Client information data desensitization method, device, medium and system

Abstract

The invention provides a client information data desensitization method, device, medium and system, and the method comprises the steps: processing a cooked corpus through employing a perceptron model, obtaining a corpus model, and obtaining a mapping relation between each preset binary sequence and an address in training data; under the condition that the type of the to-be-desensitized data in the client information data is the address, the corpus model is adopted to perform desensitization processing on all the addresses in the client information data to obtain the first desensitized client information data, so that compared with an existing scheme, different consideration of the complexity of the addresses is not needed, desensitization processing can be accurately performed on all the addresses, and the desensitization processing efficiency is improved. According to the method, the desensitization efficiency is improved, and the problem that in an existing scheme, due to different types of sensitive data of clients, part of the types of the sensitive data cannot be recognized through a fixed grammar rule for desensitization is solved.

Description

Technical Field

[0001] The present application relates to the technical field of data desensitization, and in particular to a method, device, medium and system for desensitizing customer information data. Background Art

[0002] Security control has always been an important part of governance, and data desensitization belongs to the category of security control. For Internet companies and traditional industries, data security has always been a very important and sensitive topic. Data desensitization refers to the deformation of certain sensitive information through desensitization rules to achieve reliable protection of sensitive privacy data. According to regulations, data involving customer security data or some commercial sensitive data, such as ID card number, mobile phone number, card number, customer number, address and other personal information, all need to be desensitized.

[0003] However, due to the different types of sensitive data of customers, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization in existing solutions. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device, medium and system for desensitizing customer information data, so as to at least solve the problem in existing solutions that due to the different types of customer sensitive data, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization.

[0005] In order to achieve the above object, according to one aspect of the present application, a method for desensitizing customer information data is provided, the method comprising:

[0006] Obtain customer information data;

[0007] The familiar corpus is processed by using a perceptron model to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and addresses in the training data, wherein the familiar corpus includes multiple address levels and all addresses in each of the address levels;

[0008] When the type of the data to be desensitized in the customer information data is an address, the corpus model is used to perform desensitization processing on all addresses in the customer information data to obtain first desensitized customer information data.

[0009] Optionally, all addresses in the customer information data are desensitized using the corpus model to obtain first desensitized customer information data, including:

[0010] In the case that the target address identical to the customer information data exists in the corpus model, all the target addresses in the customer information data are desensitized to obtain the first desensitized customer information data.

[0011] Optionally, before using the perceptron model to process the familiar corpus to obtain the corpus model, the method further includes:

[0012] Acquire raw corpus, where the raw corpus includes multiple addresses of the training data;

[0013] Performing hierarchical division processing on all addresses of the raw corpus to obtain a division result;

[0014] The addresses belonging to the same level in the division result are stored in the corresponding same position to obtain the familiar language material.

[0015] Optionally, after acquiring the customer information data, the method further includes:

[0016] When the type of data to be desensitized in the customer information data is non-address, a Trie tree is used to store at least the telephone number, email address, ID number and bank card number to obtain a final Trie tree.

[0017] Optionally, after obtaining the final Trie tree, the method further includes:

[0018] When the type of the data to be desensitized in the customer information data is a telephone number, and when the number of digits of the current sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current sequence is a telephone number, and performing desensitization processing on the current sequence;

[0019] When the type of the data to be desensitized in the customer information data is a bank card number, and when the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence;

[0020] When the type of data to be desensitized in the customer information data is the identification card number, and when the number of digits of the current number series in the customer information data is equal to the third preset number of digits, it is determined that the type of the current number series is the identification card number, and the current number series is desensitized.

[0021] Optionally, after obtaining the final Trie tree, the method further includes:

[0022] When the type of the data to be desensitized in the customer information data is an email address, and when a target preset symbol exists in a current sentence in the customer information data, it is determined that the type of the current sentence is an email address, and the current sentence is desensitized.

[0023] Optionally, the method further comprises:

[0024] When the number of bytes of the text paragraph of the customer information data is greater than or equal to the first preset number of bytes, the text paragraph is segmented according to the second preset number of bytes, or the text paragraph is segmented at a target position in the text paragraph using a preset symbol, and the target position is the position where the number of bytes of the text paragraph of the customer information data is equal to the first preset number of bytes.

