutes, thereby lowering operational costs for insurers and increasing transparency for policyholders.

The invention discloses an efficient data availability and invisibility implementation system, which comprises a trusted data client, a network space security center, a router, a big data platform and a block chain system, and is characterized in that the router is connected with the trusted data client, the network space security center and the big data platform, and the big data platform is connected with the block chain system. According to the invention, the data security client is provided to realize the data availability and invisibility function, and the data learning analysis plug-in platform is provided through the data security client, so that rapid fusion of the data security client and a data analysis system of a client can be realized, and the data can be taken to a user side for analysis and learning; therefore, various defects caused by the fact that the data needs to be learned and analyzed at the data provider are avoided.

The invention discloses a dynamic tracheal stenosis diagnosis system based on dynamic DR combined with deep learning, which relates to the technical field related to medical image intelligent analysis and comprises a dynamic DR imaging module, a time sequence data processing module, a deep learning analysis module and a visual report module. The invention further discloses a tracheal stenosis dynamic diagnosis method based on dynamic DR combined with deep learning. The tracheal stenosis dynamic diagnosis method comprises the steps of dynamic DR imaging, data processing, deep learning analysis and report generation. According to the method, 3D convolution and time sequence modeling are fused, real-time segmentation and tracking of a dynamic image are achieved, the performance is superior to that of a traditional 2DCNN or an independent time sequence model, data collection is optimized through a pulse perspective mode, respiratory gating and motion correction are combined, double challenges of dynamic image quality and algorithm robustness are solved, key data in a structured report can be filed in batches, and real-time segmentation and tracking of the dynamic image are achieved. A standardized database is provided for tracheal stenosis cause distribution, a thermodynamic diagram and a curve graph generated by the system provide high-quality annotation data, and iterative optimization of an algorithm is promoted.

The invention provides a data analysis method for personalized diagnosis and intervention of student learning questions, and the method comprises the following steps: S1, collecting an answer event sequence of a target student in a target question, the answer event sequence comprising an operation type and a corresponding timestamp; and S2, performing process segmentation on the answer event sequence based on a time interval of adjacent answer events, a submission event, a jump event and a rollback event. According to the method, process segmentation, step alignment and abnormal evidence extraction are carried out on the answering event sequence in the answering process of the student, and the learning abnormity of the student is specifically positioned to the corresponding standard question solving step and the trigger evidence thereof, so that a structured learning question positioning result is formed; therefore, a learning analysis result is converted from a single result index into traceable and explainable process diagnosis information, coarse-grained judgment only depending on an answer result or an abnormal score is avoided, and usability and pertinence of learning problem analysis are improved.

The invention discloses a ship lockage rolling pre-planning automation system, and belongs to the technical field of intelligent transportation and waterway transportation. According to the system, firstly, a digital twin navigation environment synchronized with a physical world in real time is constructed; in the environment, an initial scheduling strategy is generated through forward reinforcement learning training, the system allows a scheduling expert to intervene a pre-plan generated by the initial strategy, and operation of the pre-plan is collected to form an expert demonstration track; and then, analyzing the track based on reverse reinforcement learning, reversely reasoning the implicit decision preference of the expert and quantifying the implicit decision preference into an expert reward function, and finally, fusing the expert reward function with the initial reward function. The ship lockage scheduling method solves the technical problems that an existing scheduling system depends on a solidified rule, cannot learn expert implicit knowledge and is poor in adaptability, and the intelligent level, efficiency and safety of ship lockage scheduling in a complex dynamic environment are remarkably improved by constructing a closed-loop learning framework of man-machine co-evolution.

The invention discloses a lake water quality multi-parameter real-time monitoring method based on the Internet of Things. The method comprises the following steps: acquiring environmental indexes through a data acquisition device, and binding the environmental indexes with position depth information to generate three-dimensional data; cleaning data by adopting distributed calculation and marking anomalies, extracting normal data subsets, and reporting and converging the normal data subsets into regional feature parameters; dividing monitoring zones based on geographical and hydrological characteristics, and analyzing a parameter trend to construct a local dynamic sensing model; when it is detected that parameter linkage is abnormal, auxiliary parameter weights are analyzed through fusion learning to generate comprehensive monitoring indexes; and extracting full-depth data, performing dimension reduction to obtain global perception link data, and finally generating an optimized layout scheme through classification prediction analysis. According to the invention, multi-parameter and multi-depth accurate real-time monitoring and dynamic early warning of lake water quality are realized, the layout of monitoring equipment is optimized through data driving, and the monitoring efficiency and the management scientificity are obviously improved.

