Student behavior management method and system based on AI identification

Abstract

The invention provides a student behavior management method and system based on AI identification. The method comprises the following steps: acquiring a monitoring image of a target student, and performing AI identification on the monitoring image to obtain personal state data, character type data and social circle data corresponding to the target student; obtaining an in-school state trend of the target student according to the personal state data and the character type data; and when the school state trend is a descending trend, bullying source analysis is performed on the social circle data to generate a bullying traceability result, and the bullying traceability result is used for feeding back the students who have the social bullying behavior with the target student to the school. According to the method provided by the invention, the bullying person can be identified more accurately by analyzing the monitoring image.

Description

Technical Field

[0001] This application relates to the field of AI recognition technology, and in particular, to a method and system for managing student behavior based on AI recognition. Background Art

[0002] With the development of information technology, especially the popularization of social media platforms, the communication methods among teenagers have changed significantly. Social networks have not only become an indispensable part of their daily lives but also a place where "social bullying" occurs. Different from traditional physical or verbal bullying, social bullying is more manifested as spreading rumors, excluding others, posting insulting content, etc. in the online environment, and its characteristics of concealment and wide spread bring new challenges to school administrators.

[0003] To address this issue, related technologies have proposed using the data characteristics of social networks themselves to detect possible bullying behaviors. These methods generally include but are not limited to: text analysis, sentiment computing, social graph construction, etc.

[0004] However, in actual applications, the above methods still face many limitations. On the one hand, due to the high diversity of language expressions on social networks and often accompanied by regional cultural characteristics such as metaphors and slang, it is difficult to achieve accurate semantic understanding simply by relying on keyword matching; on the other hand, social bullying often occurs in specific situations, and not all negative emotions mean the occurrence of a bullying incident. Therefore, simple emotion indicators are not sufficient to accurately depict complex real situations.

[0005] Therefore, how to accurately identify social bullying is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0006] This application provides a method and system for managing student behavior based on AI recognition to solve the technical problem of inaccurate identification of social bullying.

[0007] In the first aspect, this application provides a method for managing student behavior based on AI recognition, including:

[0008] Obtaining the monitoring images of the target student, and performing AI recognition on the monitoring images to obtain the personal status data, personality type data, and social circle data corresponding to the target student;

[0009] Obtaining the on-campus status trend of the target student according to the personal status data and the personality type data;

[0010] When the on-campus status trend is a downward trend, analyzing the source of bullying for the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behaviors with the target student.

[0011] Optionally, the personal status data includes class status data and after-class status data. Based on the personal status data and the personality type data, an on-campus status trend is obtained, including:

[0012] Based on the class status data, a class status trend is obtained; based on the after-class status data, an after-class status trend is obtained;

[0013] Based on the personality type data, a target fusion method is determined among multiple fusion methods, where the multiple fusion methods can be obtained based on historical personal status data;

[0014] Based on the target fusion method, the class status trend and the after-class status trend are fused to obtain an on-campus status trend. Optionally, the social circle data includes social graph slices corresponding to multiple moments, and the social graph slice corresponding to any moment represents the social relationship of the target student at that moment;

[0015] Analyzing the source of bullying in the social circle data to generate a bullying traceability result includes:

[0016] Obtaining the moment when an anomaly occurs, where the moment when an anomaly occurs can be obtained based on the on-campus status trend;

[0017] Based on the moment when an anomaly occurs, a target graph slice is determined among the social graph slices corresponding to the multiple moments, where the target graph slice is the social graph slice corresponding to the moment when an anomaly occurs;

[0018] Performing a comparative analysis on the target graph slice to obtain a bullying traceability result.

[0019] Optionally, the obtaining the moment when an anomaly occurs includes:

[0020] Determining a target inflection point corresponding to the on-campus status trend, where the target inflection point is the point where the on-campus status trend starts to change;

[0021] Analyzing the target inflection point to obtain the inflection moment corresponding to the target inflection point, and using the inflection moment as the moment when an anomaly occurs.

[0022] Optionally, after analyzing the source of bullying in the social circle data to generate a bullying traceability result, it further includes:

[0023] Generating a student bullying warning instruction based on the bullying traceability result corresponding to the target student, where the student bullying warning instruction is used to prompt the teacher that the target student is being socially bullied.

