57 results about "Behavioral traits" patented technology

The invention relates to the technical field of language barrier evaluation, and discloses a deep learning-based language barrier executive intervention effect evaluation method. The method comprises the following steps: acquiring real-time voice data and facial expression data of a language barrier patient in intervention training through a multi-modal data acquisition device to form an original behavior feature set; performing acoustic feature hierarchical analysis on the voice data by adopting a time sequence feature extraction network to generate a voice time sequence feature vector; performing micro-expression dynamic capture on the expression data through a three-dimensional convolutional neural network to generate an expression state feature vector; inputting the two types of vectors into a multi-modal feature fusion layer to carry out cross-modal correlation analysis, and generating a comprehensive behavior evaluation matrix; on the basis of the matrix, an intervention effect analysis model driven by an attention mechanism is adopted, a behavior improvement degree index of the current intervention stage is calculated, accurate evaluation of the intervention effect is achieved, and support is provided for dynamic adjustment of language barrier rehabilitation intervention.

The embodiment of the invention discloses a background Alpha rhythm-based audio-visual perception ability evaluation method, which comprises the following steps of: acquiring an individual Alpha peak frequency of each subject, and setting various background rhythm stimuli for each subject; enabling each subject to execute various audio-visual perception tasks while the various background rhythms are stimulated, so as to collect electroencephalogram data and behavior data of each subject when the various audio-visual perception tasks are executed under the stimulation of the various background rhythms; performing feature extraction on the electroencephalogram data of each subject to obtain electroencephalogram features; performing feature extraction on the behavior data of each subject to obtain behavior features; screening electroencephalogram features related to improvement of the behavior features from the electroencephalogram features to obtain effective electroencephalogram features; and based on the effective electroencephalogram characteristics of each subject under the individual Alpha rhythm stimulation and the effective electroencephalogram characteristics of each subject under each contrast rhythm stimulation, calculating the modulation index of each subject so as to quantitatively evaluate the adjustment effect of the individual Alpha rhythm stimulation on the audiovisual perception ability of each subject.

This technology provides ways to diversify the behavioral characteristics of non-player characters. [Solution] In the acquisition step, situation information indicating the game status of a non-player character is acquired. In the output step, control information for controlling the actions of the non-player character is output based on the situation information and predetermined reference information. The non-player character is associated with a first character or other non-player character with whom it collaborates in a virtual space. The reference information is configured to allow referencing action information about the actions of the first character or other non-player character, and situation information at the timing when the control corresponding to the action information is executed. The action information is acquired sequentially via input from the first player operating the first character, or sequentially in response to actions taken by the first character or other non-player characters, and managed in an information processing system.

This invention discloses a method and system for early warning of abnormal behavior among college students based on multi-source data fusion, relating to the technical field of image processing. The method involves real-time acquisition of video streams of individuals within a target area, extraction of behavioral features from the video streams for classification, and a determination list based on these behavioral features. Multi-source data is collected, and anomaly coefficients are calculated based on the multi-source data. If the anomaly coefficient exceeds a threshold, the target individual is identified as exhibiting abnormal behavior. Anomaly warnings are then issued based on the behavioral features corresponding to the abnormal behavior. By calculating the anomaly coefficient through multi-source data fusion, a preliminary screening of hidden risks is achieved, reducing the detection delay and computational redundancy caused by traditional full-process monitoring analysis. Furthermore, real-time video stream acquisition and cloud-based behavioral feature extraction are initiated for high-risk individuals, avoiding the limitations of audio noise interference and single visual modalities, and improving the robustness of identification and the real-time performance of warnings in complex monitoring scenarios.

