Privacy information automatic identification and encryption method and system based on user behavior characteristics

Abstract

The invention discloses a privacy information automatic identification and encryption method and system based on user behavior characteristics. The method comprises the following steps: obtaining a to-be-processed client log corresponding to a client; determining a privacy identification model according to the historical behavior data of the client; performing field identification on the client log according to the privacy identification model to obtain a plurality of privacy fields; and according to an encryption parameter corresponding to the client, performing encryption processing on the privacy field to obtain an encrypted client log. Visibly, according to the method and the device, self-adaptive precise log privacy protection and encryption based on user behaviors can be realized, the security and compliance of client log uploading are improved, and the risk of data leakage caused by missing identification of privacy fields is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for automatic identification and encryption of privacy information based on user behavior characteristics. Background Technology

[0002] With the rapid growth of client-side data security and privacy compliance requirements, enterprises and developers are increasingly focusing on protecting user data privacy through accurate log processing. A key technical challenge is how to adaptively identify and encrypt privacy fields to reduce the risk of data leakage. Existing technologies typically collect client-side log data, scan and process sensitive fields using fixed privacy rules or general encryption algorithms, and upload logs based on standard protocols to meet basic compliance requirements. However, existing solutions lack dynamic analysis of historical behavioral data and adaptive determination of privacy identification models, making it difficult to accurately identify privacy fields in logs and optimize encryption parameters. Commonly used static privacy protection strategies cannot adapt to changes in user behavior, resulting in insufficient accuracy in log privacy protection. This makes them susceptible to data leakage due to missed identification of privacy fields, limiting the security and compliance of client-side log uploads. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for automatic identification and encryption of privacy information based on user behavior characteristics, which can realize accurate log privacy protection and encryption based on user behavior, improve the security and compliance of client log uploads, and reduce the risk of data leakage due to failure to identify privacy fields.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for automatically identifying and encrypting privacy information based on user behavior characteristics, the method comprising: Retrieve the client logs to be processed corresponding to the client; A privacy identification model is determined based on the client's historical behavior data; The client logs are analyzed using the privacy identification model to obtain multiple privacy fields. Based on the encryption parameters corresponding to the client, the privacy field is encrypted to obtain the encrypted client log.

[0005] As an optional implementation, in the first aspect of the invention, the client log includes multiple log event texts.

[0006] As an optional implementation, in the first aspect of the present invention, the historical behavior data includes historical operation data of multiple users on the client; the historical operation data includes multiple operation behaviors and corresponding operation text, operation location and operation time.

[0007] As an optional implementation, in the first aspect of the invention, determining the privacy identification model based on the client's historical behavior data includes: For each operation behavior in each of the historical operation data, the operation location and operation time corresponding to the operation behavior are determined as the operation parameters of the operation behavior; The behavioral parameters are input into the trained privacy behavior recognition model to obtain the privacy probability that the behavioral parameters belong to privacy processing behaviors; the privacy behavior recognition model is a classifier model, which is trained on a training dataset that includes multiple training user operation parameters and corresponding privacy processing behavior annotations; The operation text corresponding to all operation behaviors whose privacy probability is greater than a preset probability threshold is determined as privacy text; A privacy identification model is determined based on the user parameters and privacy text corresponding to each of the historical operation data.

[0008] As an optional implementation, in the first aspect of the invention, determining the privacy identification model based on the user parameters of the user corresponding to each of the historical operation data and the privacy text includes: For each candidate privacy recognition model, the privacy text corresponding to all the historical operation data is input into the candidate privacy recognition model to obtain the average value of the output privacy probability, and thus the prediction accuracy. The average value of the parameter similarity between the user parameter annotations in the training data of the candidate privacy recognition model and the user parameters of the user corresponding to each historical operation data is calculated to obtain the user similarity. The priority of the candidate privacy recognition model is obtained by calculating the product of the prediction accuracy and the user similarity. The candidate privacy identification model with the highest priority is determined as the privacy identification model.

[0009] As an optional implementation, in the first aspect of the present invention, the privacy recognition model is a neural network model, which is trained using a training dataset that includes multiple training user texts and corresponding privacy annotations.

