Data monitoring and early warning platform under mobile office environment

Abstract

The application provides a data monitoring and early warning platform in a mobile office environment, relates to the technical field of data monitoring, and comprises: a data monitoring module for monitoring and acquiring a plurality of office data of current operation of a user, and acquiring a plurality of data characteristic information and a plurality of operation types; a sensitivity analysis module for performing sensitivity analysis according to the plurality of data characteristic information; a volatility analysis module for monitoring a plurality of network parameters and performing volatility analysis; and a risk early warning module for performing data risk analysis calculation according to data sensitivity parameters, a plurality of network volatility information and a plurality of operation types, and performing early warning according to data risk parameters. The application can solve the technical problem of unstable data transmission and risk caused by great network condition fluctuation affecting data security in the prior art, and improve the safety and stability of data transmission in a mobile office scene by dynamically monitoring network volatility and performing multidimensional comprehensive evaluation.

Description

Technical Field

[0001] This application relates to the field of data monitoring technology, and in particular to a data monitoring and early warning platform for mobile office environments. Background Technology

[0002] With the rapid development of information technology, mobile office has become an important part of modern office work. In a mobile office environment, users typically transmit data and collaborate remotely via public or mobile networks. However, the instability of the network environment, the security of data transmission, and the real-time requirements of operations pose serious challenges to data management and information security. Uploading and downloading sensitive data (such as financial statements, contracts, and work reports) is a core operation in mobile office work. Data transmission over public networks is highly vulnerable to cybersecurity threats such as man-in-the-middle attacks and malicious traffic hijacking. Mobile office work often relies on public Wi-Fi and mobile 4G / 5G networks, which are subject to performance fluctuations (such as bandwidth degradation and increased latency) and security risks (such as data theft). Existing data monitoring methods lack real-time monitoring of network parameters such as upload bandwidth, download bandwidth, and latency, especially in public network environments with significant network fluctuations, failing to quantify the impact of volatility on data transmission security.

[0003] In summary, existing technologies suffer from technical problems such as data instability and risks in data transmission due to network fluctuations in public or wireless network environments during mobile office scenarios, which affect data security. Summary of the Invention

[0004] The purpose of this application is to provide a data monitoring and early warning platform for mobile office environments, in order to solve the technical problems in the existing technology where data security is affected by network fluctuations in public or wireless network environments during mobile office scenarios, leading to unstable data transmission and risks.

[0005] In view of the above problems, this application provides a data monitoring and early warning platform for mobile office environments. The platform includes: a data monitoring module, used to monitor and acquire multiple office data currently being used by a user in a mobile office environment, and to acquire multiple data feature information and multiple operation types of the multiple office data, wherein the operation types include upload or download; a sensitivity analysis module, used to perform sensitivity analysis on the multiple office data based on the multiple data feature information to obtain data sensitivity parameters; a volatility analysis module, used to monitor multiple network parameters of the network environment in the mobile office environment, obtain multiple network parameter sequences, and perform volatility analysis on each sequence to obtain multiple network volatility information, wherein the multiple network parameters include upload bandwidth, download bandwidth, and latency; and a risk early warning module, used to perform data risk analysis calculations based on the data sensitivity parameters, multiple network volatility information, and multiple operation types to obtain data risk parameters and issue early warnings.

[0006] The technical solution provided in this application has at least the following technical effects or advantages:

[0007] By performing sensitivity analysis on multiple characteristics of office data, real-time monitoring of network parameters such as upload bandwidth, download bandwidth, and latency, and analyzing their volatility, the stability and quality of the network environment are assessed. Combining data sensitivity parameters, network volatility information, and operation type, the risks of data operations are comprehensively evaluated, and early warnings are issued accordingly to reduce the risk of data leakage or damage, thereby improving the security and stability of data transmission in mobile office scenarios.

[0008] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0010] Figure 1This is a schematic diagram of the data monitoring and early warning platform in the mobile office environment of this application;

[0011] Figure 2 This is a flowchart illustrating the risk warning process in the data monitoring and early warning platform under the mobile office environment of this application.

