65 results about "Training outcome" patented technology

A server-based video game system maintains a number of video game characters having computer-simulated genetic (“digenetic”) structures that prescribe a number of physical and cognitive performance traits and characteristics for the video game characters. The system allows end users to establish remote online access to the game characters (via, e.g., the Internet). The results of competitions and training activities are based upon the game characters' digenetics, the conditions of the game environment, and the game characters' current levels of physical and cognitive development. The game characters' performance capabilities and cognition are updated continuously in response to the results of competitions and training activities. Competition and training results can be processed by the game servers and transmitted to the end user presentation devices for graphics rendering. In this manner, the video game system need not be burdened by network latency and other delays. The game system also supports game character breeding; game character evolution based upon the digenetics; and game character buying, trading, selling, and collecting.

The invention discloses a semantic segmentation model training and image segmentation method and device, and calculating equipment, and belongs to the technical field of computer vision. The method comprises the steps: inputting a training image to a semantic segmentation model, and obtaining the preliminary result of semantic segmentation of the training image outputted by the semantic segmentation model; carrying out the fusion of local candidate regions according to the weak supervision information and the local candidate regions selected from the training image, and obtaining a correction result of semantic segmentation of the training image; correcting the parameters of the semantic segmentation model according to the preliminary result and a correction result; carrying out the iteration of the training steps till the training result of the semantic segmentation model meets a preset convergence condition. According to the scheme of the invention, the method achieves the direct supervision at the pixel level, also can optimize the semantic segmentation model in an end-to-end manner, and also can improve the result of segmented branches according to the judgment of the local candidate regions.

The invention discloses a text classification method based on a generative multi-task learning model, and the method comprises the steps: alternately decoding a multi-task classification model througha training set according to a multi-label classification task and a hierarchical classification task, obtaining a current-moment semantic feature related to a coding moment feature vector through decoding, and carrying out the training of the current-moment semantic feature; optimizing the multi-task classification model according to the training result and the real label set to obtain an optimized multi-task classification model; inputting data in a to-be-detected set into the optimized multi-task classification model, and respectively obtaining classification labels to finish classification. The semantic association between the multi-label classification model and the hierarchical classification model is constructed through a multi-task mechanism in the training process of the model, sothat the semantic association between two sub-task prediction results can be enhanced, and the purpose of improving the sub-task classification performance is achieved.

The invention provides a data model training method. The method includes the steps of acquiring a training set which includes multiple pieces of training data, and conducting first privacy protectionprocessing on the training set to obtain an encrypted training set; using the encrypted training set for training a data model to obtain model parameters; conducting second privacy protection processing on the obtained model parameters to obtain encryption model parameters, and generating an encrypted data model based on the encryption model parameters. Through the first privacy protection processing, the encrypted training set is different from the training set, and training results of the encrypted training set and the training set are the same; through the second privacy protection processing, the encryption model parameters are different from the model parameters, and the encrypted data model generated based on the encryption model parameters and the data model generated based on the model parameters have the same output results under the same input. The invention further provides a data model training device and computer equipment.

The invention provides a billiard training aiding system and method. The system comprises a camera, a client side and a server. A user can select a training subject according to needs, then training results are obtained by comparative analysis of the moving track of each billiard ball and the evaluating standard corresponding to the billiard training subject, the problem that only a striking advice is given by a billiard training system in the prior art is solved, the level of a trainee is obtained, and an advice is given. According to the scheme, whether striking of an athlete is successful is accurately judged according to billiard ball striking data, scientific grading of the level of athlete is promoted, accurate analysis on striking of the athlete is promoted, then striking of the athlete can be guided in a pointed mode, detailed correcting advices are given, and training effect is improved.

A ball training apparatus comprising: a ball; a ball-holding net to wrap around the ball; a tether cord that is connected to the net; a cord handle and cord length adjuster that is connected to one end of the cord; a spinning connector structure which is installed on the ball-holding net to connect the ball-holding net to the cord and allows the ball to spin relative to the cord. Thus the ball inside the net can rotate freely without intertwining with the tether. The handle of this ball training apparatus can either be attached to the user's waist belt or be held in hand and therefore the user can achieve various training outcomes.