[0010] As an optional implementation, in the first aspect of the invention, the step of performing field identification on the client logs according to the privacy identification model to obtain multiple privacy fields includes: Each field in each log event text of the client log is input into the privacy identification model to obtain the predicted privacy level corresponding to each field; For each log time text, calculate the weighted average of the predicted privacy level of all fields in the log time text to obtain the text privacy level corresponding to the log event text; The log event text whose text privacy level is greater than a preset first level threshold is identified as a privacy field; Fields in log event texts whose predicted privacy level is greater than the second threshold but less than the first threshold are identified as privacy fields. All fields of the log event text whose text privacy level is less than the second threshold are not considered privacy fields.

[0011] As an optional implementation, in the first aspect of the present invention, the calculation weight corresponding to each predicted privacy level is the product of a first weight and a second weight; the first weight is proportional to the number of characters in the corresponding field; and the second weight is proportional to the proportion of the occurrence of the field type of the corresponding field in the client log.

[0012] A second aspect of this invention discloses an automatic identification and encryption system for privacy information based on user behavior characteristics, the system comprising: The acquisition module is used to acquire the client logs to be processed corresponding to the client. The determination module is used to determine a privacy identification model based on the client's historical behavior data; The identification module is used to identify fields in the client logs according to the privacy identification model to obtain multiple privacy fields; The encryption module is used to encrypt the privacy field according to the encryption parameters corresponding to the client to obtain the encrypted client log.

[0013] As an optional implementation, in a second aspect of the invention, the client log includes multiple log event texts.

[0014] As an optional implementation, in a second aspect of the present invention, the historical behavior data includes historical operation data of multiple users on the client; the historical operation data includes multiple operation behaviors and corresponding operation text, operation location and operation time.

[0015] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method of the privacy identification model based on the client's historical behavior data, including: For each operation behavior in each of the historical operation data, the operation location and operation time corresponding to the operation behavior are determined as the operation parameters of the operation behavior; The behavioral parameters are input into the trained privacy behavior recognition model to obtain the privacy probability that the behavioral parameters belong to privacy processing behaviors; the privacy behavior recognition model is a classifier model, which is trained on a training dataset that includes multiple training user operation parameters and corresponding privacy processing behavior annotations; The operation text corresponding to all operation behaviors whose privacy probability is greater than a preset probability threshold is determined as privacy text; A privacy identification model is determined based on the user parameters and privacy text corresponding to each of the historical operation data.

[0016] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method of the privacy identification model based on the user parameters of the user corresponding to each of the historical operation data and the privacy text, including: For each candidate privacy recognition model, the privacy text corresponding to all the historical operation data is input into the candidate privacy recognition model to obtain the average value of the output privacy probability, and thus the prediction accuracy. The average value of the parameter similarity between the user parameter annotations in the training data of the candidate privacy recognition model and the user parameters of the user corresponding to each historical operation data is calculated to obtain the user similarity. The priority of the candidate privacy recognition model is obtained by calculating the product of the prediction accuracy and the user similarity. The candidate privacy identification model with the highest priority is determined as the privacy identification model.

[0017] As an optional implementation, in the second aspect of the present invention, the privacy recognition model is a neural network model, which is trained using a training dataset that includes multiple training user texts and corresponding privacy annotations.

[0018] As an optional implementation, in a second aspect of the invention, the specific method by which the identification module performs field identification on the client logs according to the privacy identification model to obtain multiple privacy fields includes: Each field in each log event text of the client log is input into the privacy identification model to obtain the predicted privacy level corresponding to each field; For each log time text, calculate the weighted average of the predicted privacy level of all fields in the log time text to obtain the text privacy level corresponding to the log event text; The log event text whose text privacy level is greater than a preset first level threshold is identified as a privacy field; Fields in log event texts whose predicted privacy level is greater than the second threshold but less than the first threshold are identified as privacy fields. All fields of the log event text whose text privacy level is less than the second threshold are not considered privacy fields.

[0019] As an optional implementation, in a second aspect of the invention, the calculation weight corresponding to each predicted level of privacy is the product of a first weight and a second weight; the first weight is proportional to the number of characters in the corresponding field; and the second weight is proportional to the proportion of the occurrences of the field type of the corresponding field in the client log.