[0012] Figure labeling: Data monitoring module 11, sensitivity analysis module 12, volatility analysis module 13, risk warning module 14. Detailed Implementation

[0013] This application provides a data monitoring and early warning platform for mobile office environments, addressing the technical problem in existing technologies where network fluctuations in public or wireless network environments during mobile office scenarios often affect data security, leading to unstable and risky data transmission. By performing sensitivity analysis on multiple characteristic information of office data, real-time monitoring of network parameters such as upload bandwidth, download bandwidth, and latency, and analyzing their fluctuations, the stability and quality of the network environment are assessed. Combining data sensitivity parameters, network fluctuation information, and operation type, the risks of data operations are comprehensively evaluated, and early warnings are issued accordingly, reducing the risk of data leakage or damage and improving the security and stability of data transmission in mobile office scenarios.

[0014] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0015] For examples, please refer to the appendix. Figure 1 This application provides a data monitoring and early warning platform for mobile office environments, wherein the data monitoring and early warning platform for mobile office environments includes:

[0016] The data monitoring module 11 is used to monitor and acquire multiple office data currently operated by the user in a mobile office environment, and to acquire multiple data feature information and multiple operation types of the multiple office data, wherein the operation types include uploading or downloading.

[0017] Specifically, a mobile work environment refers to scenarios where users complete work tasks in non-traditional workplaces (such as at home, in a coffee shop, or while traveling) using mobile devices (such as laptops, tablets, and mobile phones) and various network methods (such as Wi-Fi, mobile data, and VPN). For example, employees might access the company's internal system via remote desktop or upload contract documents using public Wi-Fi in a hotel. Data monitoring tools (such as Windows API, cloud storage SDKs, and file system monitoring tools) can be used to track all user actions in the mobile work environment in real time, such as editing documents, sending emails, and uploading or downloading files. For instance, traffic analysis tools can be used to capture the data streams of uploads and downloads.

[0018] Extract the core features of each monitored data point, including data path, data object, and data target. Data path refers to the data's storage location or access path, i.e., its storage location on a local or remote system; data object is the customer or entity associated with the data determined through file content analysis, such as determining that the file belongs to customer A; data target refers to important business indicators within the data (such as contract amount or project budget), such as the total amount of the associated project.

[0019] This data collection captures the specific operations a user is currently performing on various office data, including uploading and downloading. These multiple operation types refer to the user's behavioral categorization of data; common operation types include uploading (transferring local files to a server or cloud) and downloading (retrieving files from remote resources to local storage). By collecting user data characteristics and operation types in a mobile office environment, this data collection helps understand data flow dynamics and user behavior patterns.

[0020] Sensitivity analysis module 12 is used to perform sensitivity analysis on the multiple office data based on the multiple data feature information to obtain data sensitivity parameters.

[0021] Specifically, multiple data feature information constitutes multidimensional information describing the characteristics of office data, including data paths, data objects, and data targets. Based on this data feature information, sensitivity parameters for office data are calculated to quantify its potential risks. An office data sensitivity classifier is constructed based on historical data to classify office data into different sensitivity levels (e.g., low, medium, and high sensitivity). Multiple data feature information is input into the office data sensitivity classifier, and different weights are assigned to paths, customers, and amounts according to the importance of the features. These multiple features are then combined into a single sensitivity parameter, and the sensitivity of each data point is calculated, resulting in multiple sensitivity levels for various office data sets. The average of these multiple sensitivity levels is calculated to obtain the data sensitivity parameter. Higher total amounts, more important customers, and more confidential paths result in higher sensitivity parameters. Sensitivity analysis is a process of assessing and determining data sensitivity, involving in-depth analysis of data feature information to identify potential risks and sensitivities within the data. Through quantitative analysis of data features and calculation of sensitivity parameters, not only is accurate sensitivity assessment of office data achieved, but also the identification of data with high sensitivity is ensured, thereby better protecting this data.