[0025] According to another aspect of the present application, a desensitizing device for customer information data is provided, the device comprising:

[0026] A first acquisition unit, used to acquire customer information data;

[0027] A first processing unit is used to process the familiar corpus using a perceptron model to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and addresses in the training data, wherein the familiar corpus includes multiple address levels and all addresses in each of the address levels;

[0028] The second processing unit is used to, when the type of the data to be desensitized in the customer information data is an address, use the corpus model to desensitize all addresses in the customer information data to obtain first desensitized customer information data.

[0029] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the methods described.

[0030] According to another aspect of the present application, a customer information data desensitizing system is provided, the system comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the described methods.

[0031] By applying the technical solution of the present application, a corpus model is obtained by processing a familiar corpus using a perceptron model to obtain a mapping relationship between each preset binary number series and the address in the training data. When the type of data to be desensitized in the customer information data is an address, the corpus model is used to desensitize all addresses in the customer information data to obtain first desensitized customer information data. Compared with the existing scheme, there is no need to consider the complexity of the address, so as to accurately desensitize each address, thereby improving the desensitization efficiency, and further solving the problem that due to the different types of sensitive data of the customer, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization in the existing scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0033] Figure 1 A schematic diagram of a process for desensitizing customer information data provided according to an embodiment of the present application is shown;

[0034] Figure 2 A schematic diagram showing a process of another method for desensitizing customer information data provided according to an embodiment of the present application;

[0035] Figure 3 A structural block diagram of a customer information data desensitization device provided according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0040] Trie tree, also known as dictionary tree or prefix tree, is a data structure used to store and retrieve string data. The characteristic of Trie tree is that each node contains a character and a pointer to a child node. By moving along the path of the character sequence in the tree, string data can be efficiently retrieved, inserted, and deleted. Trie tree is often used in applications such as search engines, spell checking, string matching, etc.

[0041] The perceptron model is a binary classification algorithm that is used to divide input data into two categories. It updates the weights iteratively so that the classifier can correctly classify the input data.

[0042] As introduced in the background technology, security control has always been an important part of governance, and data desensitization belongs to the category of security control. For Internet companies and traditional industries, data security has always been a very important and sensitive topic. Data desensitization refers to the deformation of certain sensitive information through desensitization rules to achieve reliable protection of sensitive privacy data. According to regulations, data desensitization is required for customer security data or some commercial sensitive data, such as identity card number, mobile phone number, card number, customer number, address and other personal information. In order to solve the problem that the existing solutions have different types of sensitive data of customers, so that some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization, the embodiments of the present application provide a method, device, medium and system for desensitizing customer information data.

[0043] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0044] In this embodiment, a method for desensitizing customer information data running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] Figure 1 It is a flowchart of a method for desensitizing customer information data provided according to an embodiment of the present application.

[0046] like Figure 1 As shown, the method comprises the following steps:

[0047] Step S101, obtaining customer information data;

[0048] In one embodiment of the present application, after obtaining the customer information data in step S101, the method further includes:

[0049] When the type of data to be desensitized in the above customer information data is non-address, a Trie tree is used to store at least the telephone number, email address, ID number and bank card number to obtain a final Trie tree.

[0050] Specifically, the format of the Trie tree is:

[0051] “Username: Xiao A Home address: No. 202, Alley C, No. 111, B Street Bank card number: 11111111111111111111 Contact number: 12345678910 Email: AAA1@222.com”;

[0052] Among them, "Username: Xiao A Home address: No. 202, Alley C, No. 111, B Street, Bank card number" is the text segment to be parsed by the model, "111111111111111111111" is the segment matched by the bank card number rule, "Contact number", "12345678910" is the segment matched by the telephone number rule, and "email", "AAA1@222.com" is the segment matched by the email rule.

