Student privacy data protection method based on artificial intelligence

Abstract

The invention relates to a student privacy data protection method based on artificial intelligence, and the method comprises the steps: configuring a plurality of databases which are used for storing student privacy data of different privacy levels, and each database is configured with a corresponding access level; obtaining student privacy data, and carrying out privacy grade division on the student privacy data; storing the divided student privacy data into a database of a corresponding privacy level; responding to an access request of a user, acquiring user information of the user, and judging whether the access permission of the user is higher than or equal to the access level of the accessed database; if yes, calling corresponding student privacy data; if not, refusing to access; according to the method, data are classified in an artificial intelligence mode, leakage of student privacy data possibly caused by manual data classification is avoided, and due to the fact that the artificial intelligence mode is adopted for data classification, the classification accuracy is higher than that of manual classification, and the classification efficiency is higher.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and in particular to a student privacy data protection method based on artificial intelligence. Background Art

[0002] With the rapid development of electronic technology and multimedia technology, schools use more online systems to collect student information. Among the collected information, there is no lack of student privacy data. Under normal circumstances, the online system grants corresponding permissions to specific groups to avoid bypassing the level of access to obtain student privacy data. However, in the step of collecting information, the existing technology usually requires the designated person to classify and enter the information into the online system. When entering the information into the system, the designated person has enough time to collect the above student privacy data privately; at the same time, the above authorization process does not make special data. When encountering malicious theft of high-authority users' information and obtaining data with corresponding permissions, student privacy data cannot be well protected. Therefore, the current protection of student privacy data is still not in place. Summary of the invention

[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present disclosure is to provide a student privacy data protection method based on artificial intelligence to solve the problem that the prior art cannot protect student privacy well.

[0004] The present invention discloses a method for protecting student privacy data based on artificial intelligence, comprising the following steps:

[0005] Database configuration: configuring a number of databases, wherein the databases are used to store student privacy data of different privacy levels, and each of the databases is respectively configured with a corresponding access level;

[0006] Privacy level classification: obtaining student privacy data and classifying the student privacy data into privacy levels;

[0007] Privacy data storage: store the divided student privacy data in a database with corresponding privacy levels;

[0008] Access call verification: In response to the user's access request, obtain the user's user information and determine whether the user's access rights are not lower than the access level of the accessed database; if so, call the corresponding student privacy data; if not, deny access.

[0009] Preferably, the privacy level classification specifically includes the following steps:

[0010] Establish the corresponding relationship between keywords and privacy levels;

[0011] The student privacy data is obtained, keywords in the student privacy data are extracted by using a text extraction model based on CRNN, and the privacy level of the student privacy data is divided according to the privacy level to which the keywords belong.

[0012] Preferably, in the access call verification:

[0013] Before calling the corresponding student privacy data, determine whether the user information of the user matches the student privacy data. If so, directly call the student privacy data; if not, perform corresponding desensitization processing on the called student privacy data.

[0014] Preferably, it is determined whether there is an attribution relationship between the user information of the user and the student privacy data. If so, the user information of the user matches the student privacy data; if not, the user information of the user does not match the student privacy data.

[0015] Preferably, the access call verification also includes:

[0016] A student privacy data call frequency threshold is preset, the frequency of the user calling the student privacy data within time t is monitored, and it is determined whether the frequency of the user calling the student privacy data is not less than the student privacy data call frequency threshold. If so, an alarm signal is issued and the user's access rights are reduced; if not, the monitoring operation is maintained.

[0017] Preferably, after determining that the frequency of a user calling student privacy data is not less than the student privacy data calling frequency threshold, the correlation between the student privacy data called by the user within time t is monitored, and the correlation is clustered according to at least one common indicator of the student privacy data to obtain a clustering result; a quantity threshold of the clustering results is preset to determine whether the quantity of the clustering results is greater than the quantity threshold, and if so, the user's access permission is disabled; if not, the monitoring operation is maintained.

[0018] The present disclosure also provides a student privacy data protection system, which includes the student privacy data protection method based on artificial intelligence as described above.

[0019] The present disclosure also provides a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the student privacy data protection method based on artificial intelligence as described above.

