A terminal data analysis system for smart campus management

By designing a terminal data analysis system, recording and analyzing students' reasons for going out and timestamp data, evaluating the regression time and setting thresholds, the problem of smart campus management system in judging abnormal behaviors of students leaving school is solved, and the safety of student management is improved.

CN118378221BActive Publication Date: 2024-11-01NANJING CHANGSHEN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410471263.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-18
Publication Date
2024-11-01
Estimated Expiration
2044-04-18

AI Technical Summary

Technical Problem

The smart campus management system lacks methods to analyze the reasons for students going out, which makes it difficult to judge abnormal behaviors of students leaving school.

Method used

Design a terminal data analysis system, including personnel database, terminal acquisition module, data analysis module, abnormality analysis module and exception processing module, and evaluate the regression time-stamp data by recording and analyzing students' reasons for going out and entering and leaving timestamp data, and set the regression threshold. Determine whether there is any abnormality and notify the student's tutor.

Benefits of technology

It has realized the supervision and abnormal handling of students' outdoor behavior, and improved the safety of student management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118378221B_ABST
    Figure CN118378221B_ABST
Patent Text Reader

Abstract

The present invention discloses a terminal data analysis system for smart campus management, which relates to the field of terminal data analysis management, including: the personnel database stores student information data; the terminal acquisition module records the reason for going out and the entry and exit timestamp data of each student; the data analysis module evaluates the regression time under each reason for going out; the abnormal analysis module determines whether the duration of the reason for going out of the student going out on that day exceeds the regression threshold, if not, it is determined that no processing is done, if so, it is determined whether there is an abnormality to be confirmed; the abnormality handling module is used to determine whether there is an abnormality to be confirmed, and when it is determined to be abnormal, obtain the contact information of the tutor of the class where the student going out is located, and notify the student managed by the tutor that an abnormality occurs. The advantages of the present invention are: it can realize the supervision and abnormality handling of students' outing behavior, and improve the safety of student management.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of terminal data analysis management, and in particular to a terminal data analysis system for smart campus management. Background Art

[0002] The terminal data analysis system for smart campus management is a system used to process and analyze student and faculty behavior data. It collects data through various terminal devices, such as access control records, attendance records, etc., and applies data cleaning, storage, analysis and visualization technologies to help school administrators understand student activities, optimize resource allocation and services, and improve education quality and student satisfaction.

[0003] At present, the smart campus management system mainly manages students on campus. For students who go out of school, due to the involvement of students' personal privacy, there is a lack of a set of regression features that analyze the reasons for students going out based on their historical behaviors, and a method to judge whether there is any abnormality for each student based on the regression features after analysis. Summary of the invention

[0004] In order to solve the above technical problems, a terminal data analysis system for smart campus management is provided. This technical solution solves the problem that the current smart campus management system mainly manages students on campus. For students who go out of school, due to the involvement of students' personal privacy, there is a lack of a set of regression features that analyze the reasons for students going out based on their historical behaviors, and judge whether there is any abnormality for each student based on the analyzed regression features.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A terminal data analysis system for smart campus management, comprising:

[0007] Personnel database: the personnel database is used to store the student information data in the current campus;

[0008] The terminal collection module is electrically connected to the personnel database. The terminal collection module is used to verify the identity information of each outgoing student based on the student information in the personnel database, and record the outgoing reason and entry and exit timestamp data of each student;

[0009] A data analysis module, which is electrically connected to the personnel database and the terminal acquisition module. The data analysis module is used to evaluate the regression time for each out-of-town reason based on the historical students' out-of-town reasons and entry and exit timestamp data;

[0010] The abnormality analysis module is electrically connected to the data analysis module. The abnormality analysis module is used to take the return time under each reason for going out as the return threshold, and determine whether the time for going out of the reason for the student going out on that day exceeds the return threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed.