[0022] The volatility analysis module 13 is used to monitor various network parameters of the network environment in the mobile office environment, obtain multiple network parameter sequences, perform volatility analysis on each sequence, and obtain multiple network volatility information. The various network parameters include upload bandwidth, download bandwidth, and latency.

[0023] Specifically, network monitoring tools are used to monitor the network environment in a mobile office setting, collecting various network parameters at different times, including upload bandwidth, download bandwidth, and latency. This results in multiple network parameter sequences, a set of values ​​collected at multiple time points, reflecting the trend of network performance changes over time. Mobile office environments typically utilize public networks, i.e., networks accessed via public Wi-Fi or shared networks. Because these networks are not fully controlled by individuals or businesses, they are vulnerable to external threats (such as network intrusion and data theft). When someone intrudes into the network, it causes fluctuations in the upload bandwidth, download bandwidth, and latency of the public network.

[0024] Statistical analysis of network parameter sequences is used to assess network performance stability by analyzing the magnitude of changes in these sequences. Quantitative volatility analysis results are typically expressed as standard deviation or rate of change; greater volatility indicates more unstable network performance. Abnormal network volatility (such as drastic changes in upload bandwidth or significant latency fluctuations) may indicate network intrusion, data corruption, or other security threats. In public Wi-Fi environments, attackers may cause network performance anomalies, manifesting as reduced bandwidth or increased latency, through man-in-the-middle attacks or traffic hijacking.

[0025] Volatility analysis is performed on multiple network parameter sequences, specifically upload bandwidth, download bandwidth, and latency sequences. By analyzing the fluctuations in these sequences, network instability is identified, yielding upload, download, and latency volatility information as multiple network volatility metrics. This volatility analysis accurately identifies abnormal fluctuations in upload, download, and latency, aiding in the assessment of potential security threats and helping to reduce the risk of data breaches and network intrusions in public networks.

[0026] The risk warning module 14 is used to perform data risk analysis and calculation based on the data sensitivity parameters, multiple network volatility information and multiple operation types, obtain data risk parameters, and issue warnings.

[0027] Specifically, data sensitivity parameters are calculated based on data characteristics (such as path confidentiality, customer importance, and monetary sensitivity), with higher values ​​indicating more sensitive data. Multiple network volatility information refers to a comprehensive reflection of network performance instability indicators, including upload volatility, download volatility, and latency volatility.

[0028] Based on multiple operation types, namely the specific user actions on data, such as uploading (e.g., transferring files to cloud storage) and downloading (e.g., retrieving data from a server), preset latency weights are configured for latency volatility information, and upload and download weights are configured accordingly for upload and download volatility information. Using the preset latency weights, upload weights, and download weights, upload volatility information, download volatility information, and latency volatility information are weighted and calculated. Each of these is then multiplied by its corresponding weight, and the sum of all values ​​is used to obtain the network volatility information.

[0029] Multiplying network volatility information and data sensitivity parameters yields a data risk parameter. When the risk parameter exceeds a preset threshold, an alarm mechanism is triggered, issuing a warning and alerting the user to potential risks or take security measures (such as suspending operations or switching networks). Conversely, if the risk parameter is below a certain threshold, no warning is triggered. Data risk analysis calculation refers to calculating the degree of risk of data transmission in the current network environment by comprehensively considering data sensitivity parameters, network volatility information, and operation type. Assessing the transmission risk of data in a mobile office environment through data risk parameters allows for the implementation of corresponding security measures. Understanding the degree of data risk helps in developing more effective security strategies, such as encrypting high-risk data or using more secure transmission channels.

[0030] Furthermore, the data monitoring module 11 in the data monitoring and early warning platform for the mobile office environment is also used for:

[0031] In a mobile office environment, multiple office data currently being operated by the user are monitored and acquired; the data path, data object, and data target of the multiple office data are acquired respectively and combined into multiple data feature information; and multiple operation types of the current multiple office data are acquired.