[0024] Second aspect, the present application provides a student behavior management system based on AI recognition for executing the method described in any item of the first aspect. The above system includes:

[0025] A processing module, configured to obtain a monitoring image of a target student, and perform AI recognition on the monitoring image to obtain personal status data, personality type data, and social circle data corresponding to the target student;

[0026] An analysis module, configured to obtain the on-campus status trend of the target student according to the personal status data and the personality type data;

[0027] The analysis module is further configured to: when the on-campus status trend is a downward trend, perform a bullying source analysis on the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behavior with the target student.

[0028] For the student behavior management method based on AI recognition provided by the present application, compared with obtaining the student status by analyzing social network data, some students may not use social networks. The present application determines the on-campus status of each student on campus by obtaining the monitoring image of the target student, and performs AI recognition on the monitoring image to obtain personal status data, personality type data, and social circle data corresponding to the target student; determines the change trend of the on-campus status of the target student according to the personal status data and the personality type data; when the change trend of the on-campus status is a downward trend, it indicates that the target student may be socially bullied; at this time, perform a bullying source analysis on the social circle data to generate a bullying traceability result to determine the students who socially bully the target student, and feedback the bullying traceability result to the school, so as to manage the student bullying behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0030] Figure 1 It is a schematic diagram of an application scenario of a student behavior management method based on AI recognition provided by an embodiment of the present application;

[0031] Figure 2 It is a schematic flowchart of a student behavior management method based on AI recognition provided by an embodiment of the present application;

[0032] Figure 3 It is a schematic flowchart of another student behavior management method based on AI recognition provided by an embodiment of the present application;

[0033] Figure 4Schematic structural diagram of a student behavior management system based on AI recognition provided by an embodiment of the present application;

[0034] Figure 5 Schematic structural diagram of an electronic device provided by the present application.

[0035] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Specific embodiments

[0036] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of systems and methods consistent with some aspects of the present application as detailed in the appended claims.

[0037] Related technologies have proposed various solutions to attempt to detect possible bullying behaviors by leveraging the data characteristics of social networks themselves. These methods generally include, but are not limited to:

[0038] Text analysis: By performing natural language processing (NLP) on the text information posted on social platforms to identify features such as negative emotion words and aggressive language, so as to determine whether there are signs of bullying.

[0039] Affective computing: Combining psychological theories and using machine learning algorithms to quantitatively evaluate the emotional tendencies expressed by users as a warning signal.

[0040] Social graph construction: Based on social relationships, establish an interaction model among users, analyze the structural changes within the group and the rise and fall of individual status to discover abnormal interpersonal interaction patterns.

[0041] Multimodal fusion: Integrate information from different sources, such as multimedia materials like comments, private message records, pictures, and videos, to provide a more comprehensive behavioral portrait.

[0042] However, in practical applications, the above methods still face many limitations. On the one hand, due to the high diversity of language expressions on social networks and often accompanied by regional cultural characteristics such as metaphors and slang, relying solely on keyword matching is difficult to achieve accurate semantic understanding; on the other hand, social bullying often occurs in specific situations, and not all negative emotions mean the occurrence of a bullying incident. Therefore, simple emotion indicators are not sufficient to accurately depict complex real-world situations.

[0043] In summary, although existing social network analysis techniques provide preliminary means for identifying social bullying, they still have obvious deficiencies in terms of accuracy, especially in complex and ever-changing real-world scenarios, which limits their effectiveness and reliability. Therefore, there is an urgent need for a more comprehensive method that can accurately capture real bullying behaviors in a timely manner even when some students do not use social networks, and then take appropriate intervention measures to ensure the physical and mental health of each student.

[0044] A student behavior management method based on AI recognition provided by an embodiment of this application aims to solve the above technical problems of the prior art.

[0045] As Figure 1 shown, Figure 1 FIG. is a schematic diagram of an application scenario of a student behavior management method based on AI recognition provided by this application. An application scenario of a student behavior management method based on AI recognition provided by this application is as follows: a camera device collects campus surveillance images and sends the surveillance images to an electronic device; the electronic device analyzes the surveillance images to obtain a social bullying analysis result corresponding to a student at school. When the social bullying analysis result indicates that the student is suffering from social bullying, a warning instruction is generated and sent to a teacher terminal to prompt the teacher that the student is suffering from social bullying.