The invention discloses a balance rehabilitation and cognition rehabilitation combined training system and method. Via the training system, a rehabilitation specialist sets rehabilitation contents aimed conditions of a patient, a training situation in which the patient feels less bored in the training process and adapted to the system is constructed according to the set rehabilitation contents, and balance rehabilitation and cognition rehabilitation are combined in the same situation; in the training process, the patient guides rehabilitation and cognition training by motions; and the system obtains motion data of the patient in the training process by body sensing equipment, calculates the gravity center of the patient, carries out visualized display, and displays a training result in a visual way and provides evaluation feedback timely, so that the rehabilitation specialist can guide and evaluate the patient conveniently. Thus, the patient feels less bored in the rehabilitation process, and is more confident in completing the rehabilitation training process.

The invention discloses a method for analyzing and predicting an online public opinion based on LDA topic models, and belongs to the technical field of big data analysis.The method is used for solving the problem about how to effectively organize large-scale documents, acquire evolution of topics in text sets according to the time sequence, and thus help a user track interesting topics.According to the technical scheme, documents in a corpus are firstly dispersed into corresponding time windows in a time sequence according to time information of the LDA topic models; then document sets in all the time windows are sequentially processed, training results on different time slices are obtained, and the training result of the former corpus is adopted as a prior parameter of a later corpus in the training process; finally, the change trend of the strength of all the LDA topic models over time is obtained from the training results, and a dynamic analysis and prediction function on the online public opinion is achieved.

The invention discloses a thyroid nodule recognition method based on a generative adversarial network. The thyroid nodule recognition method comprises the following steps: screening the data of a patient suffering from a thyroid disease, building a thyroid nodule database, and carrying out circling marking on thyroid nodules and thyroid parenchyma; semantic segmentation: taking U-net as a condition of a generative network to generate an adversarial network model, and inputting an annotated image to realize semantic segmentation; benign and malignant classification input: after multiple convolution, activation and pooling, using a full connection layer to integrate the extracted features to realize image benign and malignant judgment by the convolutional neural network; after the lesion isinput into the convolutional neural network, automatically classifying the lesion and outputting a benign and malignant judgment result of the lesion by the convolutional neural network; and giving areference diagnosis report of the illness state of the patient according to the training result. The thyroid nodule recognition method based on a generative adversarial network can improve thyroid nodule benign and malignant judgment accuracy of doctors, can reduce the thyroid screening time of ultrasound, can reduce the working intensity of medical staff, and can increase the satisfaction degreeof patients.

The invention relates to the field of rehabilitation evaluation, and relates to an intelligent evaluation method based on virtual reality rehabilitation training. The intelligent evaluation method comprises the steps that a movement angle evaluation mode during rehabilitation training is selected; an upper computer program acquires real-time angle data of the movement when a patient interacts witha virtual reality program; the virtual reality program drives an object in the game to move according to the angle data; real-time evaluation and feedback are performed according to the target completion condition of the patient; target data is uploaded to the upper computer program when the game is over; the training score is calculated according to a preset weight; and graphical processing is performed on the training score and other training data so as to generate an evaluation report. The invention further provides a corresponding system. The intelligent evaluation method and system realize real-time evaluation and feedback for the rehabilitation training while enabling the patient to realize limb rehabilitation training in the virtual scene by means of games, realize the evaluation report of the current session and an automatic analysis report of training results of different sessions after the rehabilitation training, and change many shortcomings of the original artificial rehabilitation evaluation.

The invention discloses an Alzheimer disease determination method and system based on ensemble learning. The method comprises the following steps: obtaining magnetic resonance imaging data of Alzheimer disease; training the magnetic resonance imaging data according to a convolutional neural network, so as to obtain a base classifier; performing ensemble learning on the base classifier, so as to obtain an ensemble classifier; obtaining a classification result serving as a determination result of Alzheimer disease according to the ensemble classifier; the system comprises an acquisition module,a training module, an ensemble module and a classification module. The method has the advantages that training is performed according to the convolutional neural network, ensemble learning is performed on a training result, finally the determination result of Alzheimer disease is obtained according to a result of ensemble learning, the early diagnosis accuracy rate and generalization capability ofAlzheimer disease are improved, and the method can be widely applied to the field of computer-aided diagnosis.

