With the increase of the number of users in the current social network platform (taking WeChat as an example), personal privacy security issues are important. This paper proposes an intelligent detection method for personal privacy disclosure in social networks. Firstly, we propose and construct the eigenvalue in social platform. Secondly, by calculating the value of user account assets, we can obtain the eigenvalue to calculate the possibility of threat occurrence and the impact of threat. Thirdly, we analyse the situation that the user may leak the privacy information and make a score. Finally, SVM algorithm is used to classify the results, and some suggestions for warning and modification are put forward. Experiments show that this intelligent detection method can effectively analyse the privacy leakage of individual users.
Today’s society is developing rapidly, and with the popularity of smartphones, the amount of private information they generate is increasing. With the occurrence of “PRISM,” Facebook user personal information leaked, and other incidents, the issue of private security has begun to attract people’s attention.
In recent years, WeChat is one of the most popular apps in China, and as of June 2019, the number of monthly active accounts on WeChat reached 1.13 billion. The huge number of users will contain a large amount of user privacy information. However, most users do not have relevant expertise or neglect information management. Therefore, when using WeChat, they do not pay attention to the protection of private information. After investigation and analysis, disclosure of account passwords, the addition of friend settings, location information, etc. during the chat process will pose a threat to user data if they are leaked and may cause economic loss or even personal harm. Because of the above problems, there have been related studies. Reference [
The innovations proposed in this article are as follows: (1) Through the investigation and analysis of WeChat, eight characteristic items in social networks are proposed and constructed, which are account passwords during chats, WeChat wallet consumption records (not friends), and WeChat wallet transfer records (friends), Moments settings of strangers, settings for nearby people, settings for adding friends, Moments settings of friends, and information acquisition of mini-program. The intelligent detection system uses these filtered feature items to calculate the risk value more efficiently and accurately. (2) Use the operations of asset identification, threat identification, and vulnerability analysis to calculate the comprehensive threat value. (3) An intelligent detection method based on SVM (Support Vector Machine) is proposed to divide the data more accurately. (4) After investigation, most of the detection software with similar functions today is oriented to enterprises, and this system is a rare intelligent detection system for individual users on the market.
The intelligent detection method of personal privacy leakage for social networks proposed in this article is always for users. There is a risk of personal privacy leakage. By obtaining user WeChat settings, asset identification, threat identification, and vulnerability analysis are performed, and the matrix is compared to obtain security. For event risk value, calculate information leakage risk coefficient according to weight. Reference [
The risks explored in this article refer to the risks of information security breaches, human or natural threats, and the use of vulnerabilities in information systems and their management systems to cause security incidents and their impact on organizations. In the current environment of high information transparency, private information cannot be in a state of zero risks [
Assets refer to any information or resources that are valuable to the unit. The value of assets does not refer to the economic value of the information system but is closely related to the business work of the organization. Asset value is the importance and sensitivity of assets and the main content of asset identification.
Asset identification includes two steps: “asset classification” and “asset assignment.” This article explores the classification of application software. Based on asset classification, further semiqualitative and semiquantitative analysis of assets is performed; that is, asset valuation is performed, to have a scientific and rational understanding of asset value. Assets are broken down into three security attribute assignments: “confidentiality assignment,” “integrity assignment,” and “availability assignment.”
It is the feature that prevents the information from being leaked to unauthorized individuals, entities, processes, or makes it useless.
It protects the accuracy and completeness of information and processing methods.
It is a feature that can be accessed and used by authorized entities once they are needed [
Threat: Potential cause of an accident that may cause damage to assets or units. Threat identification: Referring to the process of analysing the potential cause of an accident. Threat identification is divided into “threat classification” and “threat assignment” [
Vulnerability: Weakness in assets or assets that can be threatened. Compared with threats, threats are the external cause of risk, and vulnerability is the internal cause of risk. The two together form a risk. Vulnerability identification: Referring to the process of analysing and measuring the weak links of assets that may be threatened to use [
SVM refers to support vector machine, which is a common method of discrimination. In the field of machine learning, it is a supervised learning model, which is usually used for pattern recognition, classification, and regression analysis.