[0053] In one embodiment of the present application, after obtaining the final Trie tree, the method further includes:

[0054] In the case where the type of the data to be desensitized in the customer information data is a telephone number, and the number of digits of the current number sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current number sequence is a telephone number, and performing desensitization processing on the current number sequence;

[0055] In the case where the type of the data to be desensitized in the customer information data is a bank card number, and in the case where the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence;

[0056] When the type of data to be desensitized in the above-mentioned customer information data is the ID card number, and when the number of digits of the current number series in the above-mentioned customer information data is equal to the third preset number of digits, it is determined that the type of the above-mentioned current number series is the ID card number, and the above-mentioned current number series is desensitized.

[0057] Specifically, in the case where the type of data to be desensitized in the above-mentioned customer information data is a telephone number, the first preset number of digits can be 11 digits, that is, "12345678910" in the format of the above-mentioned Trie tree is a telephone number, and "12345678910" is desensitized; in the case where the type of data to be desensitized in the above-mentioned customer information data is a bank card number, the second preset number of digits is 20 digits, then "11111111111111111111" in the format of the above-mentioned Trie tree is a bank card number, and "111111111111111111111" is desensitized; in the case where the type of data to be desensitized in the above-mentioned customer information data is an ID card number, the third preset number of digits is 18 digits, then the type of the above-mentioned current number sequence is determined to be an ID card number, and the above-mentioned current number sequence is desensitized.

[0058] You can perform regular matching based on existing regular expressions.

[0059] In one embodiment of the present application, after obtaining the final Trie tree, the method further includes:

[0060] When the type of the data to be desensitized in the above customer information data is mailbox, and when there is a target preset symbol in the current sentence in the above customer information data, it is determined that the type of the above current sentence is mailbox, and the above current sentence is desensitized.

[0061] Specifically, when the type of data to be desensitized in the above customer information data is an email address, and the target preset symbol is "@", then "AAA1@222.com" in the above Trie tree format is determined to be an email address, and "AAA1@222.com" is desensitized.

[0062] Step S102, using a perceptron model to process the familiar corpus to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and the addresses in the training data, and the familiar corpus includes multiple address levels and all addresses in each of the address levels;

[0063] Specifically, the perceptron model algorithm is used to train the corpus model: clear the default named entity recognition tags of the NLP word segmentation; add preset address named entity tags, such as nd for the community name and nb for the unit building name; use the perceptron algorithm with the familiar corpus text file as input and the binary model file as output.

[0064] In one embodiment of the present application, before the perceptron model is used to process the familiar corpus in step S102 to obtain the corpus model, the method further includes:

[0065] Obtaining raw corpus, the raw corpus including multiple addresses of the training data;

[0066] Perform hierarchical division processing on all addresses in the raw corpus to obtain division results;

[0067] The addresses belonging to the same level in the above division results are stored in the corresponding same position to obtain the above familiar language material.

[0068] Specifically, raw corpus is the most original data for training the corpus model. It can be extracted from sources such as addresses in OCR-recognized certificates, takeout addresses, express delivery addresses, real estate agency addresses, etc. After removing the web page format and meaningless data (i.e. footer and header numbers, and data where numbers are incorrectly recognized as letters) from the extracted text, a general text model is used to perform preliminary part-of-speech tagging and the results are saved to a storage medium.

[0069] According to the part-of-speech information pre-annotated in the raw corpus, the text data is combined into complete address information. After excluding the province, city and county data, the remaining address data is annotated as the street named entity part-of-speech to generate a text file. In addition, the text file can be manually checked and inappropriate segmentation annotations can be modified to ensure the accuracy of the corpus.

[0070] Step S103: When the type of the data to be desensitized in the customer information data is an address, the corpus model is used to desensitize all addresses in the customer information data to obtain first desensitized customer information data.

[0071] In the above steps, the familiar corpus is processed by using the perceptron model to obtain a corpus model to obtain a mapping relationship between each preset binary number series and the address in the training data. In the case that the type of data to be desensitized in the above customer information data is an address, the above corpus model is used to desensitize all addresses in the above customer information data to obtain the first desensitized customer information data. Compared with the existing scheme, it is possible to accurately desensitize each address without considering the complexity of the address, thereby improving the desensitization efficiency, and further solving the problem that due to the different types of sensitive data of the customer, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization.