[0011] The exception handling module is electrically connected to the exception analysis module. The exception handling module is used to determine whether there is an exception to be confirmed. When it is determined to be an exception, the contact information of the tutor of the class where the out-of-town student is located is obtained, and the student managed by the tutor is notified of the exception.

[0012] Optionally, the terminal acquisition module internally includes:

[0013] The out-of-school time recording unit is used to record the reason and time of each student leaving the school, which is recorded as the out-of-school timestamp;

[0014] The school entry time recording unit is used to record the reason for each student to go out and the time of entering the school, which is recorded as the school entry timestamp;

[0015] An entry-exit array unit, the entry-exit array unit is electrically connected to the school leaving time recording unit and the school entering time recording unit, and the entry-exit matrix unit is used to combine the school leaving time stamp and the school entering time stamp with the same reason for going out into a data array of entry-exit time stamps of several reasons for going out;

[0016] The in-out matrix unit is electrically connected to the in-out array unit. The in-out matrix unit is used to combine the in-out timestamp data arrays of several outgoing reasons into an in-out matrix C:

[0017]

[0018] Among them, C is the input and output matrix, is the jth in / out timestamp in the in / out timestamp data array of the i-th out / in reason in the in / out matrix, m is the total number of out / in reasons, and n is the total number of in / out timestamps.

[0019] Optionally, the data analysis module internally includes:

[0020] The time alignment unit is used to align the entry and exit timestamps of each outgoing reason in the entry and exit matrix, so that each outgoing timestamp is in the same time dimension, and obtain the regression time between the outgoing timestamp and the incoming timestamp of each outgoing reason;

[0021] A maximum duration marking unit, which is electrically connected to the time alignment unit, is used to mark the departure time of each reason for going out as the starting time, mark the last arrival timestamp as the final time, and obtain the maximum return duration of several reasons for going out;

[0022] A feature marking unit, the feature marking unit is electrically connected to the maximum time length marking unit, and the feature marking unit is used to perform regression marking on each school entry timestamp within the maximum regression time length based on the maximum regression time length of several reasons for going out;

[0023] A regression analysis model unit, the regression analysis model unit is electrically connected to the feature marking unit, and the regression analysis model unit is used to perform regression marking based on each school entry timestamp within the maximum regression duration, and analyze the student regression distribution to determine the time period with the most regression students at each moment;

[0024] The specific regression mark for each entry timestamp within the maximum regression duration is:

[0025] Based on the maximum regression duration, the first school entry timestamp at each moment within the maximum regression duration is marked as the starting school entry feature, and the last school entry timestamp at each moment is marked as the ending school entry feature, to obtain several regression feature data within the maximum regression duration;

[0026] The specific units of the regression analysis model are:

[0027] The maximum regression duration of several reasons for going out every day in nearly a week and several regression feature data within the maximum regression duration are used as the original data set;

[0028] The original data set is divided into a training set and a validation set, with the training set accounting for 80% of the total original data set and the validation set accounting for 20% of the total original data set;

[0029] Construct regression analysis model;

[0030] Based on the regression analysis model, the regression analysis model takes several regression feature data within the maximum regression time in the training set as input, the regression analysis model takes the school entry timestamp at each moment within the maximum regression time as input, takes the number of students regressed at each moment within the maximum regression time as output, and predicts the time when the school entry timestamp within the maximum regression time due to the reason of going out corresponds to the maximum number of regression people; takes minimizing the sum of prediction target errors as the training goal; trains the regression analysis model until the sum of prediction errors reaches convergence and stops training;

[0031] Among them, the regression analysis model expression is:

[0032]

[0033] In the formula, is the maximum regression time under the predicted reason for going out, and the time when the number of people returning corresponds to the t-th school entry timestamp, is the weight vector, For paranoid items, is the loss function, is the maximum regression time, is the actual value of the regression number corresponding to the t-th school admission timestamp.

