A method for protecting number privacy

Abstract

The present application relates to the field of information security technology and provides a number privacy protection method. The method includes: receiving a communication request from a first user and parsing it to obtain communication context information; predicting the degree of privacy demand based on the communication context information to generate a first privacy demand index; performing a trigger analysis of the privacy protection mode based on the first privacy demand index to generate a first privacy protection trigger mode; based on the first privacy protection trigger mode, calling a privacy protection module to configure virtual numbers and additional protection decisions for the first user and the first recipient, and establishing communication between the first user and the first recipient based on the configuration results. The present application solves the technical problem that existing privacy protection methods fail to dynamically evaluate privacy protection needs based on actual communication contexts, and achieves the technical effect of making personalized privacy protection decisions based on communication scenarios, recipient identities, and historical communication behaviors, thereby improving the flexibility and accuracy of number privacy protection.

Description

Technical Field

[0001] The present application relates to the field of information security technology, and in particular to a method for protecting number privacy. Background Art

[0002] With the rapid development of information technology and communication networks, the privacy protection of personal information has become a growing concern. In daily life, users frequently use mobile devices for various online activities, such as ordering food, calling ride-hailing services, and social media chats. During these activities, sensitive personal information such as users' phone numbers are often transmitted to merchants, service providers, or other communication partners, posing a significant risk of privacy breaches. While traditional privacy protection methods can provide a certain degree of security, they typically rely on static number encryption or the use of virtual numbers, making them inadequate for evolving communication scenarios and dynamic privacy needs. In scenarios like food delivery, ride-hailing services, and social media chats, users' privacy needs are not static but vary depending on the communication scenario, the recipient's identity, and the time and space of communication. For example, when ordering food, users may be concerned about whether their phone number is exposed to the merchant. When calling ride-hailing services, users have more stringent privacy requirements due to the potential exposure of their personal location and the associated increased security risks. Traditional privacy protection methods lack the flexibility to address these real-world scenarios and often employ fixed privacy protection models, failing to fully consider the privacy needs of diverse scenarios. Especially for activities like social chat, where communication is frequent and scenarios vary, existing technologies often fail to dynamically adjust privacy protection strategies in real time. Therefore, how to dynamically predict privacy protection needs based on actual communication scenarios and formulate appropriate protection measures accordingly has become an urgent problem to be solved. Summary of the Invention

[0003] This application provides a number privacy protection method, which aims to solve the technical problem that existing privacy protection methods fail to dynamically evaluate privacy protection needs based on actual communication context.

[0004] In view of the above problems, this application provides a number privacy protection method.

[0005] The present application provides a number privacy protection method, the method comprising: receiving a communication request from a first user and parsing the request to obtain communication context information, wherein the communication context information includes a communication scenario, an identity of a first recipient, communication spatiotemporal information, and historical communication records; performing a privacy requirement prediction based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records to generate a first privacy requirement index; performing a trigger analysis of a privacy protection mode based on the first privacy requirement index to generate a first privacy protection trigger mode; and, based on the first privacy protection trigger mode, invoking a privacy protection module to configure a virtual number and additional protection decision for the first user and the first recipient, and establishing communication between the first user and the first recipient according to the configuration result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The above-mentioned number privacy protection method first receives and parses a communication request from a first user to obtain communication context information, wherein the communication context information includes a communication scenario, an identity of a first recipient, communication spatiotemporal information, and historical communication records; then, based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records, a privacy demand degree prediction is performed to generate a first privacy demand degree index; further, based on the first privacy demand degree index, a trigger analysis of a privacy protection mode is performed to generate a first privacy protection trigger mode; finally, based on the first privacy protection trigger mode, a privacy protection module is called to configure a virtual number and additional protection decision for the first user and the first recipient, and communication between the first user and the first recipient is established according to the configuration result, thereby solving the technical problem that existing privacy protection methods fail to dynamically evaluate privacy protection needs according to the actual communication context, realizing personalized privacy protection decisions based on the communication scenario, the identity of the recipient, and historical communication behavior, and improving the flexibility and accuracy of number privacy protection.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 The figure is a flowchart of a number privacy protection method in one embodiment.

