CN118312476A - Sensitive information detection method, device, medium and system - Google Patents

Abstract

The application discloses a sensitive information detection method, a sensitive information detection device, a sensitive information detection medium and a sensitive information detection system. The method comprises the following steps: responding to a sensitive information detection request, acquiring a file to be detected, and putting the file to be detected into a detection queue; the file type of the file to be detected comprises one or more of a text file, a picture file, an audio file and a video file; taking out the target file from the detection queue, and determining a detection algorithm according to the file type of the target file; detecting sensitive information of the target file based on a detection algorithm; and determining a detection report format according to the file type of the target file, and filling the sensitive information detection result of the target file based on the detection report format to obtain a detection report of the target file. By adopting the technical scheme, the manual participation can be reduced, the processing links are saved, the manual processing cost is saved by replacing manual operation with a machine, errors are reduced, the checking efficiency of sensitive information is improved, and the processing time of the service is greatly shortened.

Description

A sensitive information detection method, device, medium and system

Technical Field

The present application relates to the field of computer technology, and in particular to a sensitive information detection method, device, medium and system.

Background technique

With the rapid economic development in recent years, people pay more and more attention to their personal safety and property safety. Therefore, insurance services have also become the focus of people's attention.

The traditional insurance sales risk prevention and control work mode is as follows: First, after the material preparation party prepares the relevant materials, it submits an application for review to the risk control manager. After receiving the review task, the risk control manager needs to manually review each page one by one, annotate the parts that do not meet the requirements one by one, and then feedback to the applicant for modification. After the applicant makes modifications according to the feedback results, he submits the review application again, and repeats this cycle until the materials meet the relevant requirements. This process generally requires multiple rounds of repetition and communication before it can be completed, which is time-consuming and laborious. Second, the risk control manager takes the initiative to initiate an inspection of the relevant materials. The inspection method is also to manually check each material one by one. If any problems are found, the problems found in the inspection will be submitted to the problem handling department for processing. After the problem handling department completes the corresponding processing, it is also necessary to submit the rectified materials to the risk control manager for further verification and confirmation. This cycle repeats until all the problems found are rectified.

There are many problems with the existing manual review. The first is the lack of information support for pre-emptive prevention and control, which makes it difficult to implement pre-emptive prevention and control work. In addition, there is no intelligent system to support in-process control, and the review of a large amount of materials is purely manual, which is inefficient and has a high error rate. Previous risk control systems were based on the perspective of managers, mainly focused on control, and mostly handled after the event, resulting in a passive position in risk prevention and control. There are many prohibited descriptions in sales materials, and it is very difficult for managers or sales staff to fully understand all the contents. In the review of materials, it is inevitable that there will be omissions and errors in the review. Sensitive word library management lacks the support of information systems, and which words are sensitive words are only saved in various documents issued by regulators, industries, and companies. In addition, in recent years, new sales models such as short videos have not yet had effective sales risk prevention and control measures.

Summary of the invention

The embodiments of the present application provide a sensitive information detection method, device, medium and system. This solution combines many problems of current manual review and provides a technical solution for automatic sensitive information detection. The present invention provides different detection algorithms for files to be detected of different file types, and fills the detection results into the corresponding detection report after obtaining them. The sensitive information of the files to be detected can be detected and displayed, which is convenient for users to view and understand the sensitive information problems in the files to be detected. The present invention reduces manual participation, saves processing links, replaces manual work with machines, saves manual processing costs, reduces errors, improves the efficiency of sensitive information verification, and greatly shortens the processing time of the business.

The present application provides a sensitive information detection method, the method comprising:

In response to a sensitive information detection request, obtain a file to be detected, and put the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file;

Taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file;

Performing sensitive information detection on the target file based on the detection algorithm;

According to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain a detection report of the target file.

Furthermore, after putting the file to be detected into the detection queue, the method further includes:

Switching the real-time status of the sensitive information detection request from an upload status to a pending detection status;

After taking out the target file from the detection queue and determining the detection algorithm according to the file type of the target file, the method further includes:

Switching the real-time status of the sensitive information detection request from a pending detection status to a detecting status;

After obtaining the detection report of the target file, the method further includes:

Switching the real-time status of the sensitive information detection request from a detecting status to a detection completed status;

The method further comprises:

In response to the status query instruction of the sensitive information detection request, the real-time status of the sensitive information detection request is read and displayed.

Further, according to the file type of the target file, a detection report format is determined, and the sensitive information detection result of the target file is filled in based on the detection report format to obtain a detection report of the target file, including:

If the file type of the target file is a text file, the sensitive information detection result of the target file is filled into a text detection report format; wherein the text detection report format includes the amount of sensitive information and the paragraph position where the sensitive information is located;

If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information;

If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information;

If the file type of the target file is an audio file, the sensitive information detection result of the target file is filled into an audio detection report format; wherein the audio detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence where the sensitive information is located, and the start time point and the end time point of the sensitive information;

If the file type of the target file is a video file, the sensitive information detection result of the target file is filled into a video detection report format; wherein the video detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence containing the sensitive information, the start time point and the end time point of the sensitive information, and the associated video frames of the sensitive information.

Furthermore, after obtaining the detection report of the target file, the method further includes:

In response to a request to view the detection report of the target file, the detection report is obtained and displayed; and upon receiving an instruction to view details, the detected sensitive information is displayed; or upon receiving an instruction to view the source file, the target file is displayed.