The invention discloses a tourism network comment emotion analysis and service quality evaluation method. The emotion analysis comprises the following steps: preprocessing the tourism network evaluation; constructing the training set of emotion trend; training the logistic regression model, support vector machine model and naive Bayesian model with the set of emotion trend training, and outputtingthe training results Gm (x) of three basic training models ; establishing a voter model, dealing with the training results, calculating the emotional tendency of comments; after processing the tourism network review data to be tested, inputting the voter model to get the emotional tendency of each comment. By synthesizing various machine learning algorithm models, the three classification algorithm models of logistic regression, support vector machine and Bayesian are inegrated to overcome the efficiency and accuracy of a single algorithm model, improve the accuracy of emotional analysis of tourism reviews, help tourists choose tourism services, provide decision support for tourism authorities to evaluate and improve the quality of regional tourism services.

An exercise system and an adjustment method thereof. The exercise system includes an exercise machine and a data transmission device. The exercise machine includes a control unit, a storage unit, a stepless force device, a sensor, a user operation part and an operation interface. During exercise, the control unit executes the exercise program to dynamically change resistance according to a sensed location, speed or acceleration of the user operation part and an adjustable parameter. Through the data transmission device, the user can download the personal training parameter from a remote device into the exercise machine, or upload the training result and the personal training parameter to the remote device.

The invention discloses a short-term load forecasting method based on an artificial neural network improved training strategy. The method comprises steps: an Elman artificial neural network is built, and pretreatment is carried out on original data to acquire a training sample; through setting different thresholds and weights, multiple groups of training results are acquired; by comparing the acquired multiple groups of training results, a forecast data set with the minimal error is recorded; initial hidden layer neuron numbers of the Elman artificial neural network are reset, and the training sample is substituted in the network for training; and an artificial neural network model corresponding to the group with the minimal validation sample error is selected as a forecasting model. A practice of multiple related day historical load data by inputting only one time point traditionally is changed into a practice of multiple related day historical load data by inputting multiple time points. When model forecasting is substituted, an intermediate value in a forecasting value sequence is only selected to serve as a target value for forecasting of the time, and influences on forecasting by edge effects and data fluctuation are avoided.

Anterior cruciate ligament (ACL) is the most commonly injured knee ligament in sports-related activities, especially in pivoting sports. Considering that the knee is free to flex / extend but has much more limited motions about the off-axes (tibial rotation and valgus / varus), ACL injuries are often associated with excessive off-axis loadings. A pivoting / sliding mechanism combined with sagittal plane movement / exercise is described as a diagnostic tool to evaluate off-axis neuromechanical and anatomical risk factors of noncontact ACL injuries. The pivoting / sliding mechanism is also described as an intervention tool for off-axis training, possibly based on the diagnosis, to reduce the incidence of ACL injuries. Training outcome can also be evaluated using the pivoting / sliding mechanism. In general, the pivoting / sliding mechanism can be used with many sagittal plane exercise machines and used to improve off-axis control of the lower limbs and reduce lower limb injuries.

The invention relates to a multi-sensing fusion-based juvenile basic athletic ability training system. The system comprises a somatosensory game device, an array-type flexible pressure sensor, a PC, adisplay and a cloud service platform; the somatosensory game device is provided with a Kinect camera and collects the limb movement information of players during a game playing process; the array-type flexible pressure sensor collects the foot pressure information of the players during the game playing process in real time; the PC compares the features of received information with features in a standard database, uploads the identity information of the players and training results to the cloud service platform in time, and intelligently executes training prescriptions; and the cloud service platform performs model calculation and result analysis on the received basic information of the players and the training results and evaluates the athletic ability of the players so as to generate personalized training prescriptions. The invention also discloses a training method for the multi-sensor fusion-based juvenile basic athletic ability training system. With the multi-sensing fusion-basedjuvenile basic athletic ability training system and the method of the present invention adopted, a closed-loop interactive training mode is formed, the interestingness of exercise can be enhanced, andthe improvement of the physical functions and basic motor skills of juveniles can be benefitted.