The main idea of SVM can be summarized as two points: It analyses linearly separable cases. For linearly inseparable cases, by using a nonlinear mapping algorithm, a linearly inseparable sample from a low-dimensional input space is transformed into a high-dimensional feature space to make it linearly separable. It is possible to perform a linear analysis of the nonlinear features of the sample using a linear algorithm in the feature space. It constructs the optimal hyperplane in the feature space based on the structural risk minimization theory, so that the learner is globally optimized, and the expectations in the entire sample space meet a certain upper bound with a certain probability [
This intelligent detection model is divided into a data source layer, an analysis layer, and a calculation layer, as shown in Figure
Schematic diagram of detection system structure.
Then the sum of the weight values of each risk value is used to obtain the risk value, and the risk value is brought into the corresponding SVM classifier to obtain the final result.
Based on the investigation and analysis of WeChat, we selected the following conditions as the eigenvalues. The intelligent detection system uses these filtered feature items to calculate the risk value more efficiently and accurately:
The account and password are directly mentioned during the chat. If the chat history is stolen, the account and password information is leaked, and the entire account will be lost, with more illegal acts.
They require money to communicate with each other without knowing too much about the identity of the other party, have lack of security protection, and may cause economic losses.
The transfer security between friends is higher than the transfer between non-friends, but if the identity of the friend is impersonated, the identity of the transfer counterparty is unknown, so even the transfers between friends will be at risk.
The setting of a circle of strangers is divided into invisible to strangers, ten photos visible to strangers, and unlimited. If the attacker continuously obtains the user circle information for a long time, the stranger can see that the ten photos are not much different from unlimited, which will cause a large amount of information leakage for the user.
If the nearby people are not closed, the real-time location of the user will be exposed and used by criminals.
The related settings include whether you need to verify when adding as a friend. The way to search for users is divided into WeChat, mobile phone number, and QQ number, in addition to business card. Too many permissions in this regard will increase the possibility of being disturbed by strangers. Location of Moments: The attacker can further commit a crime based on the obtained positioning information, causing the user’s personal safety to be threatened Mini-Program Information Acquisition: Mini-programs usually obtain user information. If the mini-programs are used by criminals, arbitrating user information will lead to user information leakage.
Assets have security attributes such as confidentiality, integrity, and availability, which reflect the characteristics of the asset in different aspects. By quantifying the three security attributes, one can calculate a value that reflects the asset [
Among them,
It can be seen from Table
Quantification of asset identification.
Characteristic values | Security attributes | |||
---|---|---|---|---|
Conf | Int | Avail | Asset value | |
f1: Account password revealed during chat | 5 | 5 | 5 | 5 |
f2: WeChat wallet consumption records (non-friends) | 3 | 3 | 1 | 4 |
f3: WeChat wallet transfer records (friends) | 3 | 1 | 1 | 3 |
f4: Moments permissions settings for strangers | 3 | 3 | 4 | 4 |
f5: Setting nearby people | 3 | 4 | 4 | 4 |
f6: Adding friends | 3 | 4 | 3 | 4 |
f7: Moments location targeting | 1 | 1 | 3 | 3 |
f8: Using mini-program | 3 | 4 | 3 | 4 |
According to the frequency of threats, the possibility of threats is defined and divided into 5 levels. The higher the level, the higher the probability of threats.
It can be seen from Table
Frequency of threats.