[0032] Specifically, a mobile office environment refers to scenarios where users conduct work activities using portable devices (such as laptops and smartphones) in non-fixed workplaces (such as at home, in a coffee shop, or while traveling). Data monitoring tools (such as Windows API, cloud storage SDKs, and file system monitoring tools) are used to capture users' current operations in real time from their mobile devices or applications, recording the files or data being processed, resulting in multiple sets of office data, including document content, email communication records, meeting participation, and file transfer logs. The data paths (file storage locations), data objects (data ownership or business relevance, such as the relationship between data and specific customers), and data targets (content objectives or core value, usually related to the financial or functional characteristics of the business) of these multiple sets of office data are extracted. The paths, objects, and targets are then integrated into a unified feature description, forming data feature information.

[0033] This system acquires information on users' current actions regarding various office data, including uploads and downloads. Uploads refer to the process of sending local data to a server or cloud storage. For example, a user uploads an edited document to the company server. Downloads refer to the process of retrieving data from a server or cloud storage to a local device. For example, a user downloads the latest market report from the company server. By monitoring and recording user data operations in the mobile office environment in real time, key data characteristics are extracted, and operation types are identified. This provides foundational data support for subsequent data sensitivity analysis, network volatility monitoring, and data risk analysis, thereby ensuring data security and compliance.

[0034] Furthermore, the sensitivity analysis module 12 in the data monitoring and early warning platform for the mobile office environment is also used for:

[0035] Construct an office data sensitivity classifier; input the multiple data feature information into the office data sensitivity classifier to classify and obtain multiple sensitivity levels of the multiple office data; calculate the mean of the multiple sensitivity levels to obtain data sensitivity parameters.

[0036] Specifically, based on office data processing records over a period of time, data paths, data objects, and data labels are extracted to form a set of sample data feature information. This sample data feature information set undergoes preprocessing, including removing missing values, standardizing or normalizing the data, and encoding categorical variables. The preprocessed data is divided into training and testing datasets. The training dataset is used to train the classifier, while the testing dataset is used to evaluate the classifier's performance. A suitable machine learning algorithm is selected based on the characteristics of the data and the complexity of the problem, including decision trees, random forests, support vector machines (SVMs), and neural networks. The selected machine learning algorithm is trained using the training dataset. During training, the algorithm learns how to predict the sensitivity level of data based on the data feature information. Input sample features and sensitivity level labels are used, and supervised learning is used to fit the relationship between features and sensitivity levels. The classifier's performance is evaluated using the testing dataset, including accuracy, recall, and F1 score. Based on the model's performance on the testing dataset, the algorithm's parameters are adjusted to improve the classifier's performance. When the model meets the convergence requirements, training stops, and it is used as the office data sensitivity classifier.

[0037] Multiple data feature information is input into an office data sensitivity classifier to predict the sensitivity level of multiple office data being processed, resulting in multiple sensitivity levels, such as low sensitivity, medium sensitivity, and high sensitivity. For each data point, the classifier determines the sensitivity level based on the distance of the feature vector to the classification boundary. The sensitivity levels are converted into numerical values ​​(e.g., low = 1, medium = 2, high = 3), and then the average sensitivity level of all office data is calculated to quantify the overall data sensitivity risk, yielding a data sensitivity parameter. This parameter represents the sensitivity risk of the entire dataset and is used to quantify overall security requirements. By constructing an office data sensitivity classifier and calculating the data sensitivity parameter, the sensitivity of office data is quantified, potential sensitive data leakage risks are identified and addressed, and data security and compliance in mobile office environments are ensured.

[0038] Furthermore, the sensitivity analysis module 12 in the data monitoring and early warning platform for the mobile office environment is also used for:

[0039] Based on historical office data processing records, a set of sample data feature information is collected, and each sample data feature information is labeled with a sample sensitivity level to obtain a sample sensitivity level set. A random forest is used to construct an office data sensitivity classifier. The sample data feature information set and the sample sensitivity level set are divided to obtain a training set and a test set. The office data sensitivity classifier is trained under supervision using the training set and tested using the test set to determine whether it has converged. If it has converged, the training is complete and the office data sensitivity classifier is obtained; otherwise, supervised training continues until the office data sensitivity classifier converges.