[0046] It should be noted that a student behavior management method based on AI recognition provided by this application is executed by the above-mentioned electronic device. Among them, the electronic device can be a wireless terminal or a wired terminal. A wireless terminal can be a device that provides voice and / or other service data connectivity to users, a handheld device with wireless connection function, or other processing devices connected to a wireless modem. The wireless terminal can communicate with one or more core network devices via a Radio Access Network (RAN). The wireless terminal can be a mobile terminal, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal. For example, it can be a portable, pocket-sized, handheld, computer-integrated or vehicle-mounted mobile system that exchanges language and / or data with the wireless access network. For another example, the wireless terminal can also be a Personal Communication Service (PCS) phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), and other devices. The wireless terminal can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile, a remote station, a remote terminal, an access terminal, a user terminal, a user agent, a user device or user equipment, which is not limited herein. Optionally, the above terminal device can also be a smart watch, a tablet computer and other devices.

[0047] The technical solution of this application and how the technical solution of this application solves the above technical problems will be described in detail below with specific embodiments. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0048] As Figure 2 shown, a student behavior management method based on AI recognition may specifically include steps S201 to S204, where:

[0049] S201. Obtain the monitoring images of the target student, and perform AI recognition on the monitoring images to obtain the personal status data, personality type data, and social circle data corresponding to the target student.

[0050] Preprocess the monitoring images to obtain class images and after-class images, where the monitoring images carry time identifiers.

[0051] It should be noted that the class images represent the monitoring images during class time, and the after-class images represent the monitoring images during after-class time. The monitoring images are not limited to indoor or outdoor, and are only determined by the time identifier.

[0052] Preprocess the monitoring images to obtain class images and after-class images, including: the electronic device receives the monitoring images from the camera device in real time. The monitoring images include image slices corresponding to multiple time periods, and the above multiple time periods can be obtained based on the class time of the class; perform AI face recognition on the image slices to identify the student identifier corresponding to the image slice, and the student identifier represents the student who appears in the image slice; for each image slice, integrate the image slice and the student identifier corresponding to the image slice to generate the slice data corresponding to the image slice, and the slice data corresponds to the time period one by one; group the image slices according to the class time and the above multiple time periods to obtain class images and after-class images.

[0053] AI recognition includes AI mood recognition, AI personality recognition, and AI social graph analysis.

[0054] Perform AI mood recognition on the monitoring images to obtain the personal status data corresponding to the target student, including: perform AI mood recognition on the class images to obtain class status data; perform AI mood recognition on the after-class images to obtain after-class status data. Among them, the status data represents the mood parameters of the target student, and the better the mood, the higher the mood parameter.

[0055] Perform AI personality recognition on the monitoring images to obtain the personality type data corresponding to the target student. Among them, AI personality recognition can be implemented based on psychological theory models such as the Big Five Personality Traits or the 16PF (16 Personality Factors).

[0056] Perform AI social graph analysis on the monitoring images to obtain the social graph corresponding to the target student. Among them, the social graph is updated over time.

[0057] S202. Obtain the in-school status trend of the target student based on the personal status data and personality type data.

[0058] Specifically, curve fitting is performed based on the personal status data to obtain a status distribution curve; trend analysis is performed on the status distribution curve according to the personality type data to obtain the distribution trend corresponding to the status distribution curve.

[0059] Among them, the distribution trend is the change trend of the on-campus status of the target student, that is, the distribution trend is equivalent to the on-campus status trend.

[0060] For example, the Holt method / Holt-Winters method, time series decomposition method or differencing method is used to perform trend analysis on the status distribution curve. Among them, the distribution trend is a downward trend, a stable trend or an upward trend.

[0061] When the distribution trend is a downward trend, it indicates that the on-campus status of the target student deteriorates, and at this time, the probability that the target student is socially bullied is relatively high; when the distribution trend is a stable trend or an upward trend, it indicates that the on-campus status of the target student maintains the original level or even becomes better, and at this time, the probability of social bullying occurring to the target student is relatively small.