[0034] Optionally, the abnormality analysis module includes:

[0035] A threshold division unit, the threshold division unit is used to take the return time with the most return students at each moment within the maximum return time of each reason for going out as the return threshold;

[0036] The abnormality judgment unit is electrically connected to the threshold division unit. The abnormality judgment unit is used to judge whether the duration of the student's reasons for going out on that day exceeds the regression threshold based on the regression threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed.

[0037] The specific determination of whether there is an abnormality to be confirmed is:

[0038] Determine the regression threshold of the reason for going out according to the reason for going out of the student on that day, set several abnormal judgments according to the regression threshold, and when the regression time exceeds the first regression threshold, the second regression threshold or the third regression threshold, send a safety confirmation message to the student's mobile terminal through the information terminal, and the student confirms that it is safe and there is no abnormality, that is, it is judged that the current student has no abnormality;

[0039] Students confirm that they are safe and have no abnormalities as follows:

[0040] When the return time exceeds the first return threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that the safety is normal. If yes, it is determined that there is no abnormality. If not, it is determined that there is an abnormality.

[0041] If the first regression threshold determines that there is no abnormality, when the regression time exceeds the second regression threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality, if not, it is determined that there is an abnormality;

[0042] If the second regression threshold determines that there is no abnormality, when the regression time exceeds the third regression threshold, a safety confirmation message is sent to the student's mobile terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality. If not, it is determined that there is an abnormality.

[0043] Optionally, the exception handling module includes:

[0044] The abnormality notification unit is used to obtain the contact information of the tutor of the class where the abnormal student is located, and notify the tutor that the student under his / her management has an abnormality;

[0045] The abnormality elimination unit is electrically connected to the abnormality notification unit. The abnormality elimination unit is used to determine whether to eliminate the abnormality based on the student's tutor's feedback. If yes, eliminate the abnormality judgment, otherwise do not do anything.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention proposes a terminal data analysis system for smart campus management, which obtains students' information data and records the reasons for students going out and entry and exit timestamp data. The data analysis module analyzes the return time of each student under different reasons for going out, and the exception analysis module is used to determine whether the student's current out-of-town time exceeds the return threshold to determine whether there is an abnormal situation. The exception handling module notifies the instructor of the student's class and informs the student of the abnormal situation. This solution uses technical means to achieve supervision and exception handling of students' out-of-town behavior. In this way: it is possible to achieve supervision and exception handling of students' out-of-town behavior and improve the security of student management. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A framework diagram of a terminal data analysis system for smart campus management according to the present invention;

[0049] Figure 2 This is an internal flow chart of the terminal acquisition module of the present invention;

[0050] Figure 3 It is the internal flow chart of the data analysis module of the present invention;

[0051] Figure 4 This is the internal flow chart of the abnormality analysis module of the present invention;

[0052] Figure 5 This is an internal flow chart of the exception handling module of the present invention. DETAILED DESCRIPTION

[0053] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0054] Reference Figure 1 As shown, a terminal data analysis system for smart campus management includes:

[0055] Personnel database: the personnel database is used to store the student information data in the current campus;

[0056] The terminal collection module is electrically connected to the personnel database. The terminal collection module is used to verify the identity information of each outgoing student based on the student information in the personnel database, and record the outgoing reason and entry and exit timestamp data of each student;

[0057] A data analysis module, which is electrically connected to the personnel database and the terminal acquisition module. The data analysis module is used to evaluate the regression time for each out-of-town reason based on the historical students' out-of-town reasons and entry and exit timestamp data;

[0058] The abnormality analysis module is electrically connected to the data analysis module. The abnormality analysis module is used to take the return time under each reason for going out as the return threshold, and determine whether the time for going out of the reason for the student going out on that day exceeds the return threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed.

[0059] The exception handling module is electrically connected to the exception analysis module. The exception handling module is used to determine whether there is an exception to be confirmed. When it is determined to be an exception, the contact information of the tutor of the class where the out-of-town student is located is obtained, and the student managed by the tutor is notified of the exception.

