CN113868297A - Sensitive data analysis method and device, terminal equipment and storage medium - Google Patents

Abstract

The embodiments of the present application are applicable to the technical field of artificial intelligence, and provide a sensitive data analysis method, device, terminal device and storage medium. The method includes: copying the data stream returned by the server to obtain the copied data stream; Any current data in , according to the processing order M in which the current data is sampled, to

Update the determined N sampled data with a probability of , to obtain the updated N sampled data; perform sensitive word analysis on the updated N sampled data to obtain the sensitive word result. By using the above method, when the terminal device performs down-sampling on data of unknown quantity, it can provide the same sampled probability value to perform down-sampling processing on it.

Description

Sensitive data analysis method, device, terminal device and storage medium

technical field

The present application belongs to the technical field of artificial intelligence, and in particular, relates to a sensitive data analysis method, device, terminal device and storage medium.

Background technique

With the advent of the data era, the huge value contained in the data has been mined, but the protection of the sensitive data in the data has also become difficult. Among them, common sensitive data include name, ID number, address, phone number, bank account number and other data, which are all personal private information.

At present, for the data stream returned from the server to the client, because the data stream usually contains a large amount of data, it is usually necessary to downsample the data through an existing sampling method to reduce the amount of data to be processed. However, because the quantity of data contained in each data stream is unknown and the quantity contained each time is also different, using the existing downsampling method, when downsampling the data in the data stream, it is necessary to count the data in advance The total amount of data contained in the stream. Then, calculate the inverse of the total number to get the probability value of each data being sampled. Therefore, the existing down-sampling methods have certain limitations, and cannot provide the same sampled probability value to down-sample data with an unknown quantity.

SUMMARY OF THE INVENTION

The embodiments of the present application provide a sensitive data analysis method, device, terminal device, and storage medium, which can solve the limitation of the existing downsampling method, which cannot provide the same sampled probability value for an unknown amount of data. Downsampling issues.

In a first aspect, an embodiment of the present application provides a sensitive data analysis method, the method comprising:

Copy the data stream returned by the server to obtain the copied data stream;

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据；For any current data in the copy data stream, according to the processing order M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data;

Sensitive word analysis is performed on the updated N sampled data to obtain sensitive word results.

In one embodiment, the data stream returned by the server is copied to obtain the copied data stream, including:

For any data in the data stream, identify the data structure of the data;

Copy the data belonging to the target data structure in the data to obtain the copy data corresponding to the data.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据之前，还包括：In one embodiment, in the processing order M in which the current data is sampled, to

The probability of updating the determined N sampled data, and before obtaining the updated N sampled data, also includes:

If M≤N, it is determined that the current data is the sampled data, until the determined N sampled data are obtained.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据，包括：In one embodiment, according to the processing order M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data, including:

的概率确定当前数据不为采样数据，则保持已确定的N个采样数据不变；if

The probability of determining that the current data is not sampled data, then keep the determined N sampled data unchanged;

的概率确定当前数据为采样数据，则以

的概率从已确定的N个采样数据中，确定一个需被当前数据更新的替换采样数据；if

The probability of determining that the current data is sampled data, then use

The probability of determining a replacement sampled data that needs to be updated by the current data from the N sampled data that have been determined;

Update the current data to replace the sampled data to obtain the updated N sampled data.

In one embodiment, sensitive word analysis processing is performed on the updated N sampled data to obtain sensitive word results, including:

For any sampled data, identify the sampled data and generate text information;

Segment the text information to obtain multiple text segmentations of the text information;

Determine the word segmentation vectors of the multiple text segmentations according to the respective position information of the multiple text segmentations in the preset word vector library;

According to the word segmentation vector, multiple text segmentations are identified, and the sensitive word results of the sampled data are obtained.

In one embodiment, after performing sensitive word analysis processing on the updated N sampled data to obtain a sensitive word result, the method further includes:

According to the results of sensitive words, count the number of sampled data containing sensitive words;

If the ratio between the number of sampled data containing sensitive words and N is greater than the preset value, the sensitive word analysis processing is performed on the data stream returned by the server.