[0072] In one embodiment of the present application, all addresses in the above customer information data are desensitized using the above corpus model to obtain first desensitized customer information data, including:

[0073] In the case that the target address identical to the customer information data exists in the corpus model, all the target addresses in the customer information data are desensitized to obtain the first desensitized customer information data.

[0074] Specifically, when there is a target address identical to the above-mentioned customer information data in the above-mentioned corpus model, all the above-mentioned target addresses in the customer information data are desensitized, thereby ensuring the accuracy of desensitization. Because the desensitization is performed based on the mapping relationship of binary data, the processing speed is also improved compared with the existing scheme.

[0075] In one embodiment of the present application, the above method further includes:

[0076] When the number of bytes of the text paragraph of the above-mentioned customer information data is greater than or equal to the first preset number of bytes, the above-mentioned text paragraph is segmented according to the second preset number of bytes, or the above-mentioned text paragraph is segmented at the target position in the above-mentioned text paragraph using a preset symbol, and the above-mentioned target position is the position where the number of bytes of the above-mentioned text paragraph of the above-mentioned customer information data is equal to the first preset number of bytes.

[0077] Specifically, the first preset byte number can be 4000 bytes. If the byte number of the text paragraph of the above customer information data is greater than or equal to the first preset byte number, the above text paragraph is segmented according to the second preset byte number (i.e. 2000), or segmented according to the "." symbol in the natural sentence to improve the processing speed.

[0078] In addition, according to the paragraph symbol and the symbol indicating segmentation in the web page format, ”.

[0079] Beneficial effects of the method of this application:

[0080] Use natural language processing technology to generate text models through training corpus, and train the model to learn the ability to distinguish fine-grained addresses. Fine-grained addresses are divided into multiple levels, and the model can distinguish address information at a specified level. The desensitization device can customize the address information at a specified level, thereby improving the accuracy of the address desensitization results, and also meeting the needs of customizing desensitization rules at the fine-grained address level.

[0081] The desensitization rules and recognition methods are set separately. When the desensitization rules change, you only need to modify the desensitization rules without changing the recognition method. Collect the latest address data corpus and retrain the generated text model to realize the address street data recognition of the new rules.

[0082] Format the text to be desensitized. The sending device retains the format information of the text to be desensitized, and can obtain paragraph information when the receiving device processes the text to be desensitized, so that the receiving device processes a smaller text block at a time instead of the original larger text block, thereby reducing the resource usage of the receiving device.

[0083] Apply multi-mode text matching algorithm to desensitization processing. Complete desensitization processing through a single traversal, improve the speed of desensitization processing of the receiving device, and avoid blocking other devices caused by the desensitization link.

[0084] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the method for desensitizing the customer information data of the present application will be described in detail below in conjunction with specific embodiments.

[0085] This embodiment relates to a specific method for desensitizing customer information data, such as Figure 2 As shown, the following steps are included:

[0086] Obtain customer information data; perform hierarchical division processing on all addresses of the raw corpus to obtain division results; store addresses belonging to the same level in the division results to the corresponding same position to obtain the familiar corpus;

[0087] The perceptron model is used to process the familiar corpus to obtain a corpus model, which represents the mapping relationship between each preset binary number series and the addresses in the training data. The familiar corpus includes multiple address levels and all addresses in each address level.

[0088] When the type of the data to be desensitized in the customer information data is an address, desensitizing all addresses in the customer information data using a corpus model to obtain first desensitized customer information data;

[0089] Use the Trie tree to store at least the phone number, email address, ID number and bank card number to obtain the final Trie tree;

[0090] When the type of the data to be desensitized in the customer information data is a telephone number, and when the number of digits of the current sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current sequence is a telephone number, and performing desensitization processing on the current sequence to obtain second desensitized customer information data;

[0091] When the type of the data to be desensitized in the customer information data is a bank card number, and when the number of digits of the current sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current sequence is a bank card number, and performing desensitization processing on the current sequence to obtain third desensitized customer information data;

[0092] When the type of the data to be desensitized in the customer information data is an ID card number, and when the number of digits of the current sequence in the customer information data is equal to the third preset number of digits, determining that the type of the current sequence is an ID card number, and performing desensitization processing on the current sequence to obtain fourth desensitized customer information data;

[0093] When the type of data to be desensitized in the customer information data is mailbox, and when there is a target preset symbol in the current sentence in the customer information data, the type of the current sentence is determined to be mailbox, and the current sentence is desensitized to obtain the fifth desensitized customer information data.