[0060] This solution obtains students' information data and records the reasons for students going out and entry and exit timestamp data. The data analysis module analyzes the return time of each student under different reasons for going out, and the exception analysis module is used to determine whether the student's current out-of-town time exceeds the return threshold to determine whether there is an abnormal situation. The exception handling module notifies the tutor of the student's class and informs the student of the abnormal situation. This solution uses technical means to achieve supervision and exception handling of students' out-of-town behavior. In this way: it can achieve supervision and exception handling of students' out-of-town behavior and improve the security of student management.

[0061] like Figure 2 As shown, the terminal acquisition module includes:

[0062] The out-of-school time recording unit is used to record the reason and time of each student leaving the school, which is recorded as the out-of-school timestamp;

[0063] The school entry time recording unit is used to record the reason for each student to go out and the time of entering the school, which is recorded as the school entry timestamp;

[0064] An entry-exit array unit, the entry-exit array unit is electrically connected to the school leaving time recording unit and the school entering time recording unit, and the entry-exit matrix unit is used to combine the school leaving time stamp and the school entering time stamp with the same reason for going out into a data array of entry-exit time stamps of several reasons for going out;

[0065] The in-out matrix unit is electrically connected to the in-out array unit. The in-out matrix unit is used to combine the in-out timestamp data arrays of several outgoing reasons into an in-out matrix C:

[0066]

[0067] Among them, C is the input and output matrix, is the jth in / out timestamp in the in / out timestamp data array of the i-th out / in reason in the in / out matrix, m is the total number of out / in reasons, and n is the total number of in / out timestamps.

[0068] This step records and counts the students' outings to facilitate management and monitoring of their actions. By recording the reasons for each student's outings and the timestamps of entering and leaving the school, we can clearly understand the time and reasons for the students' leaving and returning to school. By combining the timestamps of leaving and entering the school according to the reasons for going out, we can form an array of timestamps of several reasons for going out, which is convenient for analyzing and comparing the actions of students with different reasons for going out.

[0069] like Figure 3 As shown, the data analysis module includes:

[0070] The time alignment unit is used to align the entry and exit timestamps of each outgoing reason in the entry and exit matrix, so that each outgoing timestamp is in the same time dimension, and obtain the regression time between the outgoing timestamp and the incoming timestamp of each outgoing reason;

[0071] A maximum duration marking unit, which is electrically connected to the time alignment unit, is used to mark the departure time of each reason for going out as the starting time, mark the last arrival timestamp as the final time, and obtain the maximum return duration of several reasons for going out;

[0072] A feature marking unit, the feature marking unit is electrically connected to the maximum time length marking unit, and the feature marking unit is used to perform regression marking on each school entry timestamp within the maximum regression time length based on the maximum regression time length of several reasons for going out;

[0073] A regression analysis model unit, the regression analysis model unit is electrically connected to the feature marking unit, and the regression analysis model unit is used to perform regression marking based on each school entry timestamp within the maximum regression duration, and analyze the student regression distribution to determine the time period with the most regression students at each moment;

[0074] The specific regression mark for each entry timestamp within the maximum regression duration is:

[0075] Based on the maximum regression duration, the first school entry timestamp at each moment within the maximum regression duration is marked as the starting school entry feature, and the last school entry timestamp at each moment is marked as the ending school entry feature, to obtain several regression feature data within the maximum regression duration;

[0076] The specific units of the regression analysis model are:

[0077] The maximum regression duration of several reasons for going out every day in nearly a week and several regression feature data within the maximum regression duration are used as the original data set;

[0078] The original data set is divided into a training set and a validation set, with the training set accounting for 80% of the total original data set and the validation set accounting for 20% of the total original data set;

[0079] Construct regression analysis model;

[0080] Based on the regression analysis model, the regression analysis model takes several regression feature data within the maximum regression time in the training set as input, the regression analysis model takes the school entry timestamp at each moment within the maximum regression time as input, takes the number of students regressed at each moment within the maximum regression time as output, and predicts the time when the school entry timestamp within the maximum regression time due to the reason of going out corresponds to the maximum number of regression people; takes minimizing the sum of prediction target errors as the training goal; trains the regression analysis model until the sum of prediction errors reaches convergence and stops training;

[0081] Among them, the regression analysis model expression is:

[0082]

[0083] In the formula, is the maximum regression time under the predicted reason for going out, and the time when the number of people returning corresponds to the t-th school entry timestamp, is the weight vector, For paranoid items, is the loss function, is the maximum regression time, is the actual value of the regression number corresponding to the t-th school admission timestamp.