[0020] A third aspect of this invention discloses another automatic identification and encryption system for privacy information based on user behavior characteristics, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the privacy information automatic identification and encryption method based on user behavior characteristics disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the method for automatic identification and encryption of privacy information based on user behavior characteristics disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention acquires client-side logs to be processed and determines a privacy identification model based on historical behavior data. It identifies privacy fields in the log fields and encrypts them based on encryption parameters. This enables precise log privacy protection and encryption based on user behavior, improving the security and compliance of client-side log uploads and reducing the risk of data leakage due to missed identification of privacy fields. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating an automatic identification and encryption method for privacy information based on user behavior characteristics disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of an automatic identification and encryption system for privacy information based on user behavior characteristics disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another privacy information automatic identification and encryption system based on user behavior characteristics disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for automatic identification and encryption of privacy information based on user behavior characteristics. By acquiring client-side logs to be processed and determining a privacy identification model based on historical behavior data, privacy fields are identified from the log fields and encrypted using encryption parameters. This enables accurate log privacy protection and encryption based on user behavior adaptation, improving the security and compliance of client-side log uploads and reducing the risk of data leakage due to missed identification of privacy fields. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating an automatic identification and encryption method for privacy information based on user behavior characteristics, as disclosed in an embodiment of the present invention. Figure 1 The described method for automatic identification and encryption of privacy information based on user behavior characteristics can be applied to data processing systems / data processing devices / data processing servers (wherein, the server includes local processing servers or cloud processing servers). Figure 1 As shown, this method for automatically identifying and encrypting privacy information based on user behavior characteristics may include the following operations: 101. Obtain the client logs to be processed corresponding to the client.

[0032] Optionally, the client log can be Android / iOS application log, web front-end log, Windows desktop log, or cross-platform Electron log; this invention does not impose any limitations.

[0033] Optionally, the client logs to be processed can be real-time uploaded logs, batch uploaded logs, or locally cached logs; this invention does not impose any limitations.

[0034] Optionally, the client logs include multiple log event texts.

[0035] 102. Determine the privacy identification model based on the client's historical behavior data. Optionally, the historical behavior data may include user operation sequences, access field records, and sensitive data interaction logs; this invention does not impose any limitations on these data.

[0036] Optionally, historical behavior data includes historical operation data of multiple users on the client.

[0037] Optionally, historical operation data includes multiple operation actions and their corresponding operation text, operation location, and operation time.

[0038] 103. Based on the privacy identification model, perform field identification on the client logs to obtain multiple privacy fields. Optionally, the privacy field may include mobile phone number, ID card number, bank card number, address, name, device ID, location coordinates, or user email address; this invention does not impose any limitations.

[0039] 104. Based on the encryption parameters corresponding to the client, encrypt the privacy fields to obtain the encrypted client logs.

[0040] Optionally, the encryption parameter can be an AES-256-GCM key, an SM4 key, or an RSA public key; this invention does not impose any limitations.

[0041] Optionally, the encryption process can be field-level encryption, whole log text encryption, or partial encryption; this invention does not impose any limitations.

[0042] As can be seen, the above-described embodiments of the invention obtain client logs to be processed and determine a privacy identification model based on historical behavior data. By identifying privacy fields in the log fields and encrypting them based on encryption parameters, the invention can achieve accurate log privacy protection and encryption based on user behavior, improve the security and compliance of client log uploads, and reduce the risk of data leakage due to the failure to identify privacy fields.

[0043] As an optional embodiment, the step above, determining the privacy identification model based on the client's historical behavior data, includes: For each operation behavior in each historical operation data, the operation location and operation time corresponding to the operation behavior are determined as the operation parameters of the operation behavior; The behavioral parameters are input into the trained privacy behavior recognition model to obtain the privacy probability that the behavioral parameters belong to the privacy processing behavior.

[0044] The operation text corresponding to all operation behaviors with a privacy probability greater than a preset probability threshold is identified as privacy text; A privacy identification model is determined based on the user parameters and privacy text corresponding to each historical operation data.

[0045] Optionally, the operation location can be page coordinates, control ID, or element path, and the operation time can be an operation timestamp; this invention does not impose any limitations.

[0046] Optionally, the privacy behavior recognition model is a classifier model, which is trained using a training dataset that includes multiple training user operation parameters and corresponding privacy processing behavior annotations.