Level | |||||
---|---|---|---|---|---|
Characteristic values | 5 | 4 | 3 | 2 | 1 |
f1: Account password revealed during chat | 10 times or above | 7∼9 | 5∼6 | 3∼4 | 0∼2 |
f2: WeChat wallet consumption records (non-friends) | 10 times or above | 5∼6 | 3∼4 | 0∼2 | |
f3: WeChat wallet transfer records (friends) | 21 times or above | 16∼20 | 11∼15 | 6∼10 | 0∼5 |
f4: Moments permissions settings for strangers | Allow strangers to see ten Moments | Not allowing strangers to see ten Moments | |||
f5: Setting nearby people | Open | Close | |||
f6: Adding friends | No verification required; can be searched through WeChat, QQ number, mobile phone number; can be added through group chat, QR code, business card | No verification required; can be searched through WeChat, QQ number, mobile phone number; not added through group chat, QR code, business card | Requires verification; can be searched by WeChat, QQ number, mobile phone number; can be added through group chat, QR code, business card | Requires verification; can be searched by WeChat, QQ number, mobile phone number; not added by group chat, QR code, business card | Requires verification; cannot be searched by WeChat, QQ, or mobile phone number; can be added through group chat, QR code, business card |
f7: Moments location targeting | 10 times or above | 7∼9 | 5∼6 | 3∼4 | 0∼2 |
f8: Using mini-program | 10 and above | 7∼9 | 5∼6 | 3∼4 | 0∼2 |
This paper uses the Common Weak Evaluation System (CVSS). The CVSS evaluation system consists of three measurement groups: the basic measurement group, the time measurement group, and the environment measurement group [
Basic metric = round_to_1_decimal (10 ∗ access vector ∗ access complexity ∗ authentication ∗ ((confidentiality impact ∗ confidentiality impact weight value) + (consistent impact ∗ consistency impact weight value) + (availability impact ∗ availability impact weight value)))
The values in Table
Calculation table of basic metrics.
Characteristic values | Related parameters | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
a1 | a2 | a3 | a4 | a5 | a6 | a7 | a8 | a9 | a10 | |
f1 | 0.7 | 0.8 | 1.0 | 1.0 | 0.5 | 1.0 | 0.25 | 1.0 | 0.25 | 5.6 |
f2 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 0.7 | 0.333 | 3.92 |
f3 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 0.7 | 0.333 | 3.92 |
f4 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 1.0 | 0.333 | 2.3 |
f5 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 1.0 | 0.333 | 2.3 |
f6 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 1.0 | 0.333 | 2.3 |
f7 | 0.7 | 0.8 | 1.0 | 0.7 | 0.25 | 0.7 | 0.25 | 0.7 | 0.5 | 3.92 |
f8 | 0.7 | 0.8 | 1.0 | 0.7 | 0.333 | 0.7 | 0.333 | 1.0 | 0.333 | 2.3 |
a1: access vector, a2: access complexity, a3: authentication, a4: confidentiality impact, a5: confidentiality impact weight value, a6: consistency impact, a7: consistency influence weight value, a8: usability impact, a9: usability impact weight value, and a10: basic measure.
Basic metric value assignment table.
Parameters | Values corresponding to different degrees | |||
---|---|---|---|---|
Access vector | Local: 0.7 | Remote: 1.0 | ||
Access complexity | Height: 0.8 | Low: 1.0 | ||
Authentication | Requires: 0.6 | Not required: 1.0 | ||
Confidentiality impact | None: 0 | Section: 0.7 | All: 1.0 | |
Confidentiality impact weight value | Normal: 0.333 | Confidentiality: 0.5 | Consistency: 0.25 | Availability: 0.25 |
Consistency impact | None: 0 | Section: 0.7 | All: 1.0 | |
Consistency influence weight value | Normal: 0.333 | Confidentiality: 0.25 | Consistency: 0.5 | Availability: 0.25 |
Usability impact | None: 0 | Section: 0.7 | All: 1.0 | |
Usability impact weight value | Normal: 0.333 | Confidentiality: 0.25 | Consistency: 0.25 | Availability: 0.5 |
The values in Table
Time measurement value calculation table.
Characteristic values | Related parameters | |||
---|---|---|---|---|
Exploitability | Repairable level | Confidentiality of the report | Time metric | |
f1: Account password revealed during chat | 1.00 | 0.87 | 0.90 | 4.4 |
f2: WeChat wallet consumption records (non-friends) | 0.85 | 1.00 | 0.90 | 3.0 |
f3: WeChat wallet transfer records (friends) | 0.85 | 1.00 | 0.90 | 3.0 |
f4: Moments permissions settings for strangers | 0.95 | 0.87 | 0.90 | 1.7 |
f5: Setting nearby people | 0.95 | 0.87 | 0.90 | 1.7 |
f6: Adding friends | 0.95 | 0.87 | 0.90 | 1.7 |
f7: Moments location targeting | 0.90 | 0.90 | 0.90 | 2.9 |
f8: Using mini-program | 0.95 | 0.87 | 0.90 | 1.7 |
Time metric value assignment table.