This application provides a method for assessing the training ability of a target audience, applicable to the field of artificial intelligence. The method includes: obtaining a first score for the target audience based on their answers in a business training system and the knowledge points in the knowledge graph at the current moment; analyzing the behavioral characteristics of the target audience during the answering process using a time-series analysis model to obtain a second score and a guessing coefficient, where the guessing coefficient characterizes the degree of guessing in the target audience's answers; correcting the second score based on the guessing coefficient to obtain a corrected second score; obtaining a comprehensive score using a gating fusion mechanism based on the first score, the guessing coefficient, and the corrected second score; and obtaining the training ability level corresponding to the comprehensive score from a preset training ability level table, using the training ability level as the assessment result of the target audience's training ability. This application also provides a device, equipment, storage medium, and program product for assessing the training ability of a target audience.

An emotion recognition method includes the following steps: acquiring video data of a human-object interaction behavior process; performing data labeling on the positions of a person and an object and the interaction behaviors and emotions expressed by the person; constructing a feature extraction model based on deep learning, extracting features of interaction between the person and the object in a time-space dimension, and detecting the position and category of the human-object interaction behavior; mapping the detected interaction behavior category into a vector form through a word vector model; and finally, constructing a fusion model based on deep learning, fusing the interaction behavior vector and the time-space interaction behavior features, and identifying the emotion expressed by the interaction person.

The invention discloses a comment emotion recognition method, device and equipment based on multi-modal fusion, and the method comprises the steps: obtaining all comment data associated with any comment target, and carrying out the preprocessing of the data, and obtaining the preprocessing data; respectively extracting image features and text features from the preprocessed data; extracting initial behavior characteristics matched with the plurality of behavior statistical items from the comment data; performing dynamic weighted fusion on each initial behavior feature to generate a target behavior feature; splicing the image features, the text features and the target behavior features to generate spliced features; and performing emotion recognition on the splicing features, and determining a comment emotion category corresponding to the comment target. Therefore, comment emotion category evaluation is carried out in combination with multi-modal data such as texts, images and behavior data, and the accuracy and comprehensiveness of emotion classification are effectively improved by carrying out dynamic weighted fusion on behavior characteristics.

This application discloses a method, apparatus, device, medium, and program product for identifying student psychological states, aiming to address the problems in existing technologies such as the difficulty in real-time, comprehensive, and dynamic identification of students' psychological states, and the lack of effective early detection and warning capabilities. The method includes: acquiring a multi-dimensional behavioral feature sequence of the target student, which includes at least an expression feature sequence, a behavioral trajectory feature sequence, and a voice emotion feature sequence; inputting the multi-dimensional behavioral feature sequence into a trained psychological state identification model to output psychological state representation information for characterizing the target student's psychological state; and identifying the target student's psychological state based on the psychological state representation information.

To provide feedback information on the behaviors performed by the person being evaluated, taking into account their behavioral characteristics. [Solution] The information processing device comprises: a first acquisition unit that acquires evaluator information relating to evaluators who evaluate the actions of the person being evaluated; a second acquisition unit that acquires behavioral information relating to the actions of the person being evaluated; a first analysis unit that analyzes methods for changing the evaluator's behavior based on the evaluator information; a second analysis unit that analyzes methods for changing the person being evaluated based on the behavioral information; a feedback generation unit that generates feedback information for the person being evaluated based on the analysis results from the first and second analysis units; and an output unit that outputs the generated feedback information.

The application discloses a radar human behavior recognition method based on multi-domain feature fusion, and is applied to the problem that the existing human behavior recognition method based on a radar sensor and a deep learning technology only adopts the features of one domain or only adopts one time-frequency analysis method in a time-frequency domain, thereby causing insufficient expression of human behavior features; three time-frequency analysis methods with different time-frequency resolutions, namely, a short-time Fourier transform, an adaptive optimal kernel time-frequency representation method and a Hann kernel reduction cross-term interference distribution, are selected; three types of frequency spectrum diagrams are obtained; then, the three types of time-frequency spectrum diagrams are combined and used by using a SE Net and a 3DCNN network in the time-frequency domain, so that human behavior features are more fully expressed; in a distance domain, a key feature is extracted by using the SE Net; and the features of two domains are combined, the mutual relationship is found, and the recognition accuracy of human behaviors is effectively improved.