[0011] Figure 2 The figure is a flow chart of generating a first privacy requirement indicator in a number privacy protection method according to an embodiment. DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a number privacy protection method to solve the technical problem that existing privacy protection methods fail to dynamically evaluate privacy protection requirements based on actual communication contexts.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the present application provides a number privacy protection method, the method comprising:

[0016] A communication request from a first user is received and parsed to obtain communication context information, wherein the communication context information includes a communication scenario, an identity of the first recipient, communication time and space information, and historical communication records.

[0017] In an embodiment of the present application, when a first user initiates a communication, a communication request sent by the first user is received, and the communication context information of the current communication is parsed from the communication request of the first user, including the communication scenario, the identity of the first recipient, the time and space information of the communication, and the historical communication records. The communication scenario refers to the specific situation or activity type in which the communication occurs. For example, the first user may be ordering takeout, calling an online car-hailing service, or engaging in social chatting, etc. Each communication scenario has different privacy requirements. When ordering takeout, the first user usually wants to hide his or her personal number to avoid harassment from the merchant or delivery personnel, so the demand for privacy protection is relatively high. In social chatting, since the conversation partners are usually acquaintances or friends, the user may not pay much attention to privacy protection and the privacy demand is relatively low. For online car-hailing services, since real-time location and security issues are involved, the privacy demand is high and strict privacy protection is required. The identity of the first recipient refers to the identity category of the communication recipient, for example, a stranger (such as the other party in the social chat), a corporate service number (such as the customer service number of an online car-hailing company or a food delivery platform). The identity of the recipient will directly affect the privacy protection decision. For strangers, the first user may be more inclined to protect privacy. Communication context information refers to the time and location of communication. For example, whether the first user's communication request occurs at night, during holidays, or in certain specific areas (such as privacy-sensitive locations) can help assess potential privacy risks. If the communication occurs in certain high-risk areas (such as financial venues or sensitive areas) or at night, privacy protection policies may need to be strengthened to prevent information leakage or abuse. Historical communication records refer to the past communication history between the first user and the recipient, including communication frequency and interaction patterns. For example, whether the first user has contacted the same recipient multiple times, whether the intervals between contacts are short, or whether there has been no contact for a long time. This historical data helps determine the privacy needs of the current communication. Frequent and long-term interactions may indicate that the recipient is trustworthy and therefore has relatively low privacy protection needs, while short-term or occasional communications may require stricter privacy protection. The above four aspects of communication context information can provide a basis for subsequent dynamic assessment of user privacy needs, help make appropriate privacy protection decisions, and ensure the accuracy and flexibility of privacy protection.

[0018] A privacy requirement degree prediction is performed based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records to generate a first privacy requirement degree index.

[0019] In one embodiment, after parsing the communication scenario, the identity of the first recipient, the communication spatiotemporal information and the historical communication records from the communication context information, a privacy demand analysis is performed based on the communication scenario, the identity of the first recipient and the historical communication records through a privacy demand classifier, that is, whether the current communication of the first user requires privacy protection is determined, and an initial privacy demand degree is generated based on the judgment result. This initial privacy demand degree is then integrated with the risk indicator identified from the communication spatiotemporal information to generate a final first privacy demand degree indicator. This first privacy demand degree indicator quantifies the current communication's demand for privacy protection. The higher the first privacy demand degree indicator, the stricter the requirement for privacy protection. According to the first privacy demand degree, the privacy protection strategy can be dynamically adjusted, such as triggering different privacy protection modes, to ensure that the first user's privacy is accurately and effectively protected.