[0047] Optionally, the privacy behavior recognition model can be a 3-layer fully connected neural network (128-dimensional input, 256-128 hidden layers, and 1-dimensional sigmoid output), trained on 500,000 labeled user operation sequences, achieving an AUC of 0.97. This invention does not impose any limitations.

[0048] Optionally, the operation text can be the content of an input box, a clicked menu item, or copied sensitive text; this invention does not impose any limitations.

[0049] As can be seen, through the above optional embodiments, by inputting the operation location and time of historical operation behavior as behavior parameters into the privacy behavior recognition model to calculate the privacy probability, selecting high-probability operation text as privacy text and combining it with user parameters to determine the privacy recognition model, accurate adaptive model selection based on behavior patterns and user characteristics is achieved, improving the targeting and accuracy of privacy field recognition, and reducing the risk of privacy misjudgment caused by the generality of the model.

[0050] As an optional embodiment, the step above, determining the privacy identification model based on the user parameters and privacy text corresponding to each historical operation data, includes: For each candidate privacy recognition model, the privacy text corresponding to all historical operation data is input into the candidate privacy recognition model to obtain the average value of the output privacy probability, and thus the prediction accuracy. The average parameter similarity between the user parameter annotations in the training data of the candidate privacy recognition model and the user parameters of the user corresponding to each historical operation data is calculated to obtain the user similarity. The priority of the candidate privacy recognition model is obtained by calculating the product of the prediction accuracy and the user similarity. The candidate privacy identification model with the highest priority is selected as the privacy identification model.

[0051] Optionally, the user parameters may include user level, account type, region, device type, or frequency of historical privacy operations; this invention does not impose any limitations.

[0052] Optionally, the similarity parameter can be cosine similarity, Euclidean distance, or Jaccard coefficient; this invention does not impose any limitation on it.

[0053] Optionally, the privacy recognition model is a neural network model, trained using a training dataset that includes multiple training user texts and corresponding privacy annotations.

[0054] Optionally, the privacy recognition model can be a BERT-BiLSTM-CRF model, a RoBERTa+CRF model, or a DeBERTa-v3-large fine-tuning model; this invention does not impose any limitations.

[0055] As can be seen, through the above optional embodiments, by calculating the product of the prediction accuracy of the candidate privacy recognition model on historical privacy text and the similarity of user parameters as the priority, the model with the highest priority is selected as the privacy recognition model, thereby achieving accurate model selection based on dual optimization of accuracy and user matching, improving the individual adaptability and reliability of privacy recognition, and reducing the risk of recognition bias caused by model mismatch.

[0056] As an optional embodiment, in the above steps, the process of identifying fields in the client logs according to the privacy identification model to obtain multiple privacy fields includes: Each field in the text of each log event in the client log is input into the privacy identification model to obtain the predicted privacy level for each field; For each log time text, calculate the weighted average of the predicted privacy levels of all fields in that log time text to obtain the text privacy level corresponding to that log event text; Log event texts with a privacy level greater than a preset first level threshold are identified as a privacy field; Fields in log event texts with a text privacy level greater than the preset second threshold but less than the first threshold, whose predicted privacy level is greater than the second threshold, are identified as privacy fields. All fields in log event text with a text privacy level below the second threshold will be excluded from privacy fields.

[0057] Optionally, the predicted privacy level is a continuous probability value of 0-1, which is not limited in this invention.

[0058] Optionally, the calculated weight corresponding to each predicted level of privacy is the product of the first weight and the second weight.

[0059] Optionally, the first level threshold can be a high-risk threshold, but this invention does not limit it.

[0060] Optionally, the second risk threshold can be a medium-risk threshold, which is not limited in this invention.

[0061] Optionally, the low-risk field that is not necessarily a privacy field can be transmitted directly in plaintext or only lightly anonymized; this invention does not impose any limitations on this.

[0062] Optionally, the first weight is proportional to the number of characters in the corresponding field.

[0063] Optionally, the second weight is proportional to the percentage of times the corresponding field type appears in the client logs.

[0064] As can be seen, through the above optional embodiments, by inputting the log event text field into the privacy recognition model to obtain the predicted privacy level and determining the privacy field based on the text privacy level classification threshold, the accurate identification of privacy fields by multi-level thresholds and weighted averages is achieved, which improves the fineness and comprehensiveness of log privacy protection and reduces the risk of privacy field omission or over-identification caused by a single threshold.