The invention discloses a disease auxiliary diagnosis system and method based on multi-modal structured causal reasoning, and belongs to the technical field of intelligent auxiliary diagnosis. The system comprises a behavior feature layer which obtains behavior features representing behavior changes of a target object by synchronously collecting audio data and facial image video data of the target object; the scale entry layer is used for mapping the behavior characteristics into entry vectors of a psychological assessment clinical scale by utilizing a regression model; the symptom latent variable layer is used for aggregating entry vectors into symptom latent variables reflecting different clinical symptom dimensions according to the corresponding relation between scale entries and the clinical symptom dimensions; a disease category layer: obtaining a disease category influence probability used for analyzing the influence of symptoms on the disease risk by constructing a structured causal relationship graph; and the result interpretation layer is used for outputting a readable causal interpretation path and a feature contribution degree based on the causal structure diagram and the disease type prediction result. According to the invention, objective, structured and interpretable auxiliary decision-making of mental disease diagnosis can be realized.

The invention relates to the field of child mental evaluation, and discloses a child mental development self-service evaluation pavilion and an evaluation method thereof.The child mental development self-service evaluation pavilion comprises a pavilion body, the front end surface of the pavilion body is rotationally connected with an opening and closing door, the front end surface of the pavilion body is electrically connected with a coded lock module, and the coded lock module is electrically connected with a display panel; a folding chair is arranged on the right side of the inner side wall of the pavilion body. Behavior characteristics such as infant touch actions, facial expressions and the like are recorded through the video recording module with the resolution of 1080P and the field angle of 120 degrees, and the pickup module collects sound signals; after the AI algorithm preprocesses the data, quantitative parameters such as the effective number of times of touching red dots, the emotion similarity and the crying duration are extracted and compared with a built-in norm database of 1-3 years old, scores are calculated through a Z-score standardization and weighted summation formula, and subjective deviation is eliminated. The age correction coefficient further ensures the comparability of evaluation results of children of different month ages, so that the evaluation accuracy is improved by more than 40% compared with the traditional manual observation.

A behavior recognition system for analyzing an audio-video signal to detect challenging behaviors in autism via behavioral features. The system includes a processor configured to segment the audio-video signal into clips of audio data and video data, each of said clips having a predefined duration and annotated with interaction styles. The processor samples and preprocesses the audio data and video data of said clips to provide square video patches and square audio patches. The processor tokenizes the square video patches to embed video positional information and video modality information, and tokenize the square audio patches to embed audio positional information and audio modality information. And, the processor predicts behaviors based on the tokenized square video patches and the tokenized square audio patches.

The application belongs to the technical field of intelligent transportation, and discloses a driver workload prediction method based on multiple physiological signals under multiple auditory disturbances, which is created by combining three music rhythms (none, slow and fast), two navigation modes (high frequency and low frequency) and three traffic scenes (regular road, school road and construction site road), collects multi-modal data including skin electricity signal EDA, electrocardiogram signal ECG, electroencephalogram signal EEG, driving behavior characteristics and subjective workload evaluation, and the data is used to build a workload prediction model, and a stacking model is adopted, and the stacking model integrates multiple machine learning models to improve the prediction performance.

This application relates to the field of artificial intelligence technology, providing a method, apparatus, device, medium, and program product for detecting behavioral events anomalies. The method includes: extracting behavioral features of a target object from a video clip and extracting global scene features from a spatiotemporal cube; fusing the behavioral features and global scene features to obtain a heterogeneous feature fusion result; generating a prediction result for future frames based on the heterogeneous feature fusion result; and performing anomaly detection on the prediction result of the future frames to obtain a behavioral event anomaly detection result. The behavioral event anomaly detection method provided in this application extracts global scene features from a spatiotemporal cube and captures behavioral features of people from video clips. It heterogeneously fuses global scene features and behavioral features, learns the consistency relationship between heterogeneous features, achieves global context perception capability, strengthens the understanding of scene-dependent complex events, and improves the accuracy of identifying complex anomalies.