[0020] Further, if Figure 2 As shown, the present application provides a method for predicting the privacy requirement based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records, and generating a first privacy requirement index, including:

[0021] The method comprises the steps of: analyzing the communication frequency characteristics between the first user and the first recipient based on the historical communication records; calling a privacy requirement classifier to perform a privacy requirement analysis on the communication scenario, the identity of the first recipient, and the communication frequency characteristics to generate an initial privacy requirement degree; performing communication risk identification based on the communication spatiotemporal information to generate a first communication risk index; and weighting the initial privacy requirement degree with the first communication risk index to generate the first privacy requirement degree index.

[0022] Preferably, communication data related to the first user and the first recipient are retrieved from the historical communication records. These data include the timestamp of the communication, the type of communication (such as phone call, text message, video call, etc.), the duration of the communication, etc. In order to more accurately reflect the communication frequency characteristics, an analysis time window is defined. For example, the past week or month can be set as the time range, which is determined according to actual business needs; then, according to business accuracy requirements, multiple analysis time windows are set, and based on these analysis time windows, the communication data related to the first user and the first recipient in the historical communication records are divided. The division of the communication data is from the time of the last communication Starting from the stamp, the communication data corresponding to multiple analysis time windows are divided forward. In each analysis time window, the number of window communications between the first user and the first recipient is counted. This window communication number can reflect the frequency of interaction between the two. By calculating the average of the window communication numbers of multiple analysis time windows and then calculating the ratio of this average to the length of the analysis time window, the communication frequency feature between the first user and the first recipient is obtained. The larger the communication frequency feature, the more likely the first user and the first recipient are acquaintances or have a strong trust relationship, and the lower the privacy protection demand. On the contrary, it means that the first user and the first recipient are relatively unfamiliar with each other, and the privacy protection demand is low. The protection demand is high; then, the communication frequency feature, the current communication scenario, and the identity of the first recipient are input into the privacy demand classifier for privacy demand analysis. The privacy demand classifier maps the communication frequency feature, the communication scenario, and the identity of the first recipient to a plane according to the learned mapping relationship, and determines whether the first user's current communication requires privacy protection. If the first user requires privacy protection, the privacy demand classifier outputs a preset privacy demand degree (such as 1, 0.5, etc.) as the initialization privacy demand degree. If the first user does not need privacy protection, the privacy demand classifier outputs 0 as the initialization privacy demand degree; then, The existing privacy risk label database is used to compare the spatiotemporal information of communications, and the specific time and location of the communication are analyzed to determine whether they are in high-risk areas or time periods, such as at night or in hospitals. Based on the analysis process, a first communication risk indicator is generated to quantify the degree of privacy risk of the current communication. Finally, the generated first communication risk indicator is weighted and summed with the initial privacy requirement (the weights of the first communication risk indicator and the initial privacy requirement are determined based on historical experience and expert decisions) to calculate the first privacy requirement indicator. This first privacy requirement indicator can reflect the comprehensive privacy protection needs and provide a basis for triggering subsequent privacy protection modes.

[0023] The privacy requirement classifier can be built based on support vector machines or other machine learning algorithms, such as decision trees, random forests, neural networks, etc. When using support vector machines to build it, historical data containing features such as communication scenarios, recipient identities, historical communication records, and corresponding privacy protection situations are first collected. The historical communication records are processed in the same way as above and combined with the communication scenarios and recipient identities to obtain sample data. These sample data are identified according to the privacy protection situation, that is, whether these sample data are privacy protected. The identified sample data are then divided into training sets and test sets, usually 80% as training sets and 20% as test sets. Subsequently, support vector machines (SVM) are selected as the classification model, and radial basis kernels (RBF kernels) are used to process nonlinear features. The parameters of the support vector machine model are initialized, including kernel functions, setting C parameters and gamma parameters, etc. The initial parameters of the model are set by random initialization or preset values. Afterwards, the support vector machine model is tested using the training set. During training, the support vector machine model constructs an optimal hyperplane to classify the data by inputting the characteristic data of the training set (including communication scenarios, recipient identities, and communication frequencies), and distinguishes between two types of data: data that requires privacy protection and data that does not. During the training process, the support vector machine optimization algorithm is also used to find the optimal classification boundary by maximizing the classification interval. Then, cross-validation is used to adjust the C parameter and gamma parameter to optimize the classification effect of the model to avoid overfitting and underfitting. After training is completed, the trained model is evaluated using the test set. Evaluation indicators such as prediction accuracy, precision, recall rate, and F1 score are calculated to ensure that the model can correctly identify whether privacy protection is required. During the evaluation process, if the performance of the model on the test set does not meet expectations, hyperparameters such as the learning rate, C value, and gamma value are adjusted to further optimize the model until the optimal effect is achieved. Finally, the trained support vector machine model is used as the output of the privacy requirement classifier for subsequent privacy requirement analysis.