[0062] S203. When the distribution trend is a downward trend, perform bullying source analysis on the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behavior with the target student.

[0063] Specifically, obtain the start time point corresponding to the downward trend; determine the transformed social circle data corresponding to the start time point, and obtain the previous social circle data of the above-mentioned transformed social circle data; compare the previous social circle data and the transformed social circle data to determine the list of personnel changes, where the list of personnel changes includes the names of students who newly appear in the transformed social circle data compared with the previous social circle data.

[0064] The student behavior management method based on AI recognition provided by the embodiments of the present application, compared with obtaining the student status through social network data analysis, some students may not use social networks. The present application determines the on-campus status of each student on campus by obtaining the monitoring images of the target student, and performs AI recognition on the monitoring images to obtain the personal status data, personality type data and social circle data corresponding to the target student; according to the personal status data, perform curve fitting to obtain a status distribution curve; according to the personality type data, perform trend analysis on the status distribution curve to obtain the distribution trend corresponding to the status distribution curve, so as to determine the change trend of the on-campus status of the target student; when the change trend of the on-campus status is a downward trend, it indicates that the target student may be socially bullied; at this time, perform bullying source analysis on the social circle data to generate a bullying traceability result to determine the students who socially bully the target student.

[0065] In one possible implementation, the personal status data includes in-class status data and after-class status data, and the status distribution curve includes an in-class status curve and an after-class status curve. Obtaining the on-campus status trend based on the personal status data and the personality type data may specifically include: a pre-fusion method or a post-fusion method.

[0066] In one possible implementation, the post-fusion method includes: obtaining the in-class status trend according to the in-class status data; obtaining the after-class status trend according to the after-class status data; determining a target fusion method from multiple fusion methods according to the personality type data, where the multiple fusion methods can be obtained based on historical personal status data; and fusing the in-class status trend and the after-class status trend according to the target fusion method to obtain the on-campus status trend. In another possible implementation, the pre-fusion method includes: performing curve fitting on the personal status data to obtain a status distribution curve; performing trend analysis on the status distribution curve according to the personality type data to obtain the distribution trend corresponding to the status distribution curve; fusing the in-class status trend and the after-class status trend according to the target fusion method to obtain a fused status curve; and performing trend analysis on the fused status curve to obtain the on-campus status trend.

[0067] Further, performing curve fitting on the in-class status data to obtain an in-class status curve includes:

[0068] The in-class status curve is the same as the in-class status curve.

[0069] Preprocessing the in-class status data to remove noise and outliers to obtain available in-class status data, ensuring the accuracy and reliability of the available in-class status data; performing curve fitting on the available in-class status data based on fitting methods such as linear regression, polynomial regression, or spline interpolation; and through this curve fitting process, generating an in-class status curve that can intuitively reflect the status change trend of students during different time periods in class.

[0070] Further, performing curve fitting on the after-class status data to obtain an after-class status curve includes:

[0071] The after-class status curve is the same as the after-class status curve.

[0072] The fitting process of the after-class status curve is the same as that of the in-class status curve, only replacing the in-class status data with the after-class status data. For the specific implementation method, refer to S202-1, which will not be elaborated in this embodiment. Further, performing trend analysis on the status distribution curve according to the personality type data to obtain the distribution trend corresponding to the status distribution curve may specifically include:

[0073] Determine a target fusion method among multiple fusion methods according to personality type data, where the multiple fusion methods can be obtained based on historical personal status data.

[0074] It can be understood that students with different personalities have different school performances under the same state. Therefore, in this embodiment, the fusion method suitable for the student is selected through the personality type data of the target student, which can ensure that there is less data loss in the state distribution curve before and after fusion.

[0075] Among them, the fusion methods include but are not limited to linear combination, polynomial fitting, or spline interpolation. When the student's personality remains unchanged, the target fusion method is determined to be a linear combination; when the student's personality changes slightly, the target fusion method is determined to be spline interpolation; when the student's personality changes significantly, the target fusion method is determined to be polynomial fitting.

[0076] Fuse the class state curve and the after-class state curve according to the target fusion method to obtain a fused state curve.