In one embodiment, if the ratio between the number of sampled data containing sensitive words and N is greater than a preset value, then performing sensitive word analysis processing on all data streams returned by the server, including:

Perform desensitization processing on the sensitive words contained in the data stream returned by the server to obtain desensitized data;

Send desensitized data to the client.

In a second aspect, an embodiment of the present application provides a sensitive data analysis device, the device comprising:

The copy module is used to copy the data stream returned by the server to obtain the copied data stream;

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据；The sampling module is used for any current data in the copy data stream, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data;

The analysis module is used to analyze and process sensitive words on the updated N sampled data to obtain the results of sensitive words.

In a third aspect, an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, any one of the above-mentioned first aspects is implemented. a method.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, wherein the computer program is executed by a processor to implement any one of the above-mentioned first aspects. method.

In a fifth aspect, an embodiment of the present application provides a computer program product, which, when the computer program product runs on a terminal device, enables the terminal device to execute the method of any one of the above-mentioned first aspects.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据。以此，可以使终端设备在接收到当前数据时即进行采样处理，而不是在接收数据流中所有的数据后在进行采样处理。并且，可避免终端设备使用现有降采样方法的局限性，使终端设备能够提供相同的被采样的概率值对数量未知的数据进行降采样处理。进而，可以使终端设备在对采样效果更好的采样数据进行敏感词分析处理时，其得到的敏感词结果更接近于整体的数据流中各个数据的敏感词结果。Compared with the prior art, the embodiment of the present application has the beneficial effect of: first, after copying the data stream returned by the server to the client, the copied data stream is processed instead of directly processing the returned data stream, which can be stored in the On the basis of analyzing and processing the sensitive words on the data stream, it can also ensure that the data returned by the server to the client will not cause delay problems. After that, based on the processing order M in which the current data in the copied data stream is sampled,

Update the determined N sampled data with the probability of , and obtain the updated N sampled data. In this way, the terminal device can perform sampling processing when receiving the current data, instead of performing sampling processing after receiving all the data in the data stream. Moreover, the limitation of using the existing down-sampling method by the terminal device can be avoided, so that the terminal device can provide the same sampled probability value to perform down-sampling processing on an unknown amount of data. Furthermore, when the terminal device performs sensitive word analysis processing on the sampled data with better sampling effect, the sensitive word result obtained by the terminal device is closer to the sensitive word result of each data in the overall data stream.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only for the present application. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

FIG. 1 is an implementation flowchart of a sensitive data analysis method provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of an implementation manner of S102 of a sensitive data analysis method provided by an embodiment of the present application;

FIG. 3 is a schematic diagram of an implementation manner of S103 of a sensitive data analysis method provided by an embodiment of the present application;

FIG. 4 is a schematic diagram of an implementation manner of S101 of a sensitive data analysis method provided by an embodiment of the present application;

FIG. 5 is an implementation flowchart of a sensitive data analysis method provided by another embodiment of the present application;

FIG. 6 is a schematic diagram of an implementation manner of S132 of a sensitive data analysis method provided by an embodiment of the present application;

FIG. 7 is a schematic structural diagram of a sensitive data analysis device provided by an embodiment of the present application;

FIG. 8 is a schematic structural diagram of a terminal device according to an embodiment of the present application.

Detailed ways

In the following description, for the purpose of illustration rather than limitation, specific details such as a specific system structure and technology are set forth in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

It is to be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described feature, integer, step, operation, element and/or component, but does not exclude one or more other The presence or addition of features, integers, steps, operations, elements, components and/or sets thereof.

In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description, and should not be construed as indicating or implying relative importance.

The sensitive data analysis method provided by the embodiments of the present application can be applied to terminal devices such as tablet computers, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, and the like. No restrictions apply.