[0024] Furthermore, the present application provides performing communication risk identification based on the communication spatiotemporal information to generate a first communication risk indicator, including:

[0025] An existing privacy risk label database is called to compare the communication spatiotemporal information to generate a comparison result; and risk calculation is performed based on the comparison result to generate the first communication risk indicator.

[0026] Optionally, when identifying communication risks based on communication spatiotemporal information, time information (such as early morning, late night, weekdays, or holidays) and geographic location information (such as the current location of the first recipient or the communication target address) are first extracted from the communication spatiotemporal information. Subsequently, a pre-built privacy risk label database is called, which stores privacy risk labels related to time and geographic location and corresponding risk scores, such as high-risk time labels and time risk scores (late night period: 0.6 points), high-risk area labels and location risk scores (high-incidence areas of fraud: 0.9 points, sensitive geographic locations: 0.6 points), etc. Subsequently, the extracted time information and geographic location information are compared with the privacy risk label database. Based on the time when the communication occurs, it is checked whether it falls within the range of high-risk time labels. Based on the geographic location information, it is checked whether it is located in the high-risk area labels. Based on the generated comparison results, the risk of the communication spatiotemporal information of the first user is calculated, that is, the time risk score and location risk score in the comparison results are weighted and summed to obtain the first communication risk indicator of the first user, which is used for subsequent evaluation of the degree of privacy protection demand to ensure the dynamic and accurate nature of privacy protection.

[0027] A trigger analysis of a privacy protection mode is performed based on the first privacy requirement indicator to generate a first privacy protection trigger mode.

[0028] In one embodiment, after calculating the first privacy requirement indicator, the first privacy requirement indicator is matched with a privacy protection mode triggering rule. The privacy protection mode triggering rule includes a privacy requirement range corresponding to each privacy protection mode (such as the original number communication mode and the private number communication mode). By comparing the first privacy requirement indicator with the privacy requirement range of each privacy protection mode, the privacy requirement range within which the first privacy requirement indicator falls is identified, and the corresponding privacy protection mode is triggered according to the privacy requirement range. The triggered privacy protection mode will serve as the first privacy protection triggering mode and be applied in subsequent communication processes to ensure the privacy security of the communication.

[0029] Furthermore, the present application provides that the privacy protection mode includes an original number communication mode and a privacy number communication mode, and the privacy number communication mode includes a first-level privacy mode and a second-level privacy mode; wherein, the first-level privacy mode is a privacy mode that maps the original number to a virtual number, and the second-level privacy mode refers to a privacy mode that combines a virtual number with two-factor authentication.

[0030] Preferably, the privacy protection mode includes an original number communication mode and a private number communication mode. The original number communication mode is used when the first user's privacy needs are low, allowing the first user's original number to be directly used to establish communication with the recipient. In this mode, due to the low privacy risk, no additional protection measures are required, and the basic needs of the first user can be met while ensuring communication convenience. The private number communication mode is used when the first user's privacy needs are relatively high, and is used to hide the first user's original number and protect it with a virtual number. The private number communication mode is subdivided into a first-level privacy mode and a second-level privacy mode. The first-level privacy mode is used when the privacy needs are medium. It uses a virtual number mapping method to replace the first user's original number with a virtual number, and establish communication with the recipient through the virtual number. After the communication ends, the virtual number is automatically destroyed to ensure that the first user's real number is not leaked. The second-level privacy mode is used when the privacy needs are high. On the basis of virtual number mapping, it combines two-factor authentication to further enhance protection. In this mode, a one-time verification code is generated and sent to the first recipient via the virtual number. The first recipient must successfully verify before communication can be established. This mode can protect the privacy of the first user while increasing the security of communication, and is suitable for scenarios with high privacy requirements.