[0084] This solution can help school administrators understand students' departure and return situations and optimize resource allocation and activity arrangements by analyzing and regressing students' out-of-town behavior and marking them. Furthermore, its significance lies in supervising students' out-of-town behavior, improving safety management, predicting the number of returning students, and determining the time period with the most returning students at each moment, providing a data basis for subsequent analysis.

[0085] like Figure 4 As shown, the abnormal analysis module includes:

[0086] A threshold division unit, the threshold division unit is used to take the return time with the most return students at each moment within the maximum return time of each reason for going out as the return threshold;

[0087] The abnormality judgment unit is electrically connected to the threshold division unit. The abnormality judgment unit is used to judge whether the duration of the student's reasons for going out on that day exceeds the regression threshold based on the regression threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed.

[0088] The specific determination of whether there is an abnormality to be confirmed is:

[0089] Determine the regression threshold of the reason for going out according to the reason for going out of the student on that day, set several abnormal judgments according to the regression threshold, and when the regression time exceeds the first regression threshold, the second regression threshold or the third regression threshold, send a safety confirmation message to the student's mobile terminal through the information terminal, and the student confirms that it is safe and there is no abnormality, that is, it is judged that the current student has no abnormality;

[0090] Students confirm that they are safe and have no abnormalities as follows:

[0091] When the return time exceeds the first return threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that the safety is normal. If yes, it is determined that there is no abnormality. If not, it is determined that there is an abnormality.

[0092] If the first regression threshold determines that there is no abnormality, when the regression time exceeds the second regression threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality, if not, it is determined that there is an abnormality;

[0093] If the second regression threshold determines that there is no abnormality, when the regression time exceeds the third regression threshold, a safety confirmation message is sent to the student's mobile terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality. If not, it is determined that there is an abnormality.

[0094] This solution is to monitor students' outing behavior by determining the regression threshold of each reason for going out. When a student goes out, the system will determine whether to send a safety confirmation message based on the regression threshold. If the student confirms that it is safe and there is no abnormality, it will be judged as no abnormality; otherwise, it will be processed. Exception processing sets several abnormal judgments according to different regression thresholds, sends safety confirmation messages to students, and judges whether there is an abnormality based on the student's reply, thereby improving the safety level of student management.

[0095] like Figure 5 As shown, the exception handling module includes:

[0096] The abnormality notification unit is used to obtain the contact information of the tutor of the class where the abnormal student is located, and notify the tutor that the student under his / her management has an abnormality;

[0097] The abnormality elimination unit is electrically connected to the abnormality notification unit. The abnormality elimination unit is used to determine whether to eliminate the abnormality based on the student's tutor's feedback. If yes, eliminate the abnormality judgment, otherwise do not do anything.

[0098] The use process of a terminal data analysis system for smart campus management is as follows:

[0099] Step 1: Store student information data in the current campus;

[0100] Step 2: Record the reason and time of each student leaving school, and record it as the time stamp of leaving school;

[0101] Step 3: Record each student's reason for going out and time of entering school, and record it as the school entry timestamp;

[0102] Step 4: Combine the outgoing timestamps and incoming timestamps with the same outgoing reason into a data array of incoming and outgoing timestamps of several outgoing reasons;

[0103] Step 5: Combine the in-and-out timestamp data arrays of several out-and-out reasons into an in-and-out matrix C;