The invention discloses a monitoring evaluating system for badminton training. The system comprises an image acquiring module, a voice frequency acquiring module, a motion state acquiring module, a data processing module, an action standard judging module, a laser speed measuring device, an action guide and suggestion output module, an action command recognizing module, a three-dimensional teaching module, a puzzling question collecting module, a daily task completing condition collecting and evaluating module, an artificial expert answering module, a training result counting module, a man-machine operation module and a CPU. With the adoption of the system, the actions such as serving a badminton, receiving, spiking, moving and jumping can be automatically judged and guided; the three-dimensional teaching video launching function of the system is used, and the teaching video or guide video can be repeatedly played for learning, so that trainers can conveniently repeatedly view and understand each action, and as a result, the training efficiency is improved, and the action standard is improved, and meanwhile, the workload of a coach is decreased.

Systems and methods to provide a training solution for a trainee are disclosed. In some embodiments the method comprises receiving a training requirement comprising a training outcome and a training configuration wherein the training configuration defines a trainee state, determining a training environment based on a relevancy function of the training environment to the training outcome, determining a training content based on a relationship function of the training content to the trainee state and determining a training solution comprising the training environment and the training content. In some embodiments, the relationship function comprises a POMDP model and the relevancy function comprises a best fit curve.

Disclosed is an operation simulation training method based on a virtual reality technology. The method includes the following steps: sending operation training teaching material information to users in a medical school; carrying out operation simulation training by the users according to the operation training teaching material information, and obtaining training result information of the users; generating level evaluation reports of operation training in the medical school and an individual level evaluation report of each user according to the training result information of the users; sendingthe individual level evaluation reports of the users to the corresponding users according to a first preset period; and sending the level evaluation reports of operation training in the medical school to the designated users according to a second preset period.

The invention discloses a game type remote upper limb rehabilitation training system and method based on Kinect, and belongs to the technical field of medical rehabilitation. The game type remote upper limb rehabilitation training system comprises a mode selection module, a Kinect data acquisition and bone recognition module, a limb operation interaction module, a monitoring feedback end, a remote data transmission module and a training result feedback module. The mode selection module is used for selecting a rehabilitation action range mode; the Kinect data acquisition and bone recognition module is used for analyzing upper limb rehabilitation action; the limb operation interaction module is used for giving a game command when the upper limb rehabilitation action of a user meets a preset action standard corresponding to the mode selection module, and triggering a Russian grid game to conduct corresponding action in response to the game command; the monitoring feedback end is connected with the limb operation interaction module and used for enabling a doctor to view relevant data of the upper limb rehabilitation action and the Russian grid game and input feedback opinion data; the remote data transmission module is used for transmitting the relevant data and the feedback opinion data; the training result feedback module is used for feeding the relevant data and the feedback opinion data back to the user.

The invention discloses an attentional bias image training device and method. The device comprises a training content module, an interaction module and a control module, wherein the training content module is used for obtaining a training task and displaying multiple groups of image information in a preset time according to the training task; the interaction module is used for obtaining feedback information in a process of selecting relatively positive image from the multiple groups of image information by a testee in the preset time, wherein the feedback information comprises a using time andselection information; and the control module is used for obtaining a training result paying attention to the positive information according to the feedback information, and / or controlling the training content module to enter the next training task. According to the device, multiple training tasks can be trained by the testee so as to obtain the feedback information, so that the aim of cultivating behavior habits of positive attentional bias is realized according to the feedback information, help is provided to establish a more healthy cognition mode, and the mental health level is effectively improved; and the device is simple and easy to realize.

The invention discloses a data processing method and device for qualification authentication, and belongs to the technical field of information processing. The data processing method comprises the steps of acquiring and verifying user identity information; receiving a qualification authentication category selected by the user; if the qualification authentication category is a first qualification authentication category, sending a qualification certificate uploading request to the user; receiving a qualification certificate uploaded by a user, and verifying the qualification certificate according to authentication data acquired from an external data source; providing qualification training corresponding to the selected qualification authentication category for the user, assessing a trainingresult, and generating an electronic skill certificate correspondingly after the assessment is passed; and performing validity period monitoring on the qualification certificate and the electronicskill certificate. According to the method, various defects of traditional artificial qualification authentication are overcome, the working efficiency of qualification authentication is improved, andthe method is suitable for various data processing application scenes related to qualification authentication.