[0094] By using a perceptron model to process familiar corpus, a corpus model is obtained to obtain a mapping relationship between each preset binary number series and the address in the training data. When the type of data to be desensitized in the above customer information data is an address, the above corpus model is used to desensitize all addresses in the above customer information data to obtain the first desensitized customer information data. Compared with the existing scheme, there is no need to consider the complexity of the address, so as to accurately desensitize each address, thereby improving the desensitization efficiency, and further solving the problem that due to the different types of sensitive data of the customer, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization in the existing scheme.

[0095] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0096] The embodiment of the present application also provides a desensitizing device for customer information data. It should be noted that the desensitizing device for customer information data of the embodiment of the present application can be used to perform the desensitizing method for customer information data provided by the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and those that have been described will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0097] The following is an introduction to the desensitizing device for customer information data provided in an embodiment of the present application.

[0098] Figure 3 1 is a structural block diagram of a desensitizing device for customer information data provided according to an embodiment of the present application. Figure 3 As shown, the device comprises:

[0099] A first acquisition unit 31 is used to acquire customer information data;

[0100] A first processing unit 32 is used to process the familiar corpus using a perceptron model to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and addresses in the training data, and the familiar corpus includes multiple address levels and all addresses in each of the address levels;

[0101] The second processing unit 33 is used for, when the type of the data to be desensitized in the customer information data is an address, using the corpus model to desensitize all addresses in the customer information data to obtain first desensitized customer information data.

[0102] In the above-mentioned device, a corpus model is obtained by processing a familiar corpus using a perceptron model to obtain a mapping relationship between each preset binary number series and the address in the training data. When the type of data to be desensitized in the above-mentioned customer information data is an address, the above-mentioned corpus model is used to desensitize all addresses in the above-mentioned customer information data to obtain the first desensitized customer information data. Compared with the existing scheme, it is possible to accurately desensitize each address without considering the complexity of the address, thereby improving the desensitization efficiency, and thus solving the problem that due to the different types of sensitive data of the customer, part of the type of sensitive data cannot be identified by fixed grammatical rules for desensitization in the existing scheme.

[0103] In one embodiment of the present application, the second processing unit includes a first processing module,

[0104] The first processing module is used for desensitizing all the target addresses in the customer information data when there is a target address identical to the customer information data in the corpus model to obtain the first desensitized customer information data.

[0105] In one embodiment of the present application, the apparatus further includes a second acquisition unit, a third processing unit and a fourth processing unit. Before the perceptron model is used to process the familiar corpus to obtain the corpus model,

[0106] The second acquisition unit is used to acquire raw corpus, where the raw corpus includes multiple addresses of the training data;

[0107] The third processing unit is used to perform hierarchical division processing on all addresses in the raw corpus to obtain a division result;

[0108] The fourth processing unit is used to store the addresses belonging to the same level in the above division results into the corresponding same position to obtain the above familiar language material.

[0109] In one embodiment of the present application, the above-mentioned device also includes a fifth processing unit. After obtaining the customer information data, the fifth processing unit is used to use a Trie tree to store at least a telephone number, an email address, an ID number and a bank card number when the type of data to be desensitized in the above-mentioned customer information data is non-address, so as to obtain a final Trie tree.

[0110] In one embodiment of the present application, the apparatus further includes a sixth processing unit, a seventh processing unit and an eighth processing unit. After obtaining the final Trie tree,

[0111] The sixth processing unit is used for, when the type of the data to be desensitized in the customer information data is a telephone number, and when the number of digits of the current sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current sequence is a telephone number, and performing desensitization processing on the current sequence;

[0112] The seventh processing unit is used for, when the type of the data to be desensitized in the customer information data is a bank card number, and when the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence;

[0113] The eighth processing unit is used to determine that the type of the current number sequence is the ID card number and perform desensitization processing on the current number sequence when the type of data to be desensitized in the above-mentioned customer information data is the ID card number and the number of digits of the current number sequence in the above-mentioned customer information data is equal to the third preset number of digits.