[0104] Step 6: Align the entry and exit timestamps of each reason for going out in the entry and exit matrix so that each departure timestamp is in the same time dimension, and obtain the regression time between the departure timestamp and the arrival timestamp of each reason for going out;

[0105] Step 7: Mark the departure time of each reason for going out as the starting time, mark the last arrival timestamp as the final time, and obtain the maximum regression duration of several reasons for going out;

[0106] Step 8: Based on the maximum regression time of several reasons for going out, each school entry timestamp within the maximum regression time is marked as regression;

[0107] Step 9: Mark each school admission timestamp within the maximum regression time for regression, and analyze the student regression distribution to determine the time period with the most regression students at each moment;

[0108] Step 10: Take the regression time with the most students returning at each moment within the maximum regression time of each reason for going out as the regression threshold;

[0109] Step 11: Based on the regression threshold, determine whether the duration of the students going out on that day exceeds the regression threshold. If not, determine not to do anything. If yes, determine whether there is an abnormality to be confirmed.

[0110] Step 12: Obtain the contact information of the tutor of the class where the abnormal student belongs, and notify the tutor that the student he manages has an abnormality;

[0111] Step 13: Determine whether the anomaly has been eliminated based on the student's instructor's feedback. If yes, eliminate the anomaly judgment. Otherwise, do nothing.

[0112] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A terminal data analysis system for smart campus management, characterized in that: include: Personnel database: the personnel database is used to store the student information data in the current campus; The terminal collection module is electrically connected to the personnel database. The terminal collection module is used to verify the identity information of each outgoing student based on the student information in the personnel database, and record the outgoing reason and entry and exit timestamp data of each student; A data analysis module, which is electrically connected to the personnel database and the terminal acquisition module. The data analysis module is used to evaluate the regression time for each out-of-town reason based on the historical students' out-of-town reasons and entry and exit timestamp data; The abnormality analysis module is electrically connected to the data analysis module. The abnormality analysis module is used to take the return time under each reason for going out as the return threshold, and determine whether the time for going out of the reason for the student going out on that day exceeds the return threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed. The exception handling module is electrically connected to the exception analysis module. The exception handling module is used to determine whether there is an exception to be confirmed. When it is determined to be an exception, the contact information of the tutor of the class where the out-of-town student is located is obtained, and the student managed by the tutor is notified of the abnormality. The data analysis module includes: The time alignment unit is used to align the entry and exit timestamps of each outgoing reason in the entry and exit matrix, so that each outgoing timestamp is in the same time dimension, and obtain the regression time between the outgoing timestamp and the incoming timestamp of each outgoing reason; A maximum duration marking unit, which is electrically connected to the time alignment unit, is used to mark the departure time of each reason for going out as the starting time, mark the last arrival timestamp as the final time, and obtain the maximum return duration of several reasons for going out; A feature marking unit, the feature marking unit is electrically connected to the maximum time length marking unit, and the feature marking unit is used to perform regression marking on each school entry timestamp within the maximum regression time length based on the maximum regression time length of several reasons for going out; A regression analysis model unit, the regression analysis model unit is electrically connected to the feature marking unit, and the regression analysis model unit is used to perform regression marking based on each school entry timestamp within the maximum regression duration, and analyze the student regression distribution to determine the time period with the most regression students at each moment; The specific regression marking for each school entry timestamp within the maximum regression duration is: Based on the maximum regression duration, the first school entry timestamp at each moment within the maximum regression duration is marked as the starting school entry feature, and the last school entry timestamp at each moment is marked as the ending school entry feature, to obtain several regression feature data within the maximum regression duration; The regression analysis model unit is specifically: The maximum regression duration of several reasons for going out every day in nearly a week and several regression feature data within the maximum regression duration are used as the original data set; The original data set is divided into a training set and a validation set, with the training set accounting for 80% of the total original data set and the validation set accounting for 20% of the total original data set; Construct regression analysis model; Based on the regression analysis model, the regression analysis model takes several regression feature data within the maximum regression time in the training set as input, the regression analysis model takes the school entry timestamp at each moment within the maximum regression time as input, takes the number of students regressed at each moment within the maximum regression time as output, and predicts the time when the school entry timestamp within the maximum regression time due to the reason of going out corresponds to the maximum number of regression people; takes minimizing the sum of prediction target errors as the training goal; trains the regression analysis model until the sum of prediction errors reaches convergence and stops training; Among them, the regression analysis model expression is: , where is the maximum regression time under the predicted reason for going out, and the time when the number of people returning is the largest corresponding to the t-th school entry timestamp. is the weight vector, For paranoid items, is the loss function, is the maximum regression time, is the actual value of the number of people regressed corresponding to the t-th school admission timestamp; The abnormal analysis module includes: A threshold division unit, the threshold division unit is used to take the return time with the most return students at each moment within the maximum return time of each reason for going out as the return threshold; The abnormality judgment unit is electrically connected to the threshold division unit. The abnormality judgment unit is used to judge whether the duration of the students' reasons for going out on that day exceeds the regression threshold based on the regression threshold. If not, it is determined that no processing is done. If so, it is determined whether there is an abnormality to be confirmed.