[0065] Let's take a specific example to illustrate the solution in the above embodiments: Smart home device A, used for physiological monitoring, submits logs periodically. A single log event text is: "User ID: 10001234, Mobile number: 138****1234, ID card number: 3101********1234, Blood glucose level: 7.8mmol / L, Blood pressure: 128 / 82mmHg".

[0066] The system first splits the log event text into fields: user_id: 10001234; Phone number: 138****1234; id_card: 3101********1234; blood_sugar: 7.8; blood pressure: 128 / 82; A privacy identification model was generated based on the historical usage data of the smart home device, and each log field was scored: user_id: 0.12 (low) phone: 0.98 (high) id_card: 0.99 (high) blood_sugar: 0.67 (medium) blood pressure: 0.71 (medium) The overall privacy level of the log event text was calculated using a weighted average, resulting in a score of 0.89. The following conclusions were drawn: 0.89 > First threshold 0.95? : No, but > Second threshold 0.65, therefore phone and id_card are directly marked as privacy fields, and blood_sugar and blood_pressure are also marked as privacy fields because they are > Second threshold 0.65.

[0067] The final set of privacy fields includes: phone, id_card, blood_sugar, and blood_pressure.

[0068] The system uses AES-256-GCM (the key is derived from the private key of smart home device A) to encrypt the above fields. The encrypted log is as follows: "user_id: 10001234,phone: ENCRYPTED_xxxx,id_card: ENCRYPTED_xxxx,blood_sugar: ENCRYPTED_xxxx,blood_pressure: ENCRYPTED_xxxx".

[0069] The encrypted logs are securely uploaded to the cloud, ensuring that sensitive information is encrypted throughout the process, meeting the requirements of the Personal Information Protection Law, and supporting subsequent compliant decryption and analysis.

[0070] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of an automatic privacy information identification and encryption system based on user behavior characteristics disclosed in an embodiment of the present invention. Figure 2 The described automatic identification and encryption system for privacy information based on user behavior characteristics can be applied to data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). For example... Figure 2 As shown, this privacy information automatic identification and encryption system based on user behavior characteristics may include: The acquisition module 201 is used to acquire the client logs to be processed corresponding to the client.

[0071] The determination module 202 is used to determine the privacy identification model based on the client's historical behavior data. The identification module 203 is used to identify fields in the client logs according to the privacy identification model to obtain multiple privacy fields. The encryption module 204 is used to encrypt the privacy field according to the encryption parameters corresponding to the client to obtain the encrypted client log.

[0072] As can be seen, the above-described embodiments of the invention obtain client logs to be processed and determine a privacy identification model based on historical behavior data. By identifying privacy fields in the log fields and encrypting them based on encryption parameters, the invention can achieve accurate log privacy protection and encryption based on user behavior, improve the security and compliance of client log uploads, and reduce the risk of data leakage due to the failure to identify privacy fields.

[0073] As an optional implementation, the client log includes multiple log event texts.

[0074] As can be seen, the content of the client logs is limited through the above optional embodiments to assist in subsequent privacy identification and encryption, to assist in achieving accurate log privacy protection and encryption based on user behavior adaptation, to improve the security and compliance of client log uploads, and to reduce the risk of data leakage due to the failure to identify privacy fields.

[0075] As an optional embodiment, historical behavior data includes historical operation data of multiple users on the client; historical operation data includes multiple operation behaviors and corresponding operation text, operation location and operation time.

[0076] As can be seen, the content of historical behavior data is limited through the above optional embodiments to assist in subsequent privacy identification, to assist in the implementation of accurate log privacy protection and encryption based on user behavior adaptation, to improve the security and compliance of client log uploads, and to reduce the risk of data leakage due to the failure to identify privacy fields.

[0077] As an optional embodiment, the determining module determines the specific method of the privacy identification model based on the client's historical behavior data, including: For each operation behavior in each historical operation data, the operation location and operation time corresponding to the operation behavior are determined as the operation parameters of the operation behavior; The behavioral parameters are input into the trained privacy behavior recognition model to obtain the privacy probability that the behavioral parameters belong to the privacy processing behavior; optionally, the privacy behavior recognition model is a classifier model, which is trained on a training dataset that includes multiple training user operation parameters and corresponding privacy processing behavior annotations; The operation text corresponding to all operation behaviors with a privacy probability greater than a preset probability threshold is identified as privacy text; A privacy identification model is determined based on the user parameters and privacy text corresponding to each historical operation data.