This invention provides an intelligent early warning method and electronic device for abnormal behavior in high-altitude operations based on human posture recognition, belonging to the field of intelligent early warning. This method can not only automatically extract the behavioral characteristics of workers, but also automatically identify and warn of abnormal behavior through a deep learning model, significantly improving the efficiency and accuracy of high-altitude operation safety monitoring. By introducing an improved HDBSCAN algorithm, the model can identify and cluster abnormal behaviors in real time in a dynamically changing work environment. Simultaneously, combined with human posture recognition technology, it can more accurately capture the dynamic posture changes of workers, thereby achieving early warning of potentially dangerous behaviors. This method overcomes the limitations of manual identification and slow response in traditional high-altitude operation safety monitoring methods, providing an efficient, accurate, and innovative solution for intelligent safety management of high-altitude operations.

According to the method, the semantic unit set converted from the interview video, the set dimension description and the character individual are input into the large language model, so that the large language model can identify the relationship of the character in the semantic unit set under the dimension; and combining the triple sets of different dimensions to obtain a large number of personality maps with personality background knowledge and behavior logic for constructing the individual. Through the question of the interviewee agent, the interviewee agent finds individual information based on the personality map, the big language model obtains an inference chain based on the individual information, and the big language model obtains an answer based on the inference chain and the question, so that the QRA is obtained. The QRA data pair set can make reasoning by utilizing established role information and background knowledge, replies conforming to role playing behavior characteristics and language characteristics are made for questions proposed by a user, and therefore a role playing agent supporting high role loyalty and context consistency can be obtained after supervising and fine tuning training is conducted on a large language model.

The application discloses a behavior monitoring method and device, electronic equipment and storage medium. The method comprises the following steps: extracting at least one second audio from a first audio; the first audio represents the sound emitted by at least two monitoring objects; each second audio in the at least one second audio corresponds to the sound emitted by one monitoring object in the at least two monitoring objects; inputting each second audio in the at least one second audio and the corresponding second video into a first setting model to obtain the first behavior characteristics corresponding to each monitoring object in the at least two monitoring objects; and matching the first behavior characteristics corresponding to each monitoring object in the at least two monitoring objects with first setting behavior characteristics to obtain a first behavior monitoring result; wherein the second video represents the video in which the corresponding monitoring object is shot. The above method can accurately and quickly locate the monitoring object with abnormal behavior, thereby improving the positioning efficiency.

The invention relates to the technical field of education, and discloses a classroom interaction evaluation and recommendation method and system, and the method comprises the steps: collecting audio and video data of a teacher in class; performing feature extraction on the audio and video data to obtain an extraction result; analyzing the extraction result by adopting a clustering algorithm to obtain an analysis result; and evaluating a classroom interaction effect according to an analysis result. According to the classroom interaction evaluation and recommendation method and system, the voice print recognition model is adopted to analyze the audio data of the teacher, the voice features of the teacher are extracted, and then the behavior features of the teacher are recognized based on the human skeleton key point detection technology. Thirdly, combining feature extraction results of the audio data and the video data, constructing teacher classroom feature vectors, and classifying the teachers by adopting a clustering algorithm, so as to divide the teachers into a passion type and a guide type; and finally, evaluating the classroom interaction effect according to the classification result and the classroom interaction index.

Embodiments of the present application provide a rat social behavior recognition method and device, computer equipment and a storage medium, relating to the technical field of animal behavior recognition, wherein the method comprises: acquiring a video comprising two rats, and determining key points of each rat in each frame of the video; in each frame of the image, according to the key points of each rat, determining the coordinate information and the skeletal posture of the key points of each rat; in each frame of the image, according to the coordinate information and the skeletal posture of the key points of each rat, extracting the behavior features of the two rats; inputting the behavior features of the two rats in each frame of the image into a behavior recognition model, and the behavior recognition model outputs the social behavior type of the two rats in each frame of the image, and the behavior recognition model is obtained by training a machine learning component with historical behavior features and corresponding social behavior types as samples. The scheme can improve the efficiency and accuracy of rat social behavior recognition.