[0031] Based on the first privacy protection trigger mode, the privacy protection module is called to configure virtual numbers and additional protection decisions for the first user and the first recipient, and communication between the first user and the first recipient is established according to the configuration results.

[0032] In one embodiment, after determining the first privacy protection trigger mode, the privacy protection policy corresponding to the mode will be obtained. If the first privacy protection trigger mode is the original number communication mode, the privacy protection module will be called to directly establish communication between the first user and the first recipient. If the first privacy protection trigger mode is the first-level privacy mode, the privacy protection module will be started to configure a virtual number to meet basic privacy protection requirements, and communication between the first user and the first recipient will be established based on the configured virtual number. If the first privacy protection trigger mode is the second-level privacy mode, a two-factor authentication module will be performed on the basis of the virtual number configuration, that is, a one-time verification code will be generated and sent to the first recipient. Only after the first recipient completes the verification code verification will the communication between the first user and the first recipient be established. Through the above process, the privacy needs of the first user can be dynamically adapted, and the virtual number and additional protection measures can be combined to provide the user with a safe and reliable communication service.

[0033] Furthermore, the present application provides a method for invoking a privacy protection module to configure a virtual number and additional protection decision for the first user and the first recipient based on the first privacy protection trigger mode, and establishing communication between the first user and the first recipient based on the configuration result, including:

[0034] If the first privacy protection trigger mode is the original number communication mode, communication is established using the original numbers of the first user and the first recipient.

[0035] Optionally, when the first privacy protection trigger mode is the original number communication mode, it means that the privacy requirements of the current communication scenario are low and no additional privacy protection measures are required. At this time, the privacy protection module will skip the virtual number mapping and additional protection configuration steps and directly create a communication channel for both parties through the real number provided by the first user. This mode is suitable for the first user to communicate with a trusted first recipient. In this way, the communication establishment process can be simplified to ensure the timeliness and convenience of communication.

[0036] Furthermore, this application also includes:

[0037] If the first privacy protection trigger mode is the first-level privacy mode in the privacy number communication mode, call the virtual number pool to perform virtual number mapping for the original numbers of the first user and the first recipient, and generate the first user virtual number and the first recipient virtual number; establish communication between the first user and the first recipient using the first user virtual number and the first recipient virtual number, and destroy the virtual number after the communication ends.

[0038] Optionally, when the first privacy protection trigger mode is the first-level privacy mode in the privacy number communication mode, the privacy protection module will call the virtual number pool to assign virtual numbers to the original numbers of the first user and the first recipient to ensure that the real numbers are not leaked. Specifically, the privacy protection module will randomly extract two virtual numbers from the virtual number pool to map the original numbers of the first user and the first recipient, respectively as the first user virtual number and the first recipient virtual number. The virtual numbers serve as intermediate identifiers to replace the real numbers in communication, thereby protecting the privacy of both parties. Subsequently, the generated first user virtual number and first recipient virtual number are used to establish a communication channel between the two parties. During the communication process, the two parties can only see each other's virtual number and cannot obtain the real number, ensuring privacy security. When the communication is completed, the privacy protection module will immediately recycle and destroy the assigned virtual number. This process ensures the temporary and one-time use of the virtual number, preventing the information from being abused or stored. Through the above process, the privacy function of protecting the user's real number through virtual numbers is realized in the first-level privacy mode, while simplifying the privacy protection mechanism to meet scenarios with medium privacy requirements.