[0040] Specifically, historical office data processing records include log information on user operations on office data over a past period, including records of accessing, editing, uploading, and downloading files or data. A set of sample data feature information, including data paths, data objects, and data targets, is extracted from these historical office data processing records. Based on the feature information of the office data and actual business needs, sensitivity level labeling standards are defined, such as experts or rules assigning a sensitivity level label (e.g., low sensitivity, medium sensitivity, high sensitivity) to each sample. According to the labeling standards, the sensitivity level of each sample data feature information is labeled, forming a sample sensitivity level set. This is the result set obtained after labeling all sample data feature information sets, containing the sensitivity level of each sample.

[0041] The set of sample data feature information and the set of sample sensitivity levels are combined to form supervised training data. The data is preprocessed, including data cleaning (removing invalid data) and data normalization (eliminating units of measurement between data). Numerical vectorization representations of paths, objects, and targets serve as input features, and sample sensitivity levels serve as corresponding output labels. An appropriate model (such as logistic regression, random forest, or deep learning models) is selected for training to construct an office data sensitivity classifier, predicting the sensitivity level of data based on data feature information. The appropriate model type is selected based on the amount of data and task complexity, such as logistic regression, random forest, support vector machine (SVM), or deep neural network (such as multilayer perceptron MLP).

[0042] Taking random forest as an example, by constructing multiple decision trees and combining the prediction results of each tree, higher classification accuracy and robustness are achieved. Random forest has excellent performance in handling classification problems and can handle a large number of features. The set of sample data feature information is quantified into numerical form to form the input feature vector, such as mapping paths to the 0-1 interval through keyword ratings, or assigning values ​​based on customer type. The dataset is divided into training and test sets (e.g., 70% training, 30% test), typically using stratified sampling to ensure a uniform distribution of sensitivity levels in both sets. Key parameters of the random forest are set to initialize the model, including the number of decision trees, the maximum depth of a single tree, and the weights of different categories. The training set is input into the random forest model for training, and the model learns how to predict the sensitivity level of data based on features. The model performance is evaluated using the test set by comparing predicted values ​​with true values, calculating the loss function (e.g., mean squared error), and updating the model parameters accordingly. Convergence conditions are set for the model, such as the validation set loss changing by less than 0.01 for five consecutive rounds or the training set accuracy reaching 95%. The office data sensitivity classifier is trained iteratively until a convergence condition is met, at which point training stops; otherwise, parameters are adjusted or the data processing process is optimized until the classifier reaches convergence. The trained model is then applied to automate sensitivity level classification. By building and training the office data sensitivity classifier using historical data records, it learns how to determine the sensitivity of data based on data characteristics, improving the ability to predict the sensitivity of unknown data.

[0043] Furthermore, the volatility analysis module 13 in the data monitoring and early warning platform for the mobile office environment is also used for:

[0044] The upload bandwidth, download bandwidth, and latency of the network environment in the mobile office environment are monitored at multiple timestamps to obtain upload bandwidth sequences, download bandwidth sequences, and latency sequences, which serve as multiple network parameter sequences. Volatility analysis is performed on the upload bandwidth sequences, download bandwidth sequences, and latency sequences to obtain upload volatility information, download volatility information, and latency volatility information, which serve as multiple network volatility information.

[0045] Specifically, this involves monitoring the upload bandwidth, download bandwidth, and latency of the network environment in a mobile office setting at different times. Upload bandwidth refers to the maximum rate at which data is transmitted from the local device to the network per unit time, measured in Mbps (megabits per second); download bandwidth refers to the maximum rate at which data is received from the network to the local device per unit time, also measured in Mbps; latency refers to the time required for data to travel from the source device to the destination device in the network, measured in milliseconds (ms). Multiple timestamps are recorded at different points in time. By recording network parameters at multiple timestamps, network stability and performance can be analyzed. The upload bandwidth, download bandwidth, and latency data from multiple timestamps are sorted by time to obtain upload bandwidth sequences, download bandwidth sequences, and latency sequences, which serve as multiple network parameter sequences for analyzing network performance trends and patterns.