Parameters | Values corresponding to different degrees | |||
---|---|---|---|---|
Exploitability | Unconfirmed: 0.85 | Proved by theory: 0.90 | Practical: 0.95 | High: 1.00 |
Repairable level | Unconfirmed: 0.87 | Proved by theory: 0.90 | Practical: 0.95 | High: 1.00 |
Confidentiality of the report | Unconfirmed: 0.90 | Unverified: 0.95 | Confirmed: 1.00 |
Environmental metric value = round_to_1_decimal ((time metric score + ((10-time metric score) ∗ incidental loss impact)) ∗ target distribution).
The values in Table
Calculation table of environmental measures.
Characteristic values | Related parameters | ||
---|---|---|---|
Collateral loss effects | Target distribution | Environmental metric value | |
f1: Account password revealed during chat | 0.5 | 1.00 | 7.2 |
f2: WeChat wallet consumption records (non-friends) | 0.3 | 0.25 | 1.3 |
f3: WeChat wallet transfer records (friends) | 0.3 | 0.25 | 1.3 |
f4: Moments permissions settings for strangers | 0.5 | 0.75 | 4.4 |
f5: Setting nearby people | 0.5 | 0.75 | 4.4 |
f6: Adding friends | 0.5 | 0.25 | 1.5 |
f7: Moments location targeting | 0.1 | 0.25 | 0.9 |
f8: Using mini-program | 0.5 | 0.75 | 4.4 |
Note: round_to_1_decimal refers to rounding to one decimal place.
Assignment table of environmental metrics.
Values corresponding to different degrees | ||||
---|---|---|---|---|
Parameters | No | Low | Middle | High |
Collateral loss effects | 0 | 0.1 | 0.3 | 0.5 |
Target distribution | 0 | 0.25 | 0.75 | 1.0 |
Vulnerability value calculation formula [
Vulnerability value calculation table.
Characteristic values | Vulnerability value |
---|---|
f1: Account password revealed during chat | 4 |
f2: WeChat wallet consumption records (non-friends) | 1 |
f3: WeChat wallet transfer records (friends) | 1 |
f4: Moments permissions settings for strangers | 3 |
f5: Setting nearby people | 3 |
f6: Adding friends | 1 |
f7: Moments location targeting | 1 |
f8: Using mini-program | 3 |
After completing asset identification, threat identification, and vulnerability identification, an appropriate model can be used to calculate the risk value of a security event caused by the vulnerability using threats. This article adopts the risk calculation model in Chinese National Standard GB/
The formula is expressed as risk value =
In the specific calculation of risk, there are three key calculation links.
According to the frequency and vulnerability of threats, calculate the probability that a threat will cause a security event using vulnerability, that is, the probability of a security event =
This system uses a two-dimensional matrix algorithm to calculate the probability of a security event, as shown in Table
Two-dimensional matrix of security event probability calculations.
Severity of vulnerability | Frequency of threats | ||||
---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | |
1 | 2 | 4 | 7 | 9 | 12 |
2 | 4 | 6 | 9 | 13 | 16 |
3 | 6 | 9 | 13 | 17 | 21 |
4 | 8 | 11 | 14 | 21 | 23 |
5 | 9 | 13 | 18 | 23 | 25 |
According to the value of the asset and the severity of the vulnerability, calculate the loss caused by the security event once it occurs, that is, the loss caused by the security event =
This system uses a two-dimensional matrix method to calculate the loss of security events, as shown in Table
Two-dimensional matrix table of security event loss calculation.
Severity of vulnerability | Asset value | ||||
---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | |
1 | 2 | 3 | 6 | 9 | 11 |
2 | 3 | 6 | 9 | 12 | 16 |
3 | 5 | 8 | 12 | 16 | 20 |
4 | 7 | 10 | 13 | 19 | 22 |
5 | 9 | 13 | 18 | 23 | 25 |
According to the calculated probability of the security event and the loss caused by the security event, calculate the risk value, that is, risk value =
The system uses the two-dimensional code matrix method to calculate the risk value of security events, as shown in Table
Two-dimensional matrix table for calculating the risk value of security events.