[0020] The advantages of the student privacy data protection method based on artificial intelligence disclosed in the present invention are:

[0021] By using artificial intelligence to classify data, we can avoid the leakage of student privacy data that may be caused by manual classification of data. In addition, since artificial intelligence is used to classify data, the classification accuracy is higher than that of manual classification, and the classification efficiency is faster;

[0022] At the same time, by verifying the user information of the user who is granted access rights and monitoring the user's search frequency and the relevance of the search content, when an abnormal situation is found, the user's access rights are promptly reduced or disabled to avoid the leakage of student privacy data, thereby achieving the purpose of protecting student privacy data. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of a student privacy data protection method based on artificial intelligence described in the present invention. DETAILED DESCRIPTION

[0024] like Figure 1 As shown, the student privacy data protection method based on artificial intelligence described in the present disclosure includes the following steps:

[0025] Database configuration: configure several databases, which are used to store student privacy data of different privacy levels, and each database is configured with a corresponding access level;

[0026] Privacy level classification: obtain students' privacy data and classify the privacy data into privacy levels;

[0027] Privacy data storage: store the divided student privacy data in a database with corresponding privacy levels;

[0028] Access call verification: In response to the user's access request, obtain the user's user information and determine whether the user's access rights are not lower than the access level of the accessed database; if so, call the corresponding student privacy data; if not, deny access.

[0029] Exemplary methods of obtaining student privacy data may include: students upload their own identity information, including name, contact information, student number, major, department, ID card information, family income, etc., through the school's information collection system; teachers upload students' school information, including grades, rewards and punishments, etc., through the school's information collection system. The identity information uploaded by students and the school information uploaded by teachers are collectively referred to as student privacy information, and the system converts student privacy information into student privacy data.

[0030] By using artificial intelligence to classify data, we can avoid the leakage of student privacy data that may be caused by manual classification of data. In addition, since artificial intelligence is used to classify data, the classification accuracy is higher than that of manual classification, and the classification efficiency is faster;

[0031] At the same time, by verifying the user information of the user who is granted access rights and monitoring the user's search frequency and the relevance of the search content, when an abnormal situation is found, the user's access rights are promptly reduced or disabled to avoid the leakage of student privacy data, thereby achieving the purpose of protecting student privacy data.

[0032] Furthermore, the privacy level classification specifically includes the following steps:

[0033] Establish the corresponding relationship between keywords and privacy levels;

[0034] Obtain student privacy data, extract keywords in the student privacy data through a CRNN-based text extraction model, and divide the privacy level of the student privacy data according to the privacy level of the keywords.

[0035] For example, the privacy level is divided into the first level privacy level, the second level privacy level and the third level privacy level; the keywords include name, student number, contact information, major, department, grades, rewards and punishments, ID card information, family income, etc. For example, three databases are configured, corresponding to the first level privacy database, the second level privacy database and the third level privacy database; name and student number correspond to the first level privacy data, contact information, major, department, grades and rewards and punishments correspond to the second level privacy data, and ID card information and family income correspond to the third level privacy data.

[0036] The CRNN-based text extraction model, as a text extraction module, is responsible for extracting keywords of student privacy data. When classifying keywords, the keywords can be classified according to preset rules. In this embodiment, name and student ID number correspond to first-level privacy data, contact information, major, department, grades and rewards and punishments correspond to second-level privacy data, and ID card information and family income correspond to third-level privacy data. Therefore, when the CRNN-based text extraction model extracts keywords, the classification module classifies the keywords into the database of the corresponding level based on the preset rules.

[0037] Further, access call verification:

[0038] Before calling the corresponding student privacy data, determine whether the user information of the user matches the student privacy data. If so, directly call the student privacy data; if not, perform corresponding desensitization processing on the called student privacy data.

[0039] Specifically, the premise for calling the corresponding student privacy data is that the user's access rights are not lower than the access level of the database being accessed. Under this premise, before calling the corresponding student privacy data, it is necessary to match the user with the student privacy data to be called. The user's user information includes the user's access level, and the access level can be determined based on the user's identity.

[0040] Exemplarily, the specific matching judgment is as follows:

[0041] Determine whether there is an attribution relationship between the user information of the user and the student privacy data. If so, the user information of the user matches the student privacy data; if not, the user information of the user does not match the student privacy data.