Please refer to FIG. 1. FIG. 1 shows an implementation flowchart of a sensitive data analysis method provided by an embodiment of the present application. The method includes the following steps:

S101. The terminal device copies the data stream returned by the server to obtain the copied data stream.

In an embodiment, the terminal device may copy the data in the data stream by using the network traffic bypass technology to obtain the copied data stream. Specifically, the above-mentioned network traffic bypass technology can monitor and copy the data flow returned by the server in real time through the "port mirroring" function in the network device of the switch, without causing delay to the data flow returned by the server. That is, the data stream originally returned by the server can be directly returned to the client.

Among them, it should be noted that the above-mentioned "port mirroring" function can be implemented through the switch network device, or a software application program that can realize the "port mirroring" function can be set in the client that receives the data stream in advance, so as to return to the server. Copy to the data stream in the client. It can be understood that, after the client side copies the data stream through the software application program of the "port mirroring" function, it can transmit the copied data stream to the terminal device for subsequent processing, which is not limited.

In this embodiment, a software application program capable of implementing a "port mirroring" function is set in the client that receives the data stream to copy the data stream. In this way, the hardware devices to be installed in the server can be reduced, and the hardware cost of the server can be reduced.

In one embodiment, the above-mentioned server can simultaneously execute data requests of multiple clients. Therefore, the data stream returned by the above-mentioned server includes but is not limited to the data stream received by a certain client, and also includes the server received by multiple clients respectively. The returned data stream is not limited.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据。S102, for any current data in the copy data stream, the terminal device according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data.

In an embodiment, the above-mentioned current data is data processed at the current moment of the terminal device. Wherein, a data stream usually includes a large amount of data, and similarly, a corresponding copy data stream should also have a large amount of data. Based on this, if the terminal device performs sensitive word analysis processing on all the data in the copied data stream, it will consume a lot of processing time. Therefore, the terminal device can perform random down-sampling on a plurality of data to obtain sampled data, so as to reduce the amount of data that needs to be analyzed and processed.

However, when randomly down-sampling multiple data in the copied data stream, it is usually necessary to ensure that the probability of each data being sampled is the same. Generally, the terminal device needs to count the total quantity of data contained in the data stream in advance, and then calculate the inverse of the total quantity to obtain the sampling probability. Then, it is determined whether each data is sampled one by one according to the sampling probability. However, this method requires the terminal device to count the total amount of all data in the data stream in advance. It should be noted that, because the amount of data contained in each data stream is not equal, when the terminal device receives multiple data in the copied data stream, it can count the data contained in the copied data stream only after it has completely received the copied data stream. the total amount of data.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据，以使终端设备可在不知数据的总数量时，即可为每个数据提供相同的采样概率。Therefore, in order to avoid the limitation that the terminal device can provide the same sampling probability for each data only after fully receiving the copied data stream. When receiving any current data in the data stream, the terminal device may determine the processing sequence M for receiving the current data stream. After that, with

Update the determined N sampled data with the probability of , to obtain the updated N sampled data, so that the terminal device can provide the same sampling probability for each data without knowing the total number of data.

Specifically, the terminal device may preset the number N of sampled data to be sampled from each copied data stream. Afterwards, if it is determined that M≤N, it is determined that the current data is the sampled data, and one sampled data among the N sampled data is obtained. Then, according to the processing sequence M of any current data, the determination process of M≤N is repeatedly performed until N sample data are obtained.