[0114] In one embodiment of the present application, the above-mentioned device also includes a ninth processing unit. After obtaining the final Trie tree, the ninth processing unit is used to, when the type of the data to be desensitized in the above-mentioned customer information data is a mailbox and when there is a target preset symbol in the current sentence in the above-mentioned customer information data, determine that the type of the above-mentioned current sentence is a mailbox and perform desensitization processing on the above-mentioned current sentence.

[0115] In one embodiment of the present application, the above-mentioned device also includes a tenth processing unit, which is used to split the above-mentioned text paragraph with a second preset byte number when the number of bytes of the text paragraph of the above-mentioned customer information data is greater than or equal to the first preset byte number, or to split the above-mentioned text paragraph at a target position in the above-mentioned text paragraph using a preset symbol, and the above-mentioned target position is the position where the number of bytes of the above-mentioned text paragraph of the above-mentioned customer information data is equal to the first preset byte number.

[0116] The above-mentioned desensitizing device for customer information data includes a processor and a memory. The above-mentioned first acquisition unit, the first processing unit, the second processing unit, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0117] The processor contains a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the kernel parameters can be adjusted to solve the problem that the existing solution cannot identify some of the sensitive data types through fixed grammatical rules for desensitization due to the different types of sensitive data of customers.

[0118] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0119] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the desensitizing method for the customer information data.

[0120] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for desensitizing the customer information data is executed when the program is running.

[0121] An embodiment of the present invention provides a device, the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and the processor implements at least the following steps when executing the program: obtaining customer information data; using a perceptron model to process a familiar corpus to obtain a corpus model, wherein the corpus model represents the mapping relationship between each preset binary series and the address in the training data, and the familiar corpus includes multiple address levels and all addresses in each of the above address levels; when the type of data to be desensitized in the above customer information data is an address, the corpus model is used to desensitize all addresses in the above customer information data to obtain first desensitized customer information data. The device in this article can be a server, PC, PAD, mobile phone, etc.

[0122] Optionally, all addresses in the customer information data are desensitized using the corpus model to obtain first desensitized customer information data, including:

[0123] In the case that the target address identical to the customer information data exists in the corpus model, all the target addresses in the customer information data are desensitized to obtain the first desensitized customer information data.

[0124] Optionally, before using the perceptron model to process the familiar corpus to obtain the corpus model, the method further includes:

[0125] Obtaining raw corpus, the raw corpus including multiple addresses of the training data;

[0126] Perform hierarchical division processing on all addresses in the raw corpus to obtain division results;

[0127] The addresses belonging to the same level in the above division results are stored in the corresponding same position to obtain the above familiar language material.

[0128] Optionally, after obtaining the customer information data, the method further includes:

[0129] When the type of data to be desensitized in the above customer information data is non-address, a Trie tree is used to store at least the telephone number, email address, ID number and bank card number to obtain a final Trie tree.

[0130] Optionally, after obtaining the final Trie tree, the method further includes:

[0131] In the case where the type of the data to be desensitized in the customer information data is a telephone number, and the number of digits of the current number sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current number sequence is a telephone number, and performing desensitization processing on the current number sequence;

[0132] In the case where the type of the data to be desensitized in the customer information data is a bank card number, and in the case where the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence;

[0133] When the type of data to be desensitized in the above-mentioned customer information data is the ID card number, and when the number of digits of the current number series in the above-mentioned customer information data is equal to the third preset number of digits, it is determined that the type of the above-mentioned current number series is the ID card number, and the above-mentioned current number series is desensitized.

[0134] Optionally, after obtaining the final Trie tree, the method further includes:

[0135] When the type of the data to be desensitized in the above customer information data is mailbox, and when there is a target preset symbol in the current sentence in the above customer information data, it is determined that the type of the above current sentence is mailbox, and the above current sentence is desensitized.