The invention discloses a learning plan analysis system based on beauty teaching, which comprises an education server, a learning analysis module and a learning strategy generation module, and is characterized in that the education server receives a learning progress result of a learner, and the learning analysis module identifies vulnerable knowledge elements by analyzing the result and generates a learning strategy; the learning strategy generation module generates supplementary learning content in a targeted manner; and meanwhile, functional modules such as a learning psychological analysis unit, a learning career analysis unit and an art learning evaluation unit are integrated, and algorithms such as deep learning, natural language processing and reinforcement learning are combined, so that functions such as learning mode recognition, beauty ability multi-dimensional evaluation, personalized learning strategy generation and career path recommendation are realized. According to the system, knowledge weak points and learning requirements of the learner can be accurately positioned, customized learning planning and feedback are provided in combination with beauty teaching characteristics, the learning efficiency and beauty literacy are effectively improved, and support is provided for comprehensive development and occupational planning of the learner.

The application provides a big data federal learning analysis method and system considering privacy protection, and belongs to the field of big data processing and data security. In view of the problems of traditional centralized analysis, such as easy data leakage, low compliance, and the problems of existing federal learning, such as insufficient privacy protection, low model accuracy, and untraceable parameter transmission, a scheme combining local hierarchical encryption desensitization, differential privacy disturbance, weighted federal aggregation and blockchain evidence is adopted. Each node encrypts and desensitizes the data according to the sensitivity level, locally completes model training, encrypts and transmits the disturbed parameters, and uploads and stores the evidence on the chain. The coordination node allocates weights according to the data volume and quality of each node, weighted aggregation generates a global model and iteratively optimizes it. The method realizes data "available but invisible", while ensuring privacy security and compliance, improves analysis accuracy and training efficiency, and can be used for multi-source sensitive data cross-domain collaborative analysis in medical treatment, finance, government affairs and the like.

The invention belongs to the technical field of intelligent education equipment, and particularly relates to an intelligent student health learning auxiliary table based on a network system and a management and control method. Student health data, student learning data, home-school interaction data and student real-time sitting posture data are collected; constructing a home-school co-management platform through a learning analysis algorithm, judging whether a sitting posture is correct or not by using an AI dynamic sitting posture recognition algorithm, if a bad sitting posture is detected, sending a correction signal through vibration reminding, pushing alarm information to the home-school co-management platform, and regularly and automatically recording height and weight data of students; homework data is synchronized to an electronic ink screen of a desk in real time to be displayed, homework completion progress and answering conditions are fed back through the intelligent desk, and a learning report is automatically generated according to an education data mining algorithm and pushed to the WeChat terminal in real time; and the user checks the learning report and the student healthy growth report and communicates with the teacher in real time through the platform. Therefore, the problems of lack of dynamic sitting posture recognition, personalized correction and the like in the prior art are solved.

The invention provides a pressure sore detection method and device, electronic equipment and a storage medium, and belongs to the technical field of medical detection. Inputting the first group of detection images into a deep learning analysis engine, so that the deep learning analysis engine obtains spectral features corresponding to the first group of detection images, and comparing the spectral features of the pressure sores at different stages with the spectral features corresponding to the first group of detection images by the deep learning analysis engine to obtain a pressure sore detection result; and outputting a pressure sore detection result. According to the method, the pressure sores are detected by fusing multi-dimensional optical information such as visible light, near-infrared light and ultraviolet light and combining a deep learning analysis engine, so that the pressure sore detection result is obtained, a basis is provided for prevention and management of the pressure sores, and the problem that the prevention and management effects of the pressure sores are poor is solved.

The present application relates to the technical field of power equipment maintenance, and discloses a method, system and device for controlling an upper limb exoskeleton for assisting in replacing an insulator. The method comprises collecting image data of the insulator and the on-site environment, and obtaining motion vector information from the image data based on a preset optical flow method; analyzing the motion vector information based on a deep learning algorithm to obtain motion state information of the insulator; obtaining adjustment motion parameters of the upper limb exoskeleton based on a dynamics fusion algorithm according to the motion state information of the insulator in combination with position information of the upper limb exoskeleton; and issuing a control instruction to the upper limb exoskeleton according to the adjustment motion parameters. The present application can monitor the dynamic changes of the insulator and the surrounding environment through the optical flow method, accurately identify and analyze the insulator in motion by combining deep learning analysis, timely adjust the motion trajectory of the upper limb exoskeleton, ensure that the upper limb exoskeleton stably and accurately tracks and operates the insulator, reduce the physical burden of the operator, and improve the operation safety and efficiency.