[0077] Specifically, standardize the class state curve and the after-class state curve to ensure that they are fused on the same time scale and state scale; based on the target fusion method, control the contribution ratio of each curve in the fusion and ensure that their sum is 1; apply the target fusion method to each time point and calculate the fused state curve.

[0078] Furthermore, to eliminate the influence of short-term fluctuations, the fused curve can be smoothed, and the finally generated fused state curve can comprehensively reflect the state changes of the student during the entire class period.

[0079] Conduct trend analysis on the fused state curve to obtain the distribution trend.

[0080] Specifically, determine whether there is an inflection point on the fused state curve. If not, determine the distribution trend to be a stable trend; if so, it indicates that the school state of the target student has increased or decreased, and it is necessary to further confirm its distribution trend. At this time, based on the smoothed situation or time series analysis method, determine the distribution trend to be an upward trend or a downward trend.

[0081] In a realizable manner, the social circle data includes social graph slices corresponding to multiple moments respectively, and the social graph slice corresponding to any moment represents the social relationship of the target student at any moment. The above-mentioned analysis of the source of bullying in the social circle data to generate a bullying traceability result can specifically include:

[0082] Obtain the moment when the anomaly appears, where the moment when the anomaly appears can be obtained according to the school state trend.

[0083] The moment when the anomaly appears is the moment corresponding to the above-mentioned inflection point.

[0084] Specifically, determine the target inflection point corresponding to the state distribution curve, where the target inflection point is the point where the distribution trend starts to change; analyze the target inflection point to obtain the inflection point moment corresponding to the target inflection point, and use the inflection point moment as the abnormal occurrence moment.

[0085] According to the abnormal occurrence moment, in the social graph slices corresponding to multiple moments, determine the target graph slice, where the target graph slice is the social graph slice corresponding to the abnormal occurrence moment.

[0086] Conduct a comparative analysis on the target graph slice to obtain the bullying traceability result.

[0087] Specifically, extract features from each target graph slice to obtain slice features, where the slice features include the connection strength of nodes, the interaction frequency between nodes, and the centrality index of key nodes, etc.; use graph theory analysis techniques to conduct a comparative analysis on each target graph slice to identify abnormal patterns or changing trends that appear in the time series or event sequence, where the abnormal patterns include but are not limited to a sudden increase in the number of social personnel, a sudden decrease in the number of social personnel, etc., and the changing trends include but are not limited to the appearance of social personnel with non-overlapping social circles. Thus, through comparative analysis, potential social bullying behaviors can be identified to obtain an analysis result, and the analysis result includes key participants and event trigger points; based on the above analysis result, generate a bullying traceability result to provide detailed information about the cause, development path, and main participants of the bullying event.

[0088] In one realizable manner, after generating the bullying traceability result by analyzing the bullying source of the social circle data in S203, it further includes:

[0089] Generate a student bullying warning instruction according to the bullying traceability result corresponding to the target student, where the student bullying warning instruction is used to prompt the teacher that the target student is socially bullied.

[0090] Specifically, first, extract key information related to the target student from the bullying traceability result, and the key information includes other students participating in the bullying, the frequency and severity of the event, etc.; then, use the determination rules corresponding to each key information to evaluate the above key information respectively, such as detecting whether there is high-frequency negative interaction or abnormal social network structure features; based on the evaluation result, trigger the condition for generating the student bullying warning instruction, where when the above condition is met, an early warning instruction containing specific warning content will be automatically generated, and the instruction details the bullying risks faced by the target student and recommends specific intervention measures for the teacher.

[0091] Furthermore, the warning instruction is immediately sent to relevant teachers through the school's internal communication system or the teacher's dedicated platform to ensure that they can quickly understand the situation and provide necessary support and protection.

[0092] As Figure 3 shown, this application provides another specific implementation manner to execute the above-mentioned student behavior management method based on AI recognition, which specifically includes:

[0093] S301. Obtain the monitoring video of the target student, and perform AI recognition on the monitoring video to obtain the personal status data, personality type data, and social circle data corresponding to the target student.

[0094] S302. Obtain the in-class status data and after-class status data according to the personal status data.

[0095] S303. Perform curve fitting according to the in-class status data to obtain the in-class status curve; and perform curve fitting according to the after-class status data to obtain the after-class status curve.