[0136] Optionally, the above method further includes:

[0137] When the number of bytes of the text paragraph of the above-mentioned customer information data is greater than or equal to the first preset number of bytes, the above-mentioned text paragraph is segmented according to the second preset number of bytes, or the above-mentioned text paragraph is segmented at the target position in the above-mentioned text paragraph using a preset symbol, and the above-mentioned target position is the position where the number of bytes of the above-mentioned text paragraph of the above-mentioned customer information data is equal to the first preset number of bytes.

[0138] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with at least the following method steps: obtaining customer information data; using a perceptron model to process a familiar corpus to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and an address in the training data, wherein the familiar corpus includes multiple address levels and all addresses within each of the above address levels; when the type of data to be desensitized in the above customer information data is an address, using the above corpus model to desensitize all addresses in the above customer information data to obtain first desensitized customer information data.

[0139] Optionally, all addresses in the customer information data are desensitized using the corpus model to obtain first desensitized customer information data, including:

[0140] In the case that the target address identical to the customer information data exists in the corpus model, all the target addresses in the customer information data are desensitized to obtain the first desensitized customer information data.

[0141] Optionally, before using the perceptron model to process the familiar corpus to obtain the corpus model, the method further includes:

[0142] Obtaining raw corpus, the raw corpus including multiple addresses of the training data;

[0143] Perform hierarchical division processing on all addresses in the raw corpus to obtain division results;

[0144] The addresses belonging to the same level in the above division results are stored in the corresponding same position to obtain the above familiar language material.

[0145] Optionally, after obtaining the customer information data, the method further includes:

[0146] When the type of data to be desensitized in the above customer information data is non-address, a Trie tree is used to store at least the telephone number, email address, ID number and bank card number to obtain a final Trie tree.

[0147] Optionally, after obtaining the final Trie tree, the method further includes:

[0148] In the case where the type of the data to be desensitized in the customer information data is a telephone number, and the number of digits of the current number sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current number sequence is a telephone number, and performing desensitization processing on the current number sequence;

[0149] In the case where the type of the data to be desensitized in the customer information data is a bank card number, and in the case where the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence;

[0150] When the type of data to be desensitized in the above-mentioned customer information data is the ID card number, and when the number of digits of the current number series in the above-mentioned customer information data is equal to the third preset number of digits, it is determined that the type of the above-mentioned current number series is the ID card number, and the above-mentioned current number series is desensitized.

[0151] Optionally, after obtaining the final Trie tree, the method further includes:

[0152] When the type of the data to be desensitized in the above customer information data is mailbox, and when there is a target preset symbol in the current sentence in the above customer information data, it is determined that the type of the above current sentence is mailbox, and the above current sentence is desensitized.

[0153] Optionally, the above method further includes:

[0154] When the number of bytes of the text paragraph of the above-mentioned customer information data is greater than or equal to the first preset number of bytes, the above-mentioned text paragraph is segmented according to the second preset number of bytes, or the above-mentioned text paragraph is segmented at the target position in the above-mentioned text paragraph using a preset symbol, and the above-mentioned target position is the position where the number of bytes of the above-mentioned text paragraph of the above-mentioned customer information data is equal to the first preset number of bytes.

[0155] The present application also provides a desensitization system for customer information data, the system comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for executing any of the above methods. The familiar corpus is processed by using a perceptron model to obtain a corpus model to obtain a mapping relationship between each preset binary series and the address in the training data. When the type of data to be desensitized in the customer information data is an address, the corpus model is used to desensitize all addresses in the customer information data to obtain the first desensitized customer information data, thereby eliminating the need to consider the complexity of the address in comparison with the existing solution, so as to accurately desensitize each address, thereby improving the desensitization efficiency, and thus solving the problem that due to the different types of sensitive data of the customer, the existing solution makes it impossible to identify part of the type of sensitive data through fixed grammatical rules for desensitization.

[0156] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0157] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0158] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0159] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0160] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0161] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0162] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0163] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules 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 technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0164] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0165] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0166] 1) The method for desensitizing customer information data of the present application processes a familiar corpus by using a perceptron model to obtain a corpus model to obtain a mapping relationship between each preset binary number series and the address in the training data. When the type of data to be desensitized in the above customer information data is an address, the above corpus model is used to desensitize all addresses in the above customer information data to obtain the first desensitized customer information data. Compared with the existing scheme, there is no need to consider the complexity of the address, so as to accurately desensitize each address, thereby improving the desensitization efficiency, and thus solving the problem that due to the different types of sensitive data of the customer, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization in the existing scheme.