[0039] Furthermore, the present application provides that if the first privacy protection trigger mode is the secondary privacy mode in the privacy number communication mode, after generating the first user virtual number and the first recipient virtual number, the method further includes:

[0040] Activate the two-factor authentication module, generate a one-time verification code, and send it to the first recipient via the first recipient's virtual number; verify the verification code received by the first recipient via the two-factor authentication module. If the verification is successful, establish communication between the first user and the first recipient using the first user's virtual number and the first recipient's virtual number, and destroy the virtual numbers after the communication ends.

[0041] Optionally, when the first privacy protection trigger mode is the secondary privacy mode in the privacy number communication mode, the privacy protection module will generate the first user virtual number and the first recipient virtual number in the same way as above. On the basis of the first user virtual number and the first recipient virtual number, the privacy protection module will activate the two-factor authentication module to further enhance the security of communication by generating and verifying a one-time verification code. Specifically, the privacy protection module will generate a one-time verification code for the current communication based on the activated two-factor authentication module. The verification code is a temporary security code used to verify the identity of the first recipient and ensure the legitimacy of the communication. Subsequently, the privacy protection module will send the generated one-time verification code to the first recipient through the first recipient virtual number. The first recipient needs to receive and correctly enter the verification code to complete identity authentication. After receiving the verification code returned by the first recipient, the two-factor authentication module will compare it with the generated verification code. If the verification code is successfully verified, it means that the identity of the first recipient is legitimate, and the privacy protection module will allow communication to be established. After the identity authentication is passed, the privacy protection module uses the first user's virtual number and the first recipient's virtual number to establish a communication channel between the two parties. During the entire communication process, the real numbers of both parties are hidden, and the communication is transferred through the virtual number to ensure privacy protection. When the communication ends, the privacy protection module recycles and destroys the used virtual number to ensure that the virtual number will not be reused or abused, thereby further enhancing the security of privacy protection. Through the above process, double privacy protection is achieved in the secondary privacy mode. Not only is the real number hidden through virtual number mapping, but the legitimacy of the recipient's identity is also ensured through two-factor authentication, effectively reducing privacy leaks and security risks.

[0042] Furthermore, the present application provides that after establishing communication between the first user and the first recipient, the method further includes:

[0043] Acquire any virtual number generated when establishing a communication; introduce a behavior monitoring module to analyze abnormal usage behavior of any virtual number in real time; wherein, the behavior monitoring module is constructed based on historical abnormal call behavior, historical abnormal SMS behavior, historical abnormal communication pattern, and historical geographical and time abnormal pattern; and issue an abnormal reminder signal based on the abnormal usage behavior.