The invention discloses a pig early disease abnormal behavior recognition system based on attitude angle trend and trajectory analysis of a low-power-consumption Bluetooth ear tag, and the system comprises an attitude angle collection terminal module, a trajectory behavior reconstruction module, a high-frequency sampling trigger module, an edge recognition and preprocessing module, and a fusion judgment and early warning output module. And a cloud platform collaboration module. The attitude angle acquisition terminal is used for acquiring three-axis attitude angle data (a yaw angle Yaw, a pitch angle Pitch and a roll angle Roll) of a pig and reducing energy consumption through an intermittent sampling mechanism; the trajectory reconstruction module constructs a virtual trajectory point string based on Bluetooth signals such as RSSI and the like, and is used for identifying spatial abnormal behavior characteristics; when the system recognizes a low-frequency attitude anomaly or a trajectory disturbance trend, a high-frequency sampling window is automatically triggered, and slight or short-time behavior changes such as local wandering, rotation behaviors and the like are further recognized; and finally, whether the pigs have behavior symptoms such as neurological diseases, abnormal breathing or highly pathogenic infectious diseases or not is judged through a fusion algorithm. The system has the capabilities of high recognition precision, low power consumption, long endurance and large-scale deployment, and can be widely applied to pig farm abnormal behavior early warning and animal health management scenes.

This invention discloses a method and system for quantifying cigarette consumer preferences based on multimodal fusion, belonging to the fields of data processing and artificial intelligence technology. The method includes: collecting voice data from consumers during smoking and behavioral observation data recorded in real-time by interviewers; processing the voice data to extract text features and acoustic emotion features; structurally encoding the behavioral observation data to generate a temporal behavioral feature sequence; and inputting the multimodal features, after time alignment, into a multimodal fusion analysis model based on an attention mechanism for fusion analysis, generating and outputting a structured quantitative index of consumer preferences. The system includes corresponding functional modules for implementing the above steps. This invention solves the problems of high cost, single insight dimension, and difficulty in quantifying conclusions associated with traditional methods. It can efficiently and accurately fuse consumers' subjective speech and objective behavioral responses in natural interview scenarios, outputting quantitative preference indicators that can directly guide cigarette product development, and has significant advantages such as low cost, high feasibility, and operable results.

To provide a system for improving the accuracy of future carrier selection, and for creating the sense of security of the entire society with little anxiety.SOLUTION: Generating and sending a questionnaire for collecting behavioral characteristics of the user to the user equipment, receiving an answer to the questionnaire sent by the user equipment and generating a behavioral characteristics profile of the user, generating and sending a game in a sugoroku form based on the generated behavioral characteristics profile, receiving a selection behavior and a piece progress of the user in the game in the sugoroku form, determining a next step based on the selection behavior and the piece progress by a AI, and sending a result to the user equipment, providing a time leap option when the game is cleared and simulating different choices, and generating and sending a report for presenting an improvement of the behavioral characteristics of the user based on a result of the simulation to the user equipment.SELECTED DRAWING: Figure 1

This disclosure provides a method, apparatus, and electronic device for information mining of online speech, relating to the field of artificial intelligence technology, specifically natural language processing, deep learning, and semantic analysis. The specific implementation scheme includes: acquiring a set of online speech samples as candidates, and mining the content semantic features and content temporal features of the online speech sample; acquiring the behavioral features of the candidate samples based on the online speech sample; predicting the probability of the candidate samples executing a target event based on the content semantic features, content temporal features, and behavioral features; and filtering out abnormal target samples from the candidate samples based on the probability of the candidate samples executing the target event. This disclosure improves the accuracy and reliability of identifying abnormal target samples by mining information from online speech, i.e., acquiring the content semantic features, content temporal features, and behavioral features of the online speech sample, and determining the probability of candidate samples executing a target event.