It can be understood that the above N can be preset for the number of sampling data for the staff, which can be set according to the actual situation, which is not limited. In this embodiment, for multiple pieces of data in the copy data stream, the terminal device may first regard the data whose processing order is in the first N sequence numbers as sampling data according to the processing order of sampling and processing the multiple pieces of data. Therefore, it can be understood that if the number of data in the copied data stream is less than N, the probability that any data in the copied data stream is determined as sampled data is 100%.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据，具体的，其可包括如下子步骤S1021-S1023，详述如下：In another embodiment, referring to FIG. 2 , when the number of data in the copied data stream exceeds N, for any current data, the terminal device may, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data. Specifically, it may include the following sub-steps S1021-S1023, which are described in detail as follows:

的概率确定当前数据是否为采样数据。例如，S1021、若终端设备以

的概率确定当前数据不为采样数据，则保持已确定的N个采样数据不变。S1022、若终端设备以

的概率确定当前数据为采样数据，则以

概率从已确定的N个采样数据中，确定一个需被当前数据更新的替换采样数据。而后，S1023、终端设备可以将当前数据更新替换采样数据，得到更新后的N个采样数据。If the processing sequence of the current data is M>N, the terminal device will

The probability of determining whether the current data is sampled data. For example, in S1021, if the terminal device starts with

The probability of determining that the current data is not sampled data, then keep the determined N sampled data unchanged. S1022. If the terminal device is

The probability of determining that the current data is sampled data, then use

The probability determines a replacement sample data that needs to be updated by the current data from among the determined N sample data. Then, in S1023, the terminal device may update the current data to replace the sampling data to obtain the updated N sampling data.

的概率确定当前数据是否被采样。若确定当前数据被采样，则需要从之前确定的前N个采样数据中随机确定一个需被当前数据更新的替换采样数据。可以理解的是，从M个数据中确定替换采样数据的概率为：

其中，

表示为从M个数据中选中已确定的N个采样数据的概率；

为从已确定的N个采样数据中随机确定的替换采样数据的概率。Specifically, if the number of data in the copied data stream exceeds N, start from the Mth (M≥N+1) data, and then proceed to

The probability of determining whether the current data is sampled. If it is determined that the current data is sampled, a replacement sampled data to be updated by the current data needs to be randomly determined from the previously determined first N sampled data. It can be understood that the probability of determining the replacement sample data from the M data is:

in,

It is expressed as the probability of selecting the determined N sampled data from the M data;

It is the probability of replacing sample data randomly determined from the determined N sample data.

Based on the above description, it can be deduced that when the current data changes, that is, when M changes, the sampling probability of the current data is equal to the sampling probability of any one of the N sampled data previously determined. .

What needs to be added is that when sampling and processing each current data, it can be processed when the current data is received, so as to avoid that the terminal device can only provide the same sampling for each data after fully receiving the copied data stream. Limitations of probability.

S103, the terminal device performs sensitive word analysis processing on the updated N sampled data to obtain a sensitive word result.

In an embodiment, the terminal device may perform sensitive word analysis processing on the sampled data based on a pre-trained sensitive word recognition model to obtain a sensitive word result. The above-mentioned sensitive word recognition model may use an existing text recognition model or text classification model to convert the data in the data stream to obtain text information. After that, use the sensitive word recognition model to segment and feature the text information to obtain the word segmentation vector. Finally, based on the word segmentation vector, output the recognition result of each word segmentation. Among them, the result can be regarded as the above sensitive word result.

Specifically, referring to FIG. 3 , in S103, the updated N sampled data is subjected to sensitive word analysis processing, and the sensitive word result is obtained, which specifically includes the following sub-steps S1031-S1034, which are described in detail as follows:

S1031. For any sampled data, the terminal device identifies the sampled data and generates text information.

S1032. The terminal device performs word segmentation on the text information to obtain multiple text word segmentations of the text information.

S1033: The terminal device determines word segmentation vectors of the multiple text segmentations according to the respective position information of the multiple text segmentations in the preset word vector library.

S1034 , the terminal device recognizes the word segmentation of multiple texts according to the word segmentation vector, and obtains the result of the sensitive word of the sampled data.

In an embodiment, the above-mentioned process of identifying data and converting it into text information is known in the art, and will not be described.