Loss caused by the security event | Probability of the security event | ||||
---|---|---|---|---|---|
1∼5 | 6∼10 | 11∼15 | 16∼20 | 21∼25 | |
1∼5 | 3 | 6 | 9 | 14 | 13 |
6∼10 | 6 | 11 | 17 | 21 | 21 |
11∼15 | 11 | 18 | 22 | 30 | 30 |
16∼20 | 15 | 21 | 31 | 40 | 55 |
21∼25 | 22 | 35 | 55 | 85 | 100 |
The risk value of each data security event is obtained from Table
Comprehensive threat calculation table.
Characteristic values | Related parameters | |||
---|---|---|---|---|
t | Weights | |||
f1: Account password revealed during chat | 4 | 5 | 9 | 0.225 |
f2: WeChat wallet consumption records (non-friends) | 3 | 3 | 6 | 0.15 |
f3: WeChat wallet transfer records (friends) | 2 | 2 | 4 | 0.1 |
f4: Moments permissions settings for strangers | 2 | 2 | 4 | 0.1 |
f5: Setting nearby people | 2 | 3 | 5 | 0.125 |
f6: Adding friends | 2 | 2 | 4 | 0.1 |
f7: Moments location targeting | 2 | 1 | 3 | 0.075 |
f8: Using mini-program | 2 | 3 | 5 | 0.125 |
All comprehensive threat value calculation formulas [
The calculation formula for the comprehensive calculation value of a single threat to an information asset:
Among them,
The intelligent detection system uses the SVM algorithm to divide the comprehensive threat value (as shown in Figure
SVM algorithm application diagram.
The specific process is as follows.
Based on the comprehensive threats mentioned above, it is worth calculating the risk value. The obtained risk values are divided into two categories. The scores of 1 to 40 are low-risk areas, and the scores of 40 to 100 are high-risk areas. Among them, 1 to 20 in the low-risk areas are defined as safe, 21 to 40 are defined as basic safety, 41 to 59 in the high-risk areas are defined as higher risks, and 60 to 100 parts are expressed as high risks.
The feature quantities of two types of risk values defined as safety and basic safety are recorded in the initial feature vector set 1. The feature quantities defined as two types of risk values of higher-risk and high-risk are recorded in the initial feature vector set 2.
Normalize the feature data items to remove the extreme data. Convert the processed two types of data formats into an input format acceptable to the SVM classifier (class vector Y, feature vector Xi)
The corresponding classifier 1 is trained using data defined as safe and basic safety as training samples, and the corresponding classifier 2 is trained using data defined as safe and basic safety as training samples.
Set the SVM parameters and use the K-fold cross-validation algorithm to find the optimal parameters. Perform asset identification, vulnerability analysis, and threat identification from the characteristic values read by the user. After risk calculation, determine the low-risk area or high-risk area based on the score and enter the corresponding risk area as a test sample. The SVM classification model performs classification judgment. Substitute the results obtained by the SVM into the Naive Bayes formula to obtain the security risk probability, and send feedback of the final results to the user.
The SVM calculation process is shown in Figure
SVM calculation process.
<label> <index (1)>: <value1> <index (2)>: <value2> ...
For example: 0 1 : 1 2 : 1 3 : 1 4 : 1 5 : 2 6 : 2 7 : 2 8 : 2.
Among them:
<label> is the category identifier of the training data set, set to 0 and 1,0 for security, and 1 for basic security.
<index> refers to 8 feature quantities of the input algorithm, which are integers.
<value> is the value of the feature code for each item and is an integer.
SVM_train implements training on training samples to obtain SVM models.
SVM classification is a prediction of the classification result of the data set according to the trained model.
Use SVM_train to train the input training data set to obtain the SVM model file. The SVM algorithm maps each input training sample, that is, an n-dimensional vector into a high-dimensional space, forming multiple scattered points, and passing the aggregation of points. The region simulates the classification hyperplane and continuously uses the newly input training sample data to make corrections, generates template files, and records the classification features.