[0096] S304. Determine the target fusion method among multiple fusion methods according to the personality type data, where the multiple fusion methods can be obtained based on historical personal status data.

[0097] S305. Fusion the in-class status curve and the after-class status curve according to the target fusion method to obtain the fusion status curve.

[0098] S306. Perform trend analysis on the fusion status curve to obtain the distribution trend.

[0099] S307. When the distribution trend is a downward trend, determine the target inflection point corresponding to the status distribution curve, where the target inflection point is the point where the distribution trend starts to change.

[0100] S308. Analyze the target inflection point to obtain the inflection point moment corresponding to the target inflection point, and use the inflection point moment as the abnormal occurrence moment;

[0101] S309. Obtain the social graph slices corresponding to multiple moments according to the social circle data; and determine the target graph slice among the social graph slices corresponding to the above multiple moments, where the target graph slice is the social graph slice corresponding to the abnormal occurrence moment.

[0102] S310. Perform comparative analysis on the target graph slice to obtain the bullying traceability result, where the bullying traceability result represents the students who have social bullying behavior with the target student.

[0103] Furthermore, to remind teachers of the possible existence of social bullying, after S310, it further includes:

[0104] S311. Generate a student bullying warning instruction based on the bullying traceability result corresponding to the target student. The student bullying warning instruction is used to prompt the teacher that the target student has been socially bullied.

[0105] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be in other sequences or performed simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0106] Furthermore, it should be noted that although the steps in the flowchart are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0107] Figure 4 It is a schematic structural diagram of a student behavior management system based on AI recognition provided by an embodiment of the present application. As Figure 4 shown, the student behavior management system 40 based on AI recognition provided by an embodiment of the present application includes:

[0108] An analysis module 401, configured to obtain the on-campus status trend of the target student according to the personal status data and the personality type data;

[0109] The analysis module 402 is further configured to: when the on-campus status trend is a downward trend, perform a bullying source analysis on the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behaviors with the target student.

[0110] In a possible implementation manner, the personal status data includes class status data and after-class status data. When the above analysis module 402 executes to obtain the on-campus status trend according to the personal status data and the personality type data, it is configured to:

[0111] Obtain the class status trend according to the class status data; obtain the after-class status trend according to the after-class status data;

[0112] Determine a target fusion method among multiple fusion methods according to personality type data, where the multiple fusion methods can be obtained based on historical personal status data;

[0113] Fuse the in-class status trend and the after-class status trend according to the target fusion method to obtain the on-campus status trend.

[0114] In a possible implementation, the social circle data includes social graph slices corresponding to multiple moments respectively, and the social graph slice corresponding to any moment represents the social relationship of the target student at any moment. When the above analysis module 402 performs the analysis of the bullying source on the social circle data to generate the bullying traceability result, it is used for:

[0115] Obtain the moment when the anomaly occurs, where the moment when the anomaly occurs can be obtained according to the on-campus status trend;

[0116] Determine the target graph slice among the social graph slices corresponding to multiple moments according to the moment when the anomaly occurs, where the target graph slice is the social graph slice corresponding to the moment when the anomaly occurs;

[0117] Perform a comparative analysis on the target graph slice to obtain the bullying traceability result.

[0118] In a possible implementation, when the above analysis module 402 performs the operation of obtaining the moment when the anomaly occurs, it is used for:

[0119] Determine the target inflection point corresponding to the on-campus status trend, where the target inflection point is the point where the on-campus status trend starts to change;

[0120] Analyze the target inflection point to obtain the inflection point moment corresponding to the target inflection point, and use the inflection point moment as the moment when the anomaly occurs.

[0121] In a possible implementation, the above system further includes:

[0122] An early warning module, which is used to generate a student bullying early warning instruction according to the bullying traceability result corresponding to the target student, where the student bullying early warning instruction is used to prompt the teacher that the target student is socially bullied.

[0123] The student behavior management system based on AI recognition provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0124] It should be understood that the above system embodiments are merely illustrative, and the system of the present application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0125] In addition, without special description, each functional unit / module in the various embodiments of the present application can be integrated in one unit / module, or each unit / module can exist physically alone, or two or more units / modules can be integrated together. The above integrated unit / module can be implemented in the form of hardware or in the form of a software program module.