This invention discloses a multi-level and self-learning analysis method and medium for testing data. The method includes: acquiring and parsing product testing result data, and creating a general intelligent agent for the parsed data; determining the type of product problem based on the general intelligent agent, creating a corresponding specialized intelligent agent according to the problem type, and assigning the product problem to at least one corresponding specialized intelligent agent; establishing at least one attribution model based on the specialized intelligent agents to perform in-depth causal analysis of the product testing data and outputting the product reasoning chain; and performing hierarchical self-learning analysis of the product problem based on the reasoning chain and an accumulated knowledge base to obtain key factors affecting the product and the correlation between these factors. This application improves the depth and accuracy of product problem analysis by assigning data to general and specialized intelligent agents, with the general intelligent agent identifying the problem type and the specialized intelligent agents conducting in-depth analysis and mining to output the product reasoning chain.

The invention provides a specific population urban park satisfaction evaluation and influence factor identification method and system based on social media big data. According to the method, through social media data collection, specific crowd accurate screening, multi-dimensional text analysis, emotion value calculation and interpretable machine learning analysis, scientific evaluation of satisfaction and quantitative presentation of influence factors are realized, and reliable technical support is provided for fine design and quality improvement of urban parks.

The invention relates to a macroeconomic multi-source heterogeneous analysis system, in particular to the technical field of deep learning analysis, and the system comprises a data source acquisition module which is used for collecting data from a plurality of heterogeneous data sources, the plurality of heterogeneous data sources comprise but are not limited to government statistical data, financial market data, international trade data, social media data, industry report data and public database data, and the data source acquisition module can acquire data from online and offline channels in real time and support regular updating and incremental loading; in the invention, through integration of the data source acquisition module, data can be efficiently and comprehensively collected from a plurality of heterogeneous data sources, including government statistical data, financial market data, social media data and the like, the wide coverage of the data sources ensures that the system can comprehensively reflect various aspects of macroeconomy, and meanwhile, the data source acquisition module can effectively and comprehensively acquire the data. And the data acquisition module supports automatic acquisition and periodically performs incremental updating, so that the timeliness and the accuracy of the data are ensured.

The invention discloses an online course teaching decision-making auxiliary method and system based on multi-modal artificial intelligence, and belongs to the crossing field of education technology science and artificial intelligence technology, and the method comprises the steps: obtaining student comment data from an online course platform, carrying out the preprocessing, and constructing a labeling data set; based on a pre-trained text classification model, comments are classified according to preset dimensions, and multi-dimensional visual analysis of teaching feedback is realized in combination with sentiment analysis and keyword extraction technologies; aiming at the negative emotion comments, guiding a large language model through prompt language engineering to carry out deep semantic analysis, and identifying specific problems and roots thereof; based on the learning analysis technology framework, a targeted teaching optimization scheme is generated; the system is composed of a data layer module, an analysis layer module and a decision layer module and executes corresponding steps in the method. The system provided by the invention can be embedded into an existing online learning platform to serve as a teacher end auxiliary tool, supports normalized and periodic teaching decision and optimization, and has the characteristics of high practicability and easiness in popularization.

The invention provides an AI-based agricultural precise planting management method and system, and relates to the technical field of data processing, and the method comprises the steps: employing a multi-modal deep learning analysis model to analyze a standardized data set, and obtaining a crop state analysis result; fusing the position judgment result and the crop state analysis result to generate a crop state evaluation result; based on the crop state evaluation result, a partition farmland regulation and control scheme is obtained through a decision generation unit; the partitioned farmland regulation and control scheme is converted into a control instruction, the control instruction is sent to a field operation unit for corresponding operation, and crop response information is obtained; and inputting the crop response information into the multi-modal deep learning analysis model, and adjusting the decision strategy of the decision generation unit. According to the invention, the accuracy, dynamic optimization capability and management efficiency of agricultural planting management can be effectively improved.