[0046] Volatility analysis is a method for assessing network performance stability. It identifies network instability by analyzing fluctuations in network parameter sequences, typically involving calculating the range of parameter values, standard deviation, or other statistical indicators. High upload bandwidth volatility indicates frequent upload rate fluctuations, potentially affecting the stability of large file transfers. Volatility analysis is performed on each data sequence separately. Randomly selected two upload bandwidth values ​​are analyzed multiple times, and the deviation ratio between the two values ​​is calculated to obtain multiple upload bandwidth volatility data. The mean of these multiple upload bandwidth volatility data is then calculated to obtain the upload volatility information. Similarly, download bandwidth and latency volatility data are calculated for each sequence, serving as multiple network volatility data. By calculating upload, download, and latency volatility data, a comprehensive assessment of network stability is made. High volatility suggests a higher probability of network security problems.

[0047] Furthermore, the volatility analysis module 13 in the data monitoring and early warning platform for the mobile office environment is also used for:

[0048] Two upload bandwidths are randomly selected from the upload bandwidth sequence, and the deviation ratio between the two upload bandwidths is calculated as the first upload bandwidth volatility information; multiple sets of upload bandwidths are randomly selected to calculate multiple upload bandwidth volatility information, and the average value is calculated to obtain upload volatility information; download volatility information and latency volatility information are then calculated based on the download bandwidth sequence and latency sequence.

[0049] Specifically, each timestamp in the upload bandwidth sequence corresponds to an upload bandwidth value, reflecting the network's data upload capability at different times. Randomly selecting two upload bandwidth values ​​from the sequence helps avoid bias and provides a more comprehensive assessment of network volatility. The deviation ratio between the two upload bandwidth values ​​is calculated, i.e., the ratio of the deviation between the two bandwidth values ​​to their mean, yielding the first upload bandwidth volatility information. This process of randomly selecting two upload bandwidth values ​​is repeated multiple times to obtain multiple sets of upload bandwidth values. For each set of randomly selected upload bandwidth values, the deviation ratio is calculated to obtain multiple upload bandwidth volatility information. The average of these multiple upload bandwidth volatility information values ​​is calculated, and the sum of all calculated upload bandwidth volatility information values ​​is then divided by the total number of volatility information values ​​to obtain the upload volatility information, representing the overall upload bandwidth volatility.

[0050] Similarly, the above steps are performed on the download bandwidth sequence and latency sequence. By calculating the deviation ratio of multiple randomly selected bandwidth value pairs and the deviation ratio of multiple randomly selected latency value pairs, download volatility information and latency volatility information are obtained. Through random sampling and mean calculation, the overall volatility characteristics of upload, download, and latency are accurately quantified, avoiding the influence of instantaneous outliers on the results. The volatility information of upload and download bandwidth and latency is used to assess the overall stability of the network.

[0051] Further details are attached. Figure 2 As shown, the risk warning module 14 in the data monitoring and early warning platform in the mobile office environment is also used for:

[0052] Based on the multiple operation types, upload weights and download weights are configured for upload volatility information and download volatility information within the multiple network volatility information sets. The latency volatility information is configured with a preset latency weight. The upload volatility information, download volatility information, and latency volatility information are weighted and calculated using the preset latency weight, upload weight, and download weight to obtain network volatility information. The network volatility information is multiplied by the data sensitivity parameter to obtain a data risk parameter. It is determined whether the data risk parameter is greater than or equal to a data risk parameter threshold. If it is, an early warning is issued; otherwise, no early warning is issued.

[0053] Specifically, based on multiple operation types of office data—that is, the specific actions users take with data in the office environment, including uploading (such as transferring files to cloud storage) and downloading (such as retrieving files from a remote server)—latency volatility information is configured with preset latency weights. These are predefined settings used to represent the importance of latency volatility information in the overall performance evaluation. Based on these preset latency weights, the upload and download weights for upload and download volatility information are calculated. The weights, dynamically calculated based on the upload and download ratios of each operation type and the preset latency weights, are used to configure the impact of upload and download volatility information on overall network volatility. Finally, by combining upload, download, and latency volatility information, each of these values ​​is multiplied by its corresponding weight and then summed to obtain a comprehensive network volatility information value.