[0167] 2) The desensitizing device for customer information data of the present application processes familiar corpus by using a perceptron model to obtain a corpus model to obtain a mapping relationship between each preset binary number series and the address in the training data. When the type of data to be desensitized in the above customer information data is an address, the above corpus model is used to desensitize all addresses in the above customer information data to obtain the first desensitized customer information data. Compared with the existing scheme, it is possible to accurately desensitize each address without considering the complexity of the address, thereby improving the desensitization efficiency, and further solving the problem that due to the different types of sensitive data of the customer in the existing scheme, some of the types of sensitive data cannot be identified by fixed grammatical rules for desensitization.

[0168] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for desensitizing customer information data, characterized in that: include: Obtain customer information data; The familiar corpus is processed by using a perceptron model to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and addresses in the training data, wherein the familiar corpus includes multiple address levels and all addresses in each of the address levels; When the type of the data to be desensitized in the customer information data is an address, the corpus model is used to perform desensitization processing on all addresses in the customer information data to obtain first desensitized customer information data.

2. The method according to claim 1, characterized in that The corpus model is used to perform desensitization processing on all addresses in the customer information data to obtain first desensitized customer information data, including: In the case that the target address identical to the customer information data exists in the corpus model, all the target addresses in the customer information data are desensitized to obtain the first desensitized customer information data.

3. The method according to claim 1, characterized in that Before using the perceptron model to process the familiar corpus to obtain the corpus model, the method further includes: Acquire raw corpus, where the raw corpus includes multiple addresses of the training data; Performing hierarchical division processing on all addresses of the raw corpus to obtain a division result; The addresses belonging to the same level in the division result are stored in the corresponding same position to obtain the familiar language material.

4. The method according to claim 1, characterized in that: After acquiring the customer information data, the method further includes: When the type of data to be desensitized in the customer information data is non-address, a Trie tree is used to store at least the telephone number, email address, ID number and bank card number to obtain a final Trie tree.

5. The method according to claim 1, characterized in that After obtaining the final Trie tree, the method further includes: When the type of the data to be desensitized in the customer information data is a telephone number, and when the number of digits of the current sequence in the customer information data is equal to the first preset number of digits, determining that the type of the current sequence is a telephone number, and performing desensitization processing on the current sequence; When the type of the data to be desensitized in the customer information data is a bank card number, and when the number of digits of the current number sequence in the customer information data is equal to the second preset number of digits, determining that the type of the current number sequence is a bank card number, and performing desensitization processing on the current number sequence; When the type of data to be desensitized in the customer information data is the identification card number, and when the number of digits of the current number series in the customer information data is equal to the third preset number of digits, it is determined that the type of the current number series is the identification card number, and the current number series is desensitized.

6. The method according to claim 1, characterized in that After obtaining the final Trie tree, the method further includes: When the type of the data to be desensitized in the customer information data is an email address, and when a target preset symbol exists in a current sentence in the customer information data, it is determined that the type of the current sentence is an email address, and the current sentence is desensitized.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: When the number of bytes of the text paragraph of the customer information data is greater than or equal to the first preset number of bytes, the text paragraph is segmented according to the second preset number of bytes, or the text paragraph is segmented at a target position in the text paragraph using a preset symbol, and the target position is the position where the number of bytes of the text paragraph of the customer information data is equal to the first preset number of bytes.

8. A desensitizing device for customer information data, characterized in that: include: A first acquisition unit, used to acquire customer information data; A first processing unit is used to process the familiar corpus using a perceptron model to obtain a corpus model, wherein the corpus model represents a mapping relationship between each preset binary number series and addresses in the training data, wherein the familiar corpus includes multiple address levels and all addresses in each of the address levels; The second processing unit is used to, when the type of the data to be desensitized in the customer information data is an address, use the corpus model to desensitize all addresses in the customer information data to obtain first desensitized customer information data.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. A desensitizing system for customer information data, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of claims 1 to 7.

Images