[0044] Optionally, when the first user establishes communication with the first recipient, the privacy protection module generates and allocates a virtual number, which is used to replace the real number to protect the privacy of both parties. The privacy protection module obtains any virtual number assigned to the communicating parties from the virtual number pool and binds it to the communication channel; then, a behavior monitoring module is introduced to monitor the usage behavior of the virtual number in real time. This module is a pre-trained model that continuously tracks the activities of the virtual number and analyzes whether it conforms to the normal usage pattern, thereby identifying abnormal usage behavior, such as abnormal call behavior (high-frequency calls, etc.) and abnormal SMS behavior (repeated SMS, etc.); when constructing the behavior monitoring module, historical abnormal call behavior, historical Historical abnormal SMS behaviors, historical abnormal communication patterns, historical geographical and temporal abnormal patterns, and historical virtual number behavior activities corresponding to these anomalies. Among them, historical abnormal call behaviors refer to the calling behaviors of virtual numbers in the historical records, including whether the number of calls per hour is abnormal and whether the call duration exceeds the normal range. For example, multiple calls in a short period of time or a single call of extremely long duration will be regarded as abnormal behaviors. Historical abnormal SMS behaviors refer to the SMS sending patterns of virtual numbers in the historical records, including whether the frequency of SMS sending is too high and whether the SMS content is repeated. For example, if a large number of SMS messages with similar content (such as spam SMS messages) are sent in a short period of time, it may be regarded as abnormal behaviors. Historical abnormal communication patterns refer to the SMS sending patterns of virtual numbers in the historical records. The communication objects and pattern characteristics of the virtual number, for example, whether multiple repeated communications with the same number exceed the normal range; if there are frequent communications with a certain number in a short period of time, or the communication frequency is significantly higher than the average level, it may be regarded as abnormal behavior; historical geographical and temporal abnormal patterns refer to the use location and time characteristics of the virtual number in the historical records, for example, whether cross-regional behavior occurs at the same time as high-frequency communication; if the communication occurs in a scenario with drastic changes in geographical location (such as across cities), or the time period is concentrated in uncommon time periods such as late at night, it may be regarded as abnormal behavior; by combining the historical virtual number behavior activities and the corresponding anomalies, dividing the training set and the verification set, and then using the training set to train the constructed initial behavior monitoring module, the training After completion, the validation set is used for verification to evaluate the anomaly detection performance of the initial behavior monitoring module. The initial behavior monitoring module can be constructed based on neural networks, support vector machines, random forests, etc. Taking the long short-term memory network (LSTM) in the neural network as an example, an initial behavior monitoring module structure is constructed, including an input layer, an LSTM hidden layer, a fully connected layer, and an output layer. During the training process, the training data is used for forward propagation, and the input virtual number behavior activities are passed through the LSTM layer layer by layer to generate the prediction results for each time step. The cross-entropy loss function is then used to calculate the error between the prediction result and the actual label (abnormal behavior). The gradient of the loss with respect to the weight of each layer is calculated layer by layer through the backpropagation algorithm.Afterwards, the Adam optimizer is used to optimize the weights to minimize the value of the loss function, and this process is repeated until the maximum number of iterations is reached or the loss function converges. After training, the initial behavior monitoring module is tested using the validation set to evaluate its detection accuracy for abnormal call behavior, SMS behavior, communication patterns, and geographical and temporal anomalies. If the accuracy meets expectations, the current initial behavior monitoring module will be output as the final behavior monitoring module. Otherwise, hyperparameters such as the learning rate, number of LSTM layers, and number of hidden units are adjusted to further optimize the detection capability of the behavior monitoring module. Finally, the behavioral activities of the virtual number collected in real time are input into the trained behavior monitoring module. Based on the input data, the module will automatically detect whether the virtual number has abnormal behavior, such as abnormal call behavior, abnormal SMS behavior, abnormal communication pattern, etc. If an anomaly exists, an abnormal usage behavior will be generated, and an abnormal reminder signal will be triggered based on this abnormal usage behavior. This signal can be sent to the first user via SMS, email, or other notification methods to alert them of potential privacy risks and remind them to check or stop related communications. In this way, the use of virtual numbers can be monitored in real time, and potential security threats or abuse can be detected and responded to promptly, thereby further enhancing privacy protection during communications.

[0045] In summary, the embodiments of the present application have at least the following technical effects:

[0046] The embodiment of the present application receives and parses the communication request of the first user to obtain communication context information, wherein the communication context information includes the communication scenario, the identity of the first recipient, the communication time and space information and the historical communication records, and then performs a privacy demand prediction based on the communication scenario, the identity of the first recipient, the communication time and space information and the historical communication records to generate a first privacy demand index; further, a trigger analysis of the privacy protection mode is performed based on the first privacy demand index to generate a first privacy protection trigger mode; finally, based on the first privacy protection trigger mode, the privacy protection module is called to configure the virtual number and additional protection decision for the first user and the first recipient, and establish communication between the first user and the first recipient according to the configuration result. These technical effects jointly solve the technical problem that the existing privacy protection method fails to dynamically evaluate the privacy protection needs according to the actual communication context, and realizes the technical effect of making personalized privacy protection decisions based on the communication scenario, the identity of the recipient and the historical communication behavior, thereby improving the flexibility and accuracy of number privacy protection.