[0078] As can be seen, through the above optional embodiments, by inputting the operation location and time of historical operation behavior as behavior parameters into the privacy behavior recognition model to calculate the privacy probability, selecting high-probability operation text as privacy text and combining it with user parameters to determine the privacy recognition model, accurate adaptive model selection based on behavior patterns and user characteristics is achieved, improving the targeting and accuracy of privacy field recognition, and reducing the risk of privacy misjudgment caused by the generality of the model.

[0079] As an optional embodiment, the determining module determines the specific method of the privacy identification model based on the user parameters and privacy text corresponding to each historical operation data, including: For each candidate privacy recognition model, the privacy text corresponding to all historical operation data is input into the candidate privacy recognition model to obtain the average value of the output privacy probability, and thus the prediction accuracy. The average parameter similarity between the user parameter annotations in the training data of the candidate privacy recognition model and the user parameters of the user corresponding to each historical operation data is calculated to obtain the user similarity. The priority of the candidate privacy recognition model is obtained by calculating the product of the prediction accuracy and the user similarity. The candidate privacy identification model with the highest priority is selected as the privacy identification model.

[0080] As can be seen, through the above optional embodiments, by calculating the product of the prediction accuracy of the candidate privacy recognition model on historical privacy text and the similarity of user parameters as the priority, the model with the highest priority is selected as the privacy recognition model, thereby achieving accurate model selection based on dual optimization of accuracy and user matching, improving the individual adaptability and reliability of privacy recognition, and reducing the risk of recognition bias caused by model mismatch.

[0081] As an optional embodiment, the privacy recognition model is a neural network model, trained on a training dataset that includes multiple training user texts and corresponding privacy annotations.

[0082] As can be seen, the model details of the privacy recognition model are defined through the above optional embodiments to assist in the subsequent identification and encryption of privacy fields, to assist in the accurate log privacy protection and encryption based on user behavior adaptation, to improve the security and compliance of client log uploads, and to reduce the risk of data leakage due to the failure to identify privacy fields.

[0083] As an optional embodiment, the identification module performs field identification on the client logs according to the privacy identification model to obtain multiple privacy fields in the following specific ways: Each field in the text of each log event in the client log is input into the privacy identification model to obtain the predicted privacy level for each field; For each log time text, calculate the weighted average of the predicted privacy levels of all fields in that log time text to obtain the text privacy level corresponding to that log event text; Log event texts with a privacy level greater than a preset first level threshold are identified as a privacy field; Fields in log event texts with a text privacy level greater than the preset second threshold but less than the first threshold, whose predicted privacy level is greater than the second threshold, are identified as privacy fields. All fields in log event text with a text privacy level below the second threshold will be excluded from privacy fields.

[0084] As can be seen, through the above optional embodiments, by inputting the log event text field into the privacy recognition model to obtain the predicted privacy level and determining the privacy field based on the text privacy level classification threshold, the accurate identification of privacy fields by multi-level thresholds and weighted averages is achieved, which improves the fineness and comprehensiveness of log privacy protection and reduces the risk of privacy field omission or over-identification caused by a single threshold.

[0085] As an optional embodiment, the calculation weight corresponding to each predicted level of privacy is the product of a first weight and a second weight; the first weight is proportional to the number of characters in the corresponding field; and the second weight is proportional to the proportion of the occurrence of the corresponding field type in the client logs.

[0086] As can be seen, through the above optional embodiments, the text privacy level is calculated by using a weighted average of the predicted privacy level calculated with the number of characters in the field and the proportion of their occurrence as weights. This achieves a dual-weight quantitative assessment based on field features, improves the objectivity and sensitivity of privacy field determination, and reduces the risk of privacy identification imbalance caused by improper weight allocation.

[0087] Example 3 Please see Figure 3 , Figure 3 This is another privacy information automatic identification and encryption system based on user behavior characteristics disclosed in the embodiments of the present invention. Figure 3 The described user behavior-based automatic identification and encryption system for privacy information is applied in data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). Figure 3 As shown, this privacy information automatic identification and encryption system based on user behavior characteristics may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the privacy information automatic identification and encryption method based on user behavior characteristics described in Embodiment 1.