[0042] Take a teacher from the School of Chemistry calling student privacy data as an example. When he calls the student privacy data of "Zhang San", a match is made between the department of "Zhang San" and the department where the teacher is located, namely the School of Chemistry. If "Zhang San" also belongs to the School of Chemistry, it is judged that the teacher and "Zhang San" are a match, and therefore the corresponding student privacy data is called normally.

[0043] If "Zhang San" does not belong to the School of Chemistry, it is determined that the teacher and "Zhang San" do not match, but because the teacher still has the authority to call the student's private data, the student's private data of "Zhang San" is desensitized when calling. For example, the desensitization process can be: partially hiding "Zhang San's" contact information, for example, hiding the middle 4 to 6 digits of "Zhang San's" mobile phone number, and hiding his specific rewards and punishments, and only showing part of the grades. The specific desensitization process can be set according to actual needs.

[0044] Furthermore, access call verification also includes:

[0045] A threshold for the frequency of calling student privacy data is preset, the frequency of the user calling student privacy data within time t is monitored, and it is determined whether the frequency of the user calling student privacy data is greater than or equal to the threshold for the frequency of calling student privacy data. If so, an alarm signal is issued and the user's access rights are reduced; if not, the monitoring operation is maintained.

[0046] Specifically, the frequency with which a user calls up student privacy data reflects the user's calling habits to a certain extent. If the calling frequency is too high, it is possible that the user has been stolen by criminals and is preparing to export student privacy data in large quantities. Therefore, a preliminary judgment can be made on the user's situation based on the calling frequency.

[0047] Exemplarily, the preset threshold for the frequency of calling student privacy data is 10 times / minute, that is, different student privacy data is called 10 times in one minute. If the teacher calls student privacy data at a frequency of 20 times / minute, since the frequency exceeds the preset threshold, an alarm signal is issued and the teacher's access rights are reduced. Specifically, reducing the teacher's access rights can be: reducing the teacher's access rights from being able to obtain secondary privacy data to only being able to obtain primary privacy data.

[0048] In one of the embodiments of the present disclosure, users with different access rights can preset different student privacy data call frequency thresholds.

[0049] Furthermore, when it is determined that the frequency of a user calling student privacy data is greater than or equal to a student privacy data calling frequency threshold, the correlation between the student privacy data called by the user within time t is monitored, and the correlation is clustered according to at least one common indicator of the student privacy data to obtain a clustering result; a quantity threshold of the clustering results is preset to determine whether the quantity of the clustering results is greater than the quantity threshold; if so, the user's access permission is disabled; if not, the monitoring operation is maintained.

[0050] Specifically, if the user's access rights are disabled when it is determined that the frequency of student privacy data calls exceeds a threshold, misjudgment is likely to occur. Therefore, in the above case, when it is determined that the frequency of student privacy data calls is too high, an alarm is issued and the user's access rights are reduced; on the basis of the above, the correlation between all student privacy data called by the user within the time period t is further monitored, and the correlation is clustered according to at least one common indicator of the student privacy data to obtain a clustering result; a threshold value for the number of clustering results is preset to determine whether the number of clustering results is greater than the threshold value. If so, the user's access rights are disabled; if not, the monitoring operation is maintained.

[0051] By determining the degree of correlation, it is determined that the correlation between the student privacy data of different students called by the user within time period t is very low. Combined with the premise that warnings have been issued and access rights have been reduced, it is found that the correlation between the student privacy data of different students called by the user is very low. It is confirmed that there is malicious calling of student privacy data, and therefore the user's access rights are prohibited.

[0052] Exemplarily, the preset threshold value of the number of clustering results is 3, that is, when the number of clustering results obtained by clustering analysis of at least one common indicator of student privacy data exceeds 3, it is confirmed that there is a malicious call to student privacy data, and therefore the user's access rights are prohibited.