In an embodiment, the terminal device may perform word segmentation and vector processing on the text information as follows: the terminal device pre-builds a preset word vector library, and sorts each text segmentation in the preset word vector library. Afterwards, for the text information, the terminal device can first regard the overall text information as a text word segmentation, and if there is a word segmentation consistent with the text segmentation in the text preset word vector library, the sequence number of the word segmentation in the preset word vector library is used as the word segmentation. Word segmentation vector, involved in subsequent processing. If there is no text word segmentation consistent with the overall text information in the preset word vector library, delete one character at the end of the text information. Then, the remaining text information is used as a text word segmentation, which is matched with the word segmentation in the preset word vector library again. If the match is successful, the remaining text information is used as a text word segmentation. If the matching fails, delete one character at the end of the remaining text information again, and repeat the above steps until a plurality of text segments and a segment vector corresponding to each text segment are obtained.

In one embodiment, the above-mentioned sensitive word results include, but are not limited to, results that contain sensitive words and results that do not contain sensitive words. Wherein, the result of containing the sensitive word is that the text segmentation in the sampled data is a sensitive word, and the sensitive word result may also include the sensitive word type of the sensitive word. Specifically, for the text information "Please confirm whether your ID number is "XXXX"". Among them, the text word segmentation in the text information can be "please", "confirm", "your", "identity number", "whether", "for", "XXXX". Among them, the sensitive word recognition model can determine the "ID card number" as the sensitive word type, and determine the corresponding "XXXX" segment as the sensitive word based on the sensitive word type.

Wherein, after determining the type of the sensitive word, the terminal device may also match the sensitive word from multiple text segmentations based on the matching rule corresponding to the type of the sensitive word. Exemplarily, the sensitive word "XXXX" may be recognized based on the format of the ID card (the format is an 18-digit number).

In an embodiment, for a plurality of text segmentations in any text information, if the above-mentioned sensitive word recognition model can only identify the sensitive word types of the text segmentations from the multiple text segmentations in the text information, but not identify the sensitive word types of the text segmentations. If any specific sensitive word is selected, the sensitive word recognition model can consider the text information as a sensitive word result that does not include the sensitive word.

Exemplarily, if the text information is "Please provide your ID number", the sensitive word recognition model can only identify the corresponding sensitive word type from the text information. For example, "ID number". However, specific sensitive words could not be identified. For example, based on the format of the ID card (the format is 18 digits), the sensitive word "XXXX" that meets the requirements cannot be identified from multiple text segmentations. Therefore, the sensitive word recognition model can output the result that no sensitive words exist in the text information.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据。以此，可以使终端设备在接收到当前数据时即进行采样处理，而不是在接收数据流中所有的数据后在进行采样处理。并且，可避免终端设备使用现有降采样方法的局限性，使终端设备能够提供相同的被采样的概率值对数量未知的数据进行降采样处理。进而，可以使终端设备在对采样效果更好的采样数据进行敏感词分析处理时，其得到的敏感词结果更接近于整体的数据流中各个数据的敏感词结果。In this embodiment, after copying the data stream returned from the server to the client, the copied data stream is processed without directly processing the returned data stream. , it can also ensure that the data returned by the server to the client will not cause delay problems. After that, based on the processing order M in which the current data in the copied data stream is sampled,

Update the determined N sampled data with the probability of , and obtain the updated N sampled data. In this way, the terminal device can perform sampling processing when receiving the current data, instead of performing sampling processing after receiving all the data in the data stream. Moreover, the limitation of using the existing down-sampling method by the terminal device can be avoided, so that the terminal device can provide the same sampled probability value to perform down-sampling processing on an unknown amount of data. Furthermore, when the terminal device performs sensitive word analysis processing on the sampled data with better sampling effect, the sensitive word result obtained by the terminal device is closer to the sensitive word result of each data in the overall data stream.

In one embodiment, referring to FIG. 4 , in S101, the data stream returned by the server is copied to obtain the copied data stream, which can be specifically implemented through the following sub-steps S1011-S1012, which are described in detail as follows:

S1011. For any data in the data stream, the terminal device identifies the data structure of the data.