[0088] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the method for automatic identification and encryption of privacy information based on user behavior characteristics described in Embodiment 1.

[0089] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the privacy information automatic identification and encryption method based on user behavior characteristics described in Embodiment 1.

[0090] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0092] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0093] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0095] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0096] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0097] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0098] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0099] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0100] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0101] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0102] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0103] Finally, it should be noted that the method and system for automatic identification and encryption of privacy information based on user behavior characteristics disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatic identification and encryption of privacy information based on user behavior characteristics, characterized in that, The method includes: Retrieve the client logs to be processed corresponding to the client; A privacy identification model is determined based on the client's historical behavior data; The client logs are analyzed using the privacy identification model to obtain multiple privacy fields. Based on the encryption parameters corresponding to the client, the privacy field is encrypted to obtain the encrypted client log. 2.The method of claim 1, wherein, The client logs include multiple log event texts. 3.The method of claim 1, wherein, The historical behavior data includes historical operation data of multiple users on the client; the historical operation data includes multiple operation behaviors and corresponding operation text, operation location and operation time. 4.The method of claim 3, wherein, The step of determining the privacy identification model based on the client's historical behavior data includes: For each operation behavior in each of the historical operation data, the operation location and operation time corresponding to the operation behavior are determined as the operation parameters of the operation behavior; The behavioral parameters are input into the trained privacy behavior recognition model to obtain the privacy probability that the behavioral parameters belong to privacy processing behaviors; the privacy behavior recognition model is a classifier model, which is trained on a training dataset that includes multiple training user operation parameters and corresponding privacy processing behavior annotations; The operation text corresponding to all operation behaviors whose privacy probability is greater than a preset probability threshold is determined as privacy text; A privacy identification model is determined based on the user parameters and privacy text corresponding to each of the historical operation data. 5.The method of claim 4, wherein, The step of determining a privacy identification model based on the user parameters of the user corresponding to each of the historical operation data and the privacy text includes: For each candidate privacy recognition model, the privacy text corresponding to all the historical operation data is input into the candidate privacy recognition model to obtain the average value of the output privacy probability, and thus the prediction accuracy. The average value of the parameter similarity between the user parameter annotations in the training data of the candidate privacy recognition model and the user parameters of the user corresponding to each historical operation data is calculated to obtain the user similarity. The priority of the candidate privacy recognition model is obtained by calculating the product of the prediction accuracy and the user similarity. The candidate privacy identification model with the highest priority is determined as the privacy identification model. 6.The method of claim 5, wherein, The privacy recognition model is a neural network model, which is trained using a training dataset that includes multiple training user texts and corresponding privacy annotations. 7.The method of claim 2, wherein, The process of identifying multiple privacy fields in the client logs based on the privacy identification model includes: Each field in each log event text of the client log is input into the privacy identification model to obtain the predicted privacy level corresponding to each field; For each log time text, calculate the weighted average of the predicted privacy level of all fields in the log time text to obtain the text privacy level corresponding to the log event text; The log event text whose text privacy level is greater than a preset first level threshold is identified as a privacy field; Fields in log event texts whose predicted privacy level is greater than the second threshold but less than the first threshold are identified as privacy fields. All fields of the log event text whose text privacy level is less than the second threshold are not considered privacy fields.

8. The method for automatic identification and encryption of privacy information based on user behavior characteristics according to claim 7, characterized in that, in, The calculation weight corresponding to each predicted level of privacy is the product of a first weight and a second weight; the first weight is proportional to the number of characters in the corresponding field; the second weight is proportional to the proportion of the occurrence of the field type of the corresponding field in the client log.

9. A system for automatic identification and encryption of privacy information based on user behavior characteristics, characterized by, The system includes: The acquisition module is used to acquire the client logs to be processed corresponding to the client. The determination module is used to determine a privacy identification model based on the client's historical behavior data; The identification module is used to identify fields in the client logs according to the privacy identification model to obtain multiple privacy fields; The encryption module is used to encrypt the privacy field according to the encryption parameters corresponding to the client to obtain the encrypted client log.

10. A system for automatic identification and encryption of privacy information based on user behavior characteristics, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the privacy information automatic identification and encryption method based on user behavior characteristics as described in any one of claims 1-8.

Images