[0054] The data risk parameter is obtained by multiplying network volatility information by a data sensitivity parameter, comprehensively reflecting the impact of network environment and data sensitivity on operational risk. The data risk parameter threshold is the benchmark value for judging data risk; exceeding this value triggers an alert. The decision to trigger an alert is made by comparing the data risk parameter with the data risk parameter threshold. If the data risk parameter is greater than or equal to the threshold, an alert is triggered, indicating a potential security risk; if the data risk parameter is less than the threshold, the risk is considered within an acceptable range, and no alert is triggered. By adjusting the weights and thresholds, alert conditions can be flexibly set according to different business needs and network environments. Combining network volatility and data sensitivity, a more comprehensive assessment of data risk is achieved, making alerts more accurate and reducing the possibility of false positives and false negatives.

[0055] Furthermore, the risk warning module 14 in the data monitoring and early warning platform under the mobile office environment is also used for:

[0056] The preset delay weight is subtracted from 1 to obtain the weight to be configured; the proportions of upload operation type and download operation type among the multiple operation types are obtained, and multiplied by the weight to be configured to obtain the upload weight and download weight of the upload volatility information and download volatility information.

[0057] Specifically, a latency weight is predefined to represent the importance or impact of latency in the overall network performance evaluation. The preset latency weight is typically a value between 0 and 1, where 0 indicates that latency has no impact on overall performance, and 1 indicates that latency has a complete impact on overall performance. Subtracting the preset latency weight from 1 yields the configurable weight, which is used for weighting upload and download volatility information, representing the relative importance of other network parameters (such as upload and download bandwidth) in the overall performance evaluation.

[0058] This refers to the proportion of upload and download operations among various operation types for retrieving multiple office data points; that is, the ratio of upload and download operations to all operations. For example, if a user uploads 4 files and downloads 6 files within a certain time period, the upload proportion is 0.4, and the download proportion is 0.6. The number of upload and download operations performed by users over a period of time is statistically analyzed, and their respective proportions are calculated. These operation proportions will be used for weighting upload and download volatility information.

[0059] The upload weight is calculated by multiplying the percentage of upload operations by the weight to be configured. Similarly, the download weight is calculated by multiplying the percentage of download operations by the weight to be configured. For example, if the preset latency weight is 0.2, the weight to be configured is 0.8. If the upload percentage is 0.4 and the download percentage is 0.6, the upload weight is 0.32 and the download weight is 0.48. By dynamically calculating the upload and download weights, the impact of different operation types in network volatility analysis can be flexibly adjusted, making the assessment more consistent with actual network usage scenarios. By combining the operation percentage and latency weight, dynamic allocation of network volatility analysis weights is achieved, improving the model's adaptability. Different operation types (upload or download) have different sensitivities to network stability, and the weight configuration method can accurately reflect user behavior characteristics.

[0060] In summary, the data monitoring and early warning platform for mobile office environments provided in this application has the following technical effects:

[0061] By performing sensitivity analysis on multiple characteristics of office data, real-time monitoring of network parameters such as upload bandwidth, download bandwidth, and latency, and analyzing their volatility, the stability and quality of the network environment are assessed. Combining data sensitivity parameters, network volatility information, and operation type, the risks of data operations are comprehensively evaluated, and early warnings are issued accordingly to reduce the risk of data leakage or damage, thereby improving the security and stability of data transmission in mobile office scenarios.