The present disclosure provides methods for assessing a skin of a subject based on gene expression analysis. In an aspect, a method for assessing a skin of a subject comprises: (a) assaying a biological sample obtained or derived from the subject to produce a data set comprising gene expression measurements of the biological sample from each of a plurality of inflammatory skin disease-associated genomic loci, e.g., lupus, psoriasis, atopic dermatitis, and / or systemic sclerosis (scleroderma) disease-associated genomic loci, wherein the plurality of inflammatory skin disease-associated genomic loci comprises at least one gene selected from the group listed in Table 1, Table 2, Table 4A-1 to 4A-20, Table 4B-1 to 4B-28, Table 4C, Table 4D, or any combination thereof; (b) analyzing the data set to classify the skin of the subject as indicative of the inflammatory skin-disease associated disease state; and (c) electronically outputting a report indicative of the classification of the skin of the subject as indicative of the inflammatory skin-disease associated disease state.

The invention relates to the technical field of monitoring management, solves the technical problem that the influence degree of each link on the water quality in the whole-process production of barreled water is difficult to quantitatively monitor in a targeted manner, and particularly discloses a monitoring management system and method for the whole-process production of barreled water, and the system comprises a whole-process data collection module and a data preprocessing module. A deep learning analysis module, a decision and resource allocation module and an equipment linkage control module; the method comprises the following steps: S100, system deployment and initialization; s200, full-process data acquisition is carried out; s300, carrying out data preprocessing and feature engineering; s400, training a deep learning model; s500, carrying out real-time monitoring and analysis; and S600, decision making and execution are carried out. The system is used for barreled water whole-process integrated monitoring management, and through real-time data acquisition, deep learning analysis and decision making, the production process is controllable, and the water quality is stably guaranteed.

A system to perform multi-modal analysis has at least three distinct characteristics: an early abstraction layer for each data modality integrating homogeneous feature cues coming from different deep learning architectures for that data modality, a late abstraction layer for further integrating heterogeneous features extracted from different models or data modalities and output from the early abstraction layer, and a propagation-down strategy for joint network training in an end-to-end manner. The system is thus able to consider correlations among homogeneous features and correlations among heterogenous features at different levels of abstraction. The system further extracts and fuses discriminative information contained in these models and modalities for high performance emotion recognition.

The invention discloses a cognition-emotion evaluation method and system based on a supernormal child core feature theoretical model. The method comprises the following steps: presenting test tasks to multiple users at a computer end; collecting correct rate data of the work memory ability, the logical reasoning ability, the spatial cognitive ability and the speech ability by utilizing a cognitive test task; collecting non-cognitive test data by using a standardized scale; combining empirical weights, machine learning feature importance and large-scale norm data to establish a fractal dimension and comprehensive total score calculation rule; training a supernormal child identification model by taking Webster child intelligence test and Rayleigh reasoning test results as effect criteria; behavior data of new users are collected, and evaluation system construction and the supernormal child identification model are optimized and iterated. According to the method, various forms of cognitive test, non-cognitive test, static test and dynamic test are combined, and machine learning is utilized to analyze accuracy and non-cognitive scale scores, so that cognitive and emotional characteristics associated with supernormal child identification are accurately mapped.

The invention relates to a smart park energy management method and system based on digital twinning. The method comprises the following steps: acquiring park basic data to construct a basic digital twinborn model; energy equipment operation data and environment information data are collected in real time, and a data set is constructed based on public time axis association; inputting the energy equipment operation data into the basic model to generate an energy input model; acquiring historical energy consumption data to construct an energy consumption data set; analyzing the energy consumption data by using a machine self-learning algorithm to generate demand prediction; generating a predictive flow model based on the energy consumption demand input energy model; generating an energy complementation demand by using a machine self-learning analysis prediction model; and finally, generating a scheduling strategy according to the energy target and the complementation demand. The method has the effects of realizing energy optimization management, improving the energy efficiency and reducing the cost.

The invention discloses a crop growth state intelligent monitoring and diagnosis system based on deep learning, which belongs to the technical field of intelligent agriculture and comprises a multi-source data acquisition module, a deep learning analysis engine, a crop growth model platform and an intelligent diagnosis decision maker. And a deep learning analysis engine accurately identifies crop growth abnormality, plant diseases and insect pests and nutrition loss from the multispectral image through a convolutional neural network and an attention mechanism. The crop growth model platform establishes a dynamic standard curve based on historical data and adapts to different environmental conditions, and the intelligent diagnosis decision-making device automatically generates accurate fertilization, irrigation and plant protection suggestions. And the scientific level and the management efficiency of agricultural production are remarkably improved.

The system according to this embodiment aims to enable users to converse naturally with local people without language learning. [Solution] The system according to the embodiment comprises a reception unit, a learning unit, an analysis unit, and a provision unit. The reception unit receives user input. The learning unit learns based on data from social media or television. The analysis unit analyzes the input words. The provision unit provides the translation results.