[0047] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0049] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A number privacy protection method, characterized in that: include: Receive and parse a communication request from a first user to obtain communication context information, wherein the communication context information includes a communication scenario, an identity of the first recipient, communication time and space information, and historical communication records; Predicting the degree of privacy requirement based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records to generate a first privacy requirement index; Performing a trigger analysis of a privacy protection mode based on the first privacy requirement indicator to generate a first privacy protection trigger mode; Based on the first privacy protection trigger mode, the privacy protection module is called to configure virtual numbers and additional protection decisions for the first user and the first recipient, and communication between the first user and the first recipient is established according to the configuration results.

2. A number privacy protection method according to claim 1, characterized in that: The privacy protection mode includes an original number communication mode and a privacy number communication mode, and the privacy number communication mode includes a first-level privacy mode and a second-level privacy mode; The first-level privacy mode is a privacy mode that maps the original number to a virtual number, and the second-level privacy mode is a privacy mode that combines the virtual number with two-factor authentication.

3. A number privacy protection method according to claim 1, characterized in that: Predicting the privacy requirement based on the communication scenario, the identity of the first recipient, the communication spatiotemporal information, and historical communication records to generate a first privacy requirement index includes: analyzing a communication frequency characteristic between the first user and the first recipient based on the historical communication records; Calling a privacy requirement classifier to perform a privacy requirement analysis on the communication scenario, the identity of the first recipient, and the communication frequency characteristics to generate an initial privacy requirement degree; Performing communication risk identification based on the communication spatiotemporal information to generate a first communication risk indicator; The initialization privacy requirement is weighted by the first communication risk indicator to generate the first privacy requirement indicator.

4. A number privacy protection method according to claim 3, characterized in that: Performing communication risk identification based on the communication spatiotemporal information to generate a first communication risk indicator includes: Calling an existing privacy risk label database to compare the communication spatiotemporal information and generate a comparison result; Perform risk calculation based on the comparison result to generate the first communication risk indicator.

5. A number privacy protection method according to claim 2, characterized in that: Based on the first privacy protection trigger mode, calling a privacy protection module to configure a virtual number and an additional protection decision for the first user and the first recipient, and establishing communication between the first user and the first recipient according to the configuration result, including: If the first privacy protection trigger mode is the original number communication mode, communication is established using the original numbers of the first user and the first recipient.

6. A number privacy protection method according to claim 5, characterized in that: Also includes: If the first privacy protection trigger mode is the first level privacy mode in the privacy number communication mode, calling the virtual number pool to perform virtual number mapping for the original numbers of the first user and the first recipient, and generating a first user virtual number and a first recipient virtual number; Communication between the first user and the first recipient is established using the first user virtual number and the first recipient virtual number, and the virtual numbers are destroyed after the communication ends.

7. A number privacy protection method according to claim 6, characterized in that: If the first privacy protection trigger mode is the secondary privacy mode in the privacy number communication mode, after generating the first user virtual number and the first recipient virtual number, the method further includes: activating a two-factor authentication module, generating a one-time verification code, and sending the code to the first recipient via the first recipient's virtual number; The verification code received by the first recipient is verified by the two-factor authentication module. If the verification is successful, communication is established between the first user and the first recipient using the first user virtual number and the first recipient virtual number, and the virtual numbers are destroyed after the communication ends.

8. A number privacy protection method according to claim 7, characterized in that: After establishing communication between the first user and the first recipient, the method further includes: Obtain any virtual number generated by establishing communication; Introducing a behavior monitoring module to analyze abnormal usage behavior of any virtual number in real time; The behavior monitoring module is constructed based on historical abnormal call behavior, historical abnormal SMS behavior, historical abnormal communication patterns, and historical geographical and temporal abnormal patterns; An abnormal reminder signal is issued based on the abnormal usage behavior.

Images