[0062] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0063] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A data monitoring and early warning platform for mobile office environments, characterized in that: include: The data monitoring module is used to monitor and acquire multiple office data of the user's current operation in a mobile office environment, and to acquire multiple data feature information and multiple operation types of the multiple office data, wherein the operation types include uploading or downloading. A sensitivity analysis module is used to perform sensitivity analysis on the multiple office data based on the multiple data feature information to obtain data sensitivity parameters; A volatility analysis module is used to monitor various network parameters of the network environment in the mobile office environment, obtain multiple network parameter sequences, perform volatility analysis on each sequence, and obtain multiple network volatility information. The various network parameters include upload bandwidth, download bandwidth, and latency. The risk warning module is used to perform data risk analysis and calculation based on the data sensitivity parameters, multiple network volatility information and multiple operation types, obtain data risk parameters, and issue warnings. Specifically, the volatility analysis module is used for: Monitor the upload bandwidth, download bandwidth, and latency of the network environment in the mobile office environment at multiple timestamps to obtain upload bandwidth sequences, download bandwidth sequences, and latency sequences, which serve as multiple network parameter sequences. Volatility analysis is performed on the upload bandwidth sequence, download bandwidth sequence, and delay sequence to obtain upload volatility information, download volatility information, and delay volatility information, which are used as multiple network volatility information. Two upload bandwidths are randomly selected from the upload bandwidth sequence, and the deviation ratio between the two upload bandwidths is calculated as the first upload bandwidth volatility information. The deviation ratio is obtained by calculating the ratio between the deviation value between the two upload bandwidths and the mean value between the two upload bandwidths. Continue to randomly sample multiple sets of upload bandwidth, calculate multiple upload bandwidth fluctuation information, and calculate the average to obtain upload fluctuation information; Based on the download bandwidth sequence and latency sequence, download volatility information and latency volatility information are calculated. The step of performing data risk analysis and calculation based on the data sensitivity parameters, multiple network volatility information, and multiple operation types to obtain data risk parameters and issue early warnings includes: Based on the multiple operation types, the upload weight and download weight of the multiple network volatility information are configured, wherein the delay volatility information is configured with a preset delay weight; Using the preset delay weight, upload weight, and download weight, the upload volatility information, download volatility information, and delay volatility information are weighted and calculated to obtain a comprehensive network volatility information value; The data risk parameter is obtained by multiplying the comprehensive network volatility information value by the data sensitivity parameter. Determine whether the data risk parameter is greater than or equal to the data risk parameter threshold. If yes, issue an early warning; otherwise, do not issue an early warning.

2. The data monitoring and early warning platform for mobile office environments according to claim 1, characterized in that, In a mobile office environment, multiple office data points of the user's current operation are monitored and acquired, along with multiple data feature information and multiple operation types of the aforementioned office data, including: In a mobile office environment, monitor and acquire multiple office data points related to the user's current actions; The data paths, data objects, and data targets of the multiple office data are obtained respectively and combined into multiple data feature information; Obtain the current operation types for the multiple office data.

3. The data monitoring and early warning platform for mobile office environments according to claim 1, characterized in that, Based on the multiple data feature information, a sensitivity analysis is performed on the multiple office data to obtain data sensitivity parameters, including: Build a sensitive classifier for office data; The multiple data feature information is input into the office data sensitivity classifier to classify and obtain multiple sensitivity levels of the multiple office data; Calculate the mean of the multiple sensitivity levels to obtain the data sensitivity parameter; The construction of an office data sensitivity classifier includes: Based on the office data processing records over a historical period, a set of sample data feature information is collected, and the sensitivity level of each sample data feature information is labeled to obtain a set of sample sensitivity levels. A random forest is used to construct a sensitive classifier for office data. The sample data feature information set and the sample sensitivity level set are divided to obtain a training set and a test set; The office data sensitivity classifier is trained under supervision using the training set and tested using the test set to determine whether it has converged. If yes, then training is complete and the office data sensitivity classifier is obtained; otherwise, supervised training continues until the office data sensitivity classifier converges.

4. The data monitoring and early warning platform for mobile office environments according to claim 1, characterized in that, Based on the multiple operation types, configure the upload weights and download weights for uploading and downloading network volatility information within the multiple network volatility information sets, including: The preset delay weight is subtracted from 1 to obtain the weight to be configured; The proportions of upload and download operation types within the multiple operation types are obtained, and then multiplied by the weights to be configured to obtain the upload weight and download weight of the upload volatility information and download volatility information.

Images