S1012: The terminal device copies the data belonging to the target data structure in the data to obtain copy data corresponding to the data.

In one embodiment, the above-mentioned data is usually sent in the form of a message. Therefore, the data structure of the above data can also be regarded as a message structure. Among them, the message structure is usually: status line <status-line>; message header <headers>; response body [<response-body>]. Among them, the status line contains information such as the version of the server HTTP protocol and the response status. The message header contains some additional information about the response message. For example, the type of response body, the length of the response body, and the encoding format used. The response body contains the specific content sent by the server to the client, that is, the content requested by the client.

Based on this, the terminal device may consider that the content in the response message is the required substantive data. Therefore, the terminal device can identify the message structure of the message, and determine the structure of the response body as the target data structure, so as to copy the data contained in the response body to obtain the copied data.

It can be understood that, when the terminal device copies part of the data in a packet based on the traffic bypass technology, the copied data can not only include the substantial data returned by the server, but also reduce the amount of data that the terminal device needs to copy.

In one embodiment, referring to FIG. 5 , in S103, the sensitive word analysis processing is performed on the updated N sample data, and after obtaining the sensitive word result, the following steps S131-S132 are further included, which are described in detail as follows:

S131. The terminal device counts the number of sampled data including the sensitive word according to the result of the sensitive word.

S132. If the ratio between the number of sampled data containing sensitive words and N is greater than a preset value, the terminal device performs sensitive word analysis processing on all data streams returned by the server.

In an embodiment, the above-mentioned preset value may be set by the staff in advance according to the actual situation, which is not limited. It should be added that if the number of data contained in the copied data stream is less than N, it can be known based on the description in S103 that all data in the copied data stream will be used as sample data. Therefore, when calculating the ratio, the ratio between the number of sampled data containing sensitive words and the total number of sampled data is calculated.

In an embodiment, based on the description in S103 above, it can be known that the terminal device can analyze and process each sampled data to obtain a sensitive word result of each sampled data. Based on this, according to the sensitive word result of each sampled data, the terminal device can count the number of sampled data containing sensitive words. Then, the ratio of the number to the number N of sampled data is calculated, and when it is determined that the ratio is greater than the preset value, it is determined that the data stream returned by the server to the client contains a large number of sensitive words, which may have the risk of leaking sensitive data.

Based on this, the terminal device can directly perform sensitive word analysis processing on the data stream returned by the server next time (in this case, the returned data stream may not be copied), and then perform data extraction on the sensitive words contained in it according to the sensitive word result. sensitive to prevent information leakage. Specifically, referring to FIG. 6 , in the process of performing sensitive word analysis on the data stream returned by the server next time, the following sub-steps S1321--S1322 are specifically included, which are described in detail as follows:

S1321 , the terminal device performs desensitization processing on the sensitive words contained in the data stream returned by the server to obtain desensitized data.

S1322, the terminal device sends the desensitization data to the client.

Among them, data desensitization refers to deforming sensitive words through desensitization rules to obtain desensitized data, so as to achieve reliable protection of sensitive words. Specifically, the desensitization rules can be set by the staff in the terminal device and the client in advance, so that the client can restore the desensitized data based on the same desensitization rules, and obtain the original data returned by the server.

Please refer to FIG. 7 . FIG. 7 is a structural block diagram of an apparatus for analyzing sensitive data provided by an embodiment of the present application. Each module included in the sensitive data analysis apparatus in this embodiment is used to execute each step in the embodiment corresponding to FIG. 1 to FIG. 6 . For details, please refer to FIG. 1 to FIG. 6 and the related descriptions in the embodiments corresponding to FIG. 1 to FIG. 6 . For convenience of explanation, only the parts related to this embodiment are shown. Referring to FIG. 7, the sensitive data analysis apparatus 700 may include: a copying module 710, a sampling module 720 and an analysis module 730, wherein:

The copying module 710 is configured to copy the data stream returned by the server to obtain the copied data stream.