The invention belongs to the technical field of intelligent traffic, and discloses a driver workload prediction method based on multiple physiological signals under various hearing interferences, which is created by combining three music rhythms (no, slow and fast), two navigation modes (high frequency and low frequency) and three traffic scenes (conventional roads, school roads and construction site roads). Multi-modal data including EDA, ECG, EEG, driving behavior characteristics and subjective workload evaluation is collected, the data is used for constructing a workload prediction model, a stacking model is adopted, and the stacking model integrates a plurality of machine learning models to improve prediction performance.

The application discloses a motion posture normative evaluation method based on behavior feature recognition, relates to the technical field of motion posture evaluation, and comprises the following steps: obtaining key point trajectories by processing visual sensor data, extracting space-time features and combining the space-time features into feature vectors. The feature vectors are mapped to a high-dimensional semantic space by a pre-trained posture semantic encoder to obtain semantic codes containing deep motion patterns; three types of internal attributes, i.e., an interpretable motion rhythm, joint coordination and posture stability, are decoded from the codes. Distance measurement is performed on the attributes in the semantic space based on standard templates, an evaluation grade is determined according to rules, and finally, a structured evaluation report combining the grade and each sub-item value is generated. The method realizes a leap from appearance features to semantic understanding, and improves the accuracy of evaluation and the interpretability of results.

Provide information while considering the user's behavioral characteristics. [Solution] The output system responds to access from the user's information processing device to the user. Rows that retrieve behavioral characteristic information related to behavioral characteristics that indicate the nature of the user's actions. A dynamic characteristics acquisition unit, behavioral characteristics, and a display screen that displays the information provided to the user. Based on the correspondence between the two, the display image is displayed in a manner corresponding to the behavioral characteristics indicated by the behavioral characteristics information. It comprises a display information output unit that outputs first display information for displaying a surface on an information processing device. .

The invention discloses a crab stress recognition method and device, equipment, a medium and a product, and relates to the field of crab culture, and the method comprises the steps: obtaining a biological voiceprint signal of a crab, a vibration signal of a water body, and a behavior image of the crab; performing feature extraction according to the biological voiceprint signal to obtain a two-dimensional spectrogram and voiceprint behavior features; performing feature extraction according to the vibration signal to obtain a vibration fusion feature; performing behavior detection according to the behavior image to obtain an image behavior probability; performing cross-modal multi-head attention alignment according to the two-dimensional spectrogram and the behavior image to obtain a voiceprint image matching result; determining a behavior probability vector according to the voiceprint image matching result, the voiceprint behavior characteristics, the vibration fusion characteristics and the image behavior probability; and determining a stress index of the crabs according to the behavior probability vector, and regulating and controlling the environment according to the stress index. According to the invention, accurate stress identification can be realized, and accurate regulation and control can be realized in an early stage.

The invention discloses a multi-dimensional perception stress and fatigue state detection method and device, a medium and equipment, and the method comprises the steps: extracting an rPPG signal and a facial behavior as a facial behavior based on a facial video of a subject during a test period, and representing the facial behavior through a facial action unit, a head posture and a gazing direction; extracting audio features based on the audio file of the subject in the test period; and on the basis of the rPPG signal, the facial behavior and the audio feature of the subject, obtaining a global multi-modal feature tensor, inputting the global multi-modal feature tensor into a double-branch prediction model, and predicting the stress state probability and the fatigue state probability of the subject during the test period. By constructing a multi-task model framework based on multi-modal data input, multi-dimensional information of physiological features, behavior features and audio features is fully and effectively fused, meanwhile, complementary information between two tasks can be effectively captured based on synchronous detection of stress and fatigue, and the model performance and psychological state detection efficiency are improved.