In this paper, the K-fold cross-validation method is used to obtain the optimal parameters by verifying the accuracy of the results. The main purpose of the verification algorithm is to divide the data set A into a training set B and a test set C. When the sample size is small, the data set A can be randomly divided into
SVM classifier 1: Step 1: Normalize the feature data Step 2: Convert the processed feature data into an input format acceptable by the classifier (feature vector Step 3: Set the SVM type to 0-SVM and the kernel function type to radial basis function (RBF) Step 4: Set the penalty factor C and kernel function parameter G Step 5: Set the K value of the K-fold cross-validation algorithm Step 6: Use the SMO algorithm to find the support vector Step 7: Build a hyperplane model from training samples Step 8: Enter the test samples for classification, and get the classification result y Step 9: Calculate the Step 10: Calculate Step 11: Calculate Step 12: Substitute the formula Step 13: return SVM classifier 2: SVM classifier 2 process is the same as SVM classifier 1.
This test system is designed to run on the Android platform. During the test phase, Android Studio is used to simulate the Android platform for various tests.
Due to the inconvenience of directly obtaining the personal privacy data of the user’s WeChat, a questionnaire was used at this stage to collect the WeChat usage of 149 users as a training sample for the SVM classifier. The specific content of the questionnaire is shown as Appendix in Supplementary Materials (available
User test assignment table is shown in Table
User test assignment table.
Characteristic values | User test related assignments |
---|---|
f1: Account password revealed during chat (6 times) | 3 |
f2: WeChat wallet consumption records (non-friends) (3 times) | 1 |
f3: WeChat wallet transfer records (friends) (10 times) | 2 |
f4: Moments permissions settings for strangers (close) | 1 |
f5: Setting nearby people (close) | 1 |
f6: Requires verification; cannot be searched by WeChat, QQ, or mobile phone number; can be added through group chat, QR code, business card | 1 |
f7: Moments location targeting (8 times) | 4 |
f8: Using mini-program (2) | 2 |
After removing the extreme data from the remaining samples, the above steps are processed and sent to the corresponding SVM classifier. The training samples are used to build a hyperplane model. When the system intelligently detects the risk leakage probability, it will automatically obtain the feature quantity, calculate the comprehensive threat value after calculation, and send it to the corresponding SVM classifier to obtain the final security risk probability.
According to this method, we processed the results of 149 user questionnaires and calculated the number of scores for each segment. The results are shown in the following Table
Number distribution of each risk area.
Level | Number |
---|---|
Safety | 60 |
Basic safety | 29 |
High-risk | 26 |
Higher-risk | 34 |
From Table
Obtain user WeChat related information through a questionnaire. As a sample, test the personal privacy leak detection value of a user’s social network, and give a warning or suggestion to get the percentage of people at each risk level, and then get the current data of whether people know and implement the degree of privacy protection in place, which aspects are of importance to people, and which aspects are ignored by people, and provide directions for the promotion of privacy protection awareness in the future. The findings are shown in Figure
Statistics of the percentage of occurrences of each feature.
At the same time, we counted the number of occurrences of high threats for each feature item (that is, the number of times assigned 4 or 5).
In Figure
Statistics of occurrence times of high threats for each feature.
The system proposed in this article is based on reading multiple characteristic values of personal WeChat and establishing a model based on three aspects of asset identification, threat identification, and vulnerability analysis. According to the risk calculation models and methods in the national standards of information security risk assessment standards, the dimension matrix table calculates the possibility of security events, the loss of security events, and the risk value of security events, determines the risk level according to the magnitude of the risk, evaluates the personal privacy leakage of the user’s online social software, gives a score, and informs the user about source of risk.
This article only mentions the scoring function in the system and the function of displaying the risk source of personal privacy leakage. In the future, more functions will be added to improve the entire system, which will also make the judgment more accurate and create a more accurate situation for the individual users, creating safe environment to use social networks.
Due to the inconvenience of directly obtaining the personal privacy data of the user’s WeChat, a questionnaire was used at this stage to collect the WeChat usage of 149 users as a training sample for the SVM classifier. The specific content of the questionnaire is given in Supplementary Materials.
The authors declare that they have no conflicts of interest.
This work was partially supported by humanities and social sciences research project of the Ministry of Education (No. 20YJAZH046) and Higher Education Research Projects (No. 2020GJZD02), “Practical training plan” (Entrepreneurship) for cross training of high-level talents.
The detailed information of the questionnaire is given in Appendix I. According to the number, it is divided into five levels: 5 is the highest, it is gradually decreased in order, and 1 is the lowest.