The invention discloses a Cognitive-behavior-therapy-based psychological capital promotion training method. The method comprises the following steps that: S1, a user starts a device and carries out registration or logging in; S2, information and psychological behavior data of the user are collected; S3, evaluation and diagnosis are carried out based on the information and psychological behavior data of the user and preset data; S4, a training solution is provided based on the evaluation and diagnosis; S5, human-machine interaction is performed and an interaction content is fed back to the S3,the steps S3, S44, and S5 are repeated successively until a training result meets the preset data; and S6, the training is completed. According to the invention, on the basis of decisive influence andeffect on all psychological disorders by negative thinking according to a cognitive behavior model, the cognition of the user is changed or the user is guided to evaluate the own thinking by using apractical and adaptive method, so that the emotional state and the behavior of the user are improved.

The invention discloses a text classification method based on a naive Bayesian model, and the method comprises the steps: collecting sample data, and constructing a training set and a verification set; selecting common information keywords, and determining characteristic attributes; establishing a naive Bayes model; training the naive Bayesian model by using the training set; verifying the training result by using the verification set, and correcting and perfecting; model application. According to the method, the text is classified on the basis of the naive Bayesian model, so that the accuracyand efficiency of classification work are improved, the workload of workers is reduced, and the smooth operation of work is ensured.

The invention discloses a computer professional technical ability verification training system. The computer professional technical ability verification training system comprises a WEB server used for installing the professional technical ability verification training system, and at least one computer terminal or smart phone terminal used for browsing and having access to the computer professional technical ability verification training system. The professional technical ability verification training system comprises a verification subject management module used for processing and storing the serial number and the subject name information of a professional technical ability verification subject, a verification topic management module capable of determining and storing the topic information according to one specific subject, an automatic test question generating module capable of automatically and rapidly selecting topics out to form a test paper meeting the requirement, a training and grading module capable of grading a student training answering result and storing the student training result, a training grade analyzing module capable of carrying out the statistics analysis on the student training result, and a user information management module for an administrator, a teacher and a student to enter. By means of the computer professional technical ability verification training system, the passing rate and the authenticated grade of professional technical ability verification of the student can be rapidly improved.

A trace guiding athletic training system mainly comprises a guiding trace forming unit and a control unit, wherein, the guiding trace forming unit can form changing guiding trace under the control of the control unit so as to guide a user to do physical training following the trace and detect the acting state of the user in match with different difficulty levels and the training models of acting traces, thus providing objective data for the reference of training results.

The invention discloses an APP-controlled football training device, which comprises a microcomputer unit, a serving unit, a goal unit, a recovery unit, a ball-up unit and an APP control terminal; The goal unit and the ball unit are connected, and the APP control terminal communicates with the microcomputer unit through wireless to control the training mode, training time, service time interval, control the speed and rotation direction of the ball during the training process; the microcomputer unit specifies the goal according to the set mode Unit; Only when the player kicks the football into the designated target goal within the specified time, is it considered a successful goal. The whole training process can be observed in real time on the APP control terminal, and the training results can be viewed on the APP control terminal after the training.

The invention discloses a fitness training resistance system and a controlling and adjusting method thereof. The fitness training resistance system comprises stepless fitness equipment and a data transmission device. The stepless fitness equipment comprises a user operating element, a stepless resistance source, a detection unit, a storage unit, an operating interface and a control unit. In a fitness course of a user, the control unit is in charge of executing a motion program and dynamically changing resistance in accordance with a position, a speed or an accelerated speed, which are detected as well as adjustable parameters. The data transmission device is connected to a far-end device; the control unit is used for selectively downloading personal training parameters from the far-en device by virtue of the data transmission device or for uploading a training result or the personal training parameters to the far-end device. Therefore, according to the fitness training resistance system provided by the invention, by storing the personal training parameters in a cloud mode, complexity of implementing individual setting by virtue of the stepless fitness equipment is simplified, so that user's fitness experience can be improved.