的概率对已确定的N个采样数据进行更新，得到更新后的N个采样数据。The sampling module 720 is configured to, for any current data in the copy data stream, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data.

The analysis module 730 is configured to perform sensitive word analysis processing on the updated N sampled data to obtain a sensitive word result.

In one embodiment, the copy module 710 is also used to:

For any data in the data stream, identify the data structure of the data; copy the data belonging to the target data structure in the data to obtain the copy data corresponding to the data.

In one embodiment, the sensitive data analysis apparatus 700 further includes:

The determining module is configured to determine that the current data is sampled data if M≤N, until the determined N sampled data are obtained.

In one embodiment, the sampling module 720 is further configured to:

的概率确定当前数据不为采样数据，则保持已确定的N个采样数据不变；若以

的概率确定当前数据为采样数据，则以

的概率从已确定的N个采样数据中，确定一个需被当前数据更新的替换采样数据；将当前数据更新替换采样数据，得到更新后的N个采样数据。if

The probability of determining that the current data is not sampled data, then keep the determined N sampled data unchanged;

The probability of determining that the current data is sampled data, then use

The probability of determining a replacement sample data that needs to be updated by the current data is determined from the determined N sample data; the current data is updated to replace the sample data to obtain the updated N sample data.

In one embodiment, the analysis module 730 is also used to:

For any sampled data, identify the sampled data and generate text information; segment the text information to obtain multiple text segmentations of the text information; determine multiple texts according to the position information of the multiple text segmentations in the preset word vector library respectively The word segmentation vector of the word segmentation; according to the word segmentation vector, identify multiple text segmentations respectively, and obtain the sensitive word results of the sampled data.

In one embodiment, the sensitive data analysis apparatus 700 further includes:

The statistics module is used to count the number of sampled data containing sensitive words according to the results of sensitive words.

The processing module is configured to perform sensitive word analysis processing on all data streams returned by the server if the ratio between the number of sampled data containing sensitive words and N is greater than a preset value.

In one embodiment, the processing module is further used to:

Perform desensitization processing on the sensitive words contained in the data stream returned by the server to obtain desensitized data; send the desensitized data to the client.

It should be understood that, in the structural block diagram of the sensitive data analysis device shown in FIG. 7 , each module is used to execute each step in the embodiment corresponding to FIG. 1 to FIG. 6 , and for the embodiment corresponding to FIG. 1 to FIG. 6 Each step of the above has been explained in detail in the above-mentioned embodiments. For details, please refer to FIG. 1 to FIG. 6 and the relevant descriptions in the corresponding embodiments of FIG. 1 to FIG.

FIG. 8 is a structural block diagram of a terminal device provided by an embodiment of the present application. As shown in FIG. 8 , the terminal device 800 in this embodiment includes: a processor 810 , a memory 820 , and a computer program 830 stored in the memory 820 and executable on the processor 810 , such as a program of a sensitive data analysis method. When the processor 810 executes the computer program 830, it implements the steps in the various embodiments of the above-mentioned sensitive data analysis methods, for example, S101 to S103 shown in FIG. 1 . Alternatively, when the processor 810 executes the computer program 830, the functions of the modules in the embodiment corresponding to FIG. 7 are implemented, for example, the functions of the modules 710 to 730 shown in FIG. describe.

Exemplarily, the computer program 830 may be divided into one or more modules, and one or more modules are stored in the memory 820 and executed by the processor 810 to implement the sensitive data analysis method provided by the embodiments of the present application. One or more modules may be a series of computer program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program 830 in the terminal device 800 . For example, the computer program 830 can implement the sensitive data analysis method provided by the embodiments of the present application.

The terminal device 800 may include, but is not limited to, a processor 810 and a memory 820 . Those skilled in the art can understand that FIG. 8 is only an example of the terminal device 800, and does not constitute a limitation on the terminal device 800, and may include more or less components than the one shown, or combine some components, or different components For example, the terminal device may also include an input and output device, a network access device, a bus, and the like.