2. According to claim 1, a terminal data analysis system for smart campus management is characterized in that: The terminal acquisition module internally includes: The out-of-school time recording unit is used to record the reason and time of each student leaving the school, which is recorded as the out-of-school timestamp; The school entry time recording unit is used to record the reason for each student to go out and the time of entering the school, which is recorded as the school entry timestamp; An entry-exit array unit, the entry-exit array unit is electrically connected to the school leaving time recording unit and the school entering time recording unit, and the entry-exit matrix unit is used to combine the school leaving time stamp and the school entering time stamp with the same reason for going out into a data array of entry-exit time stamps of several reasons for going out; The in-out matrix unit is electrically connected to the in-out array unit. The in-out matrix unit is used to combine the in-out timestamp data arrays of several outgoing reasons into an in-out matrix C: , where C is the input and output matrix, is the jth in / out timestamp in the in / out timestamp data array of the i-th out / in reason in the in / out matrix, m is the total number of out / in reasons, and n is the total number of in / out timestamps.

3. According to claim 2, a terminal data analysis system for smart campus management is characterized in that: The determination of whether there is an abnormality to be confirmed is specifically as follows: The regression threshold of the reason for going out is determined based on the reason for the student going out that day, and several abnormal judgments are set based on the regression threshold. When the regression time exceeds the first regression threshold, the second regression threshold or the third regression threshold, a safety confirmation message is sent to the student's mobile terminal through the information terminal. The student confirms that it is safe and there is no abnormality, that is, it is judged that the current student has no abnormality.

4. According to claim 3, a terminal data analysis system for smart campus management is characterized in that: The students confirmed that they were safe and normal: When the return time exceeds the first return threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that the safety is normal. If yes, it is determined that there is no abnormality. If not, it is determined that there is an abnormality. If the first regression threshold determines that there is no abnormality, when the regression time exceeds the second regression threshold, a safety confirmation message is sent to the student's mobile phone terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality, if not, it is determined that there is an abnormality; If the second regression threshold determines that there is no abnormality, when the regression time exceeds the third regression threshold, a safety confirmation message is sent to the student's mobile terminal through the information terminal to determine whether the student confirms that it is safe and there is no abnormality. If so, it is determined that there is no abnormality. If not, it is determined that there is an abnormality.

5. According to claim 4, a terminal data analysis system for smart campus management is characterized in that: The exception handling module includes: The abnormality notification unit is used to obtain the contact information of the tutor of the class where the abnormal student is located, and notify the tutor that the student under his / her management has an abnormality; The abnormality elimination unit is electrically connected to the abnormality notification unit. The abnormality elimination unit is used to determine whether to eliminate the abnormality based on the student's tutor's feedback. If yes, eliminate the abnormality judgment, otherwise do not do anything.