[0126] When the integrated unit / module is implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc.

[0127] Figure 5 It is a schematic structural diagram of the electronic device provided by the present application. As Figure 5 shown, the electronic device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the electronic device 50 further includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus 504.

[0128] In the specific implementation process, at least one processor 501 executes the computer execution instructions stored in the memory 502, so that at least one processor 501 executes the above method.

[0129] For the specific implementation process of the processor 501, reference can be made to the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0130] Unless otherwise specified, the processor 501 may be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, ASIC, etc. Unless otherwise specified, the memory 502 may be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0131] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in various embodiments of this application. The aforementioned memory includes: USB flash drive, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disc and other media that can store program codes.

[0132] The embodiments of this application also provide a computer-readable storage medium. Computer-executable instructions are stored in the computer-readable storage medium. When the processor executes the computer-executable instructions, the redundant field update method for a distributed system as described above is implemented.

[0133] The embodiments of this application also provide a computer program product, including a computer program. When the computer program is executed by the processor, the redundant field update method for a distributed system as described above is implemented.

[0134] In the above embodiments, the descriptions of the various embodiments each have their own emphasis. For parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as falling within the scope described in this specification.

[0135] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0136] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A student behavior management method based on AI recognition, characterized in that, Including: Obtain the surveillance video of the target student, and perform AI recognition on the surveillance video to obtain the personal status data, personality type data, and social circle data corresponding to the target student; Based on the personal status data and the personality type data, obtain the on-campus status trend of the target student; When the on-campus status trend is a downward trend, perform a bullying source analysis on the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behavior with the target student.

2. The method for managing student behavior based on AI recognition according to claim 1, wherein The personal status data includes in-class status data and after-class status data. Based on the personal status data and the personality type data, obtaining the on-campus status trend includes: Based on the in-class status data, obtain the in-class status trend; based on the after-class status data, obtain the after-class status trend; Based on the personality type data, determine a target fusion method among multiple fusion methods, where the multiple fusion methods can be obtained based on historical personal status data; Based on the target fusion method, fuse the in-class status trend and the after-class status trend to obtain the on-campus status trend.

3. The method for managing student behavior based on AI recognition according to claim 1, wherein The social circle data includes social graph slices corresponding to multiple moments respectively, and the social graph slice corresponding to any moment represents the social relationship of the target student at the any moment; The performing a bullying source analysis on the social circle data to generate a bullying traceability result includes: Obtain the abnormal occurrence moment, where the abnormal occurrence moment can be obtained based on the on-campus status trend; Based on the abnormal occurrence moment, determine a target graph slice among the social graph slices corresponding to the multiple moments, where the target graph slice is the social graph slice corresponding to the abnormal occurrence moment; Perform a comparative analysis on the target graph slice to obtain a bullying traceability result.

4. The method for managing student behavior based on AI recognition according to claim 3, wherein, The obtaining the abnormal occurrence moment includes: Determine the target inflection point corresponding to the on-campus status trend, where the target inflection point is the point where the on-campus status trend starts to change; Analyze the target inflection point to obtain the inflection point moment corresponding to the target inflection point, and use the inflection point moment as the abnormal occurrence moment.

5. The student behavior management method based on AI recognition according to claim 1, characterized in that After performing the bullying source analysis on the social circle data to generate a bullying traceability result, it further includes: Generate a student bullying warning instruction based on the bullying traceability result corresponding to the target student, where the student bullying warning instruction is used to prompt the teacher that the target student is socially bullied.

6. A student behavior management system based on AI recognition, which is applied to the method for managing student behavior based on AI recognition according to any one of claims 1-5, and is characterized in that, Including: A processing module, configured to obtain the surveillance video of the target student, and perform AI recognition on the surveillance video to obtain the personal status data, personality type data, and social circle data corresponding to the target student; An analysis module, configured to obtain the on-campus status trend of the target student based on the personal status data and the personality type data; The analysis module is further configured to: when the on-campus status trend is a downward trend, perform a bullying source analysis on the social circle data to generate a bullying traceability result, where the bullying traceability result is used to feedback to the school the students who have social bullying behavior with the target student.

Images