The so-called processor 810 can be a central processing unit, and can also be other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components Wait. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 820 may be an internal storage unit of the terminal device 800 , such as a hard disk or a memory of the terminal device 800 . The memory 820 may also be an external storage device of the terminal device 800 , such as a plug-in hard disk, a smart memory card, a flash memory card, etc., which are equipped on the terminal device 800 . Further, the memory 820 may also include both an internal storage unit of the terminal device 800 and an external storage device.

An embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the sensitive data analysis method in each of the foregoing embodiments is implemented. .

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the above-mentioned various implementations are implemented. Examples of sensitive data analysis methods.

In a fifth aspect, the embodiments of the present application provide a computer program product, which when the computer program product runs on a terminal device, enables the terminal device to execute the sensitive data analysis method in each of the foregoing embodiments.

The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The recorded technical solutions are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in the application. within the scope of protection.

Claims (10)

1. a sensitive data analysis method, is characterized in that, comprises: Copy the data stream returned by the server to obtain the copied data stream; For any current data in the copy data stream, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data; Sensitive word analysis processing is performed on the updated N sampled data to obtain a sensitive word result. 2. The sensitive data analysis method according to claim 1, wherein the data stream returned by the server is copied to obtain the copied data stream, comprising: For any data in the data stream, identifying the data structure of the data; Copy the data belonging to the target data structure in the data to obtain copy data corresponding to the data. 3. The sensitive data analysis method according to claim 1, characterized in that, in the processing sequence M that is sampled according to the current data, to

The probability of updating the determined N sampled data, and before obtaining the updated N sampled data, also includes: If the M≤N, the current data is determined to be the sampled data until the determined N sampled data are obtained. 4. The sensitive data analysis method according to any one of claims 1-3, characterized in that, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data, including: If said to

The probability of determining that the current data is not the sampled data, then keep the determined N sampled data unchanged; If said to

The probability of determining that the current data is the sampling data, then use

The probability of determining a replacement sample data that needs to be updated by the current data from the determined N sample data; The replacement sampling data is updated by using the current data to obtain the updated N sampling data. 5. The sensitive data analysis method according to claim 1, wherein the said updated N sampled data is subjected to sensitive word analysis processing to obtain a sensitive word result, comprising: For any sampled data, identify the sampled data, and generate text information; Segmenting the text information to obtain multiple text segmentations of the text information; Determine word segmentation vectors of the multiple text segmentations according to the respective position information of the multiple text segmentations in the preset word vector library; The multiple text word segmentations are respectively identified according to the word segmentation vector, and the sensitive word result of the sampled data is obtained. 6. The sensitive data analysis method according to claim 1, characterized in that, after performing sensitive word analysis processing on the updated N sampled data to obtain a sensitive word result, further comprising: According to the sensitive word results, count the number of sampled data containing sensitive words; If the ratio between the quantity of the sampled data including the sensitive words and the N is greater than the preset value, the sensitive word analysis processing is performed on all the data streams returned by the server. 7 . The sensitive data analysis method according to claim 6 , wherein if the ratio between the quantity of the sampled data including the sensitive word and the N is greater than a preset value, the data returned by the server Streams are processed for sensitive word analysis, including: Perform desensitization processing on the sensitive words contained in the data stream returned by the server to obtain desensitized data; The desensitized data is sent to the client. 8. A sensitive data analysis device, comprising: The copy module is used to copy the data stream returned by the server to obtain the copied data stream; Sampling module, for any current data in the copy data stream, according to the processing sequence M in which the current data is sampled, to

Update the determined N sampled data with the probability of , and obtain the updated N sampled data; An analysis module, configured to perform sensitive word analysis processing on the updated N sampled data to obtain a sensitive word result. 9. A terminal device, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that, when the processor executes the computer program, the implementation as claimed in the claims The method of any one of 1 to 7. 10 . A computer-readable storage medium storing a computer program, wherein, when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented. 11 .

Images