[0053] Specifically, the common indicators take the college as an example, and the user takes the teacher of the School of Chemistry as an example. When the frequency of the Chemistry teacher calling the student privacy data exceeds 20 times / minute, taking the calling frequency of 90 times in three minutes as an example, the correlation between the 90 student privacy data called by the Chemistry teacher within a 3-minute period is monitored, and the clustering result of the 90 student privacy data obtained through cluster analysis is 4. For example, among the 90 student privacy data, 40 people are from the School of Chemistry, 20 people are from the School of Literature, 20 people are from the School of Physics and Electronic Engineering, and 10 people are from the School of Politics and Law. It is confirmed that the Chemistry teacher has maliciously called the student privacy data, and therefore the user's access rights are prohibited.

[0054] The present invention discloses an artificial intelligence-based student privacy data protection method, which uses artificial intelligence to classify data to avoid leakage of student privacy data that may be caused by manual classification of data. In addition, since artificial intelligence is used to classify data, the classification accuracy is higher than manual classification, and the classification efficiency is faster.

[0055] At the same time, by verifying the user information of the user who is granted access rights and monitoring the user's search frequency and the relevance of the search content, when an abnormal situation is found, the user's access rights are promptly reduced or disabled to avoid the leakage of student privacy data, thereby achieving the purpose of protecting student privacy data.

[0056] The present disclosure also provides a student privacy data protection system, which is used to execute the above-mentioned artificial intelligence-based student privacy data protection method.

[0057] The present disclosure also provides a non-temporary computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned artificial intelligence-based student privacy data protection method is implemented.

[0058] In the description of the present disclosure, it is necessary to understand that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present disclosure and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present disclosure.

[0059] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present disclosure.

Claims

1. A student privacy data protection method based on artificial intelligence, characterized in that: The following steps are involved: Database configuration: configuring a number of databases, wherein the databases are used to store student privacy data of different privacy levels, and each of the databases is respectively configured with a corresponding access level; Privacy level classification: obtaining student privacy data and classifying the student privacy data into privacy levels; Privacy data storage: store the divided student privacy data in a database corresponding to the privacy level; Access call verification: In response to the user's access request, obtain the user's user information and determine whether the user's access rights are not lower than the access level of the accessed database; if so, call the corresponding student privacy data; if not, deny access.

2. According to the method for protecting student privacy data based on artificial intelligence in claim 1, it is characterized in that: The privacy level classification specifically includes the following steps: Establish the corresponding relationship between keywords and privacy levels; The student privacy data is obtained, keywords in the student privacy data are extracted by using a text extraction model based on CRNN, and the privacy level of the student privacy data is divided according to the privacy level to which the keywords belong.

3. The method for protecting student privacy data based on artificial intelligence according to claim 1 is characterized in that: In the access call verification: Before calling the corresponding student privacy data, determine whether the user information of the user matches the student privacy data. If so, directly call the student privacy data; if not, perform corresponding desensitization processing on the called student privacy data.

4. According to the method for protecting student privacy data based on artificial intelligence in claim 3, it is characterized in that: It is determined whether there is an attribution relationship between the user information of the user and the student privacy data. If so, the user information of the user matches the student privacy data; if not, the user information of the user does not match the student privacy data.

5. The method for protecting student privacy data based on artificial intelligence according to claim 1 is characterized in that: The access call verification also includes: A student privacy data call frequency threshold is preset, the frequency of the user calling the student privacy data within time t is monitored, and it is determined whether the frequency of the user calling the student privacy data is not less than the student privacy data call frequency threshold. If so, an alarm signal is issued and the user's access rights are reduced; if not, the monitoring operation is maintained.

6. The method for protecting student privacy data based on artificial intelligence according to claim 5 is characterized in that: When it is determined that the frequency of a user calling student privacy data is not less than the student privacy data calling frequency threshold, the correlation between the student privacy data called by the user within time t is monitored, and the correlation is clustered according to at least one common indicator of the student privacy data to obtain a clustering result; a number threshold of clustering results is preset, and it is determined whether the number of clustering results is greater than the number threshold; if so, the user's access permission is disabled; if not, the monitoring operation is maintained.

7. A student privacy data protection system, characterized in that: The student privacy data protection system includes the student privacy data protection method based on artificial intelligence as described in any one of claims 1 to 6.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for protecting student privacy data based on artificial intelligence as described in any one of claims 1 to 6 is implemented.

Images