Furthermore, based on the detection algorithm, sensitive information detection is performed on the target file, including:

Converting the target file into text information based on a detection algorithm;

The text information is matched with a preset sensitive word library for sensitive words.

Furthermore, the method further comprises:

In response to a query request for a sensitive word library, sensitive words in the sensitive word library are displayed according to pre-classified classification results.

Furthermore, the method further comprises:

In response to a request to report sensitive words, receiving words to be entered;

Send out vocabulary entry review requests and receive vocabulary entry review results;

If the review result is that the entry is allowed, the category to which the word to be entered belongs is determined, and the word to be entered is entered into the sensitive word library according to the category;

If the review result is that entry is not allowed, the reporting process of sensitive words will end.

The present application also provides a sensitive information detection device, the device comprising:

A module for acquiring files to be detected, used to respond to a sensitive information detection request, acquire files to be detected, and put the files to be detected into a detection queue; wherein the file types of the files to be detected include one or more of text files, picture files, audio files, and video files;

A detection algorithm determination module, used to take out a target file from the detection queue and determine a detection algorithm according to the file type of the target file;

A detection module, used to perform sensitive information detection on the target file based on the detection algorithm;

The detection report generation module is used to determine the detection report format according to the file type of the target file, and fill in the sensitive information detection result of the target file based on the detection report format to obtain the detection report of the target file.

An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the sensitive information detection method as described in the embodiment of the present application is implemented.

An embodiment of the present application also provides a sensitive information detection system, which is used to execute the sensitive information detection method as described in the embodiment of the present application.

An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the sensitive information detection method as described in the embodiment of the present application is implemented.

At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

The technical solution provided by the present application can detect and display the sensitive information of the files to be detected by using different detection algorithms for the files of different file types, and fill the detection results into the corresponding detection reports, so that users can view and understand the sensitive information problems in the files to be detected. The present invention reduces manual participation, saves processing links, replaces manual work with machines, saves manual processing costs, reduces errors, improves the efficiency of sensitive information verification, and greatly shortens the processing time of the business.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

FIG1 is a schematic diagram of a process flow of a sensitive information detection method provided in Example 1 of the present application;

FIG2 is a schematic diagram of an interaction flow of a sensitive information detection system provided in Example 1 of the present application;

FIG3 is a schematic diagram of an upload detection process of a sensitive information detection system provided in Example 1 of the present application;

FIG4 is a schematic diagram of an input detection process of a sensitive information detection system provided in Example 1 of the present application;

5 is a schematic diagram of the historical detection list interaction process of the sensitive information detection system provided in Example 1 of the present application;

FIG6 is a schematic diagram of a vocabulary query process of a sensitive information detection system provided in Example 1 of the present application;

FIG7 is a schematic diagram of a sensitive word adding process of a sensitive information detection system provided in Embodiment 1 of the present application;

FIG8 is a schematic diagram of a newly added sensitive word review process of the sensitive information detection system provided in Example 1 of the present application;

FIG9 is a schematic diagram of the structure of a sensitive information detection device provided in Example 2 of the present application;

FIG10 is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. 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 making creative work are within the scope of protection of this application.

The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

Embodiment 1

Figure 1 is a flow chart of the sensitive information detection method provided in the first embodiment of the present application. This embodiment can be applied to the case of detecting whether videos, audios, texts, and pictures contain sales-sensitive words. The method can be executed by the sensitive information detection device provided in the embodiment of the present application. The device can be implemented by software and/or hardware, and can be integrated into an electronic device for sensitive information detection.

As shown in FIG1 , the method comprises:

S110, in response to a sensitive information detection request, obtaining a file to be detected, and placing the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file;

In this solution, the sensitive information detection request may be issued by a salesperson or a user, and the request may be issued by clicking a "Sensitive Information Detection" button on the sensitive information detection system.

The files to be detected may be files uploaded by sales personnel or users, such as pictures, audios, and videos, or text content directly entered in the search bar.

In this solution, a detection queue can be maintained in the sensitive information detection system, and the detection queue can be used to put the received requests into it and detect them one by one in the order of the time they are put in.

Among them, sensitive information detection includes but is not limited to sensitive word detection. This solution can help users intelligently check whether the audio, text, and picture content of video files, audio files, common office files, and various types of pictures contain sales-sensitive words; corresponding test reports are issued for different types of files. In various test reports, you can clearly understand whether the voice, image, and text parts of this file contain sales-sensitive words, which sensitive words are contained, and highlight them to help users quickly lock sensitive words and facilitate users to quickly make corrections. The system breaks the dilemma of pure manual operation. By providing simple, fast, and convenient intelligent inspection, review, analysis and other tools, it reduces the difficulty of operation and the threshold for personnel to use, so that everyone can learn it once, understand it once, and everyone can use it. It completely subverts the working mode of manual review and improves quality and efficiency. It helps to build an intelligent line of defense for sales risk prevention and control, and at the same time fills the gap in the application of sensitive information detection in the industry.

The sensitive information detection system can be used to autonomously and proactively check whether the audio, text, and pictures of relevant documents contain sales-sensitive words. Prevent the occurrence of sales risks at the source, change passive disposal to active prevention and control, form a closed loop of risk control covering before, during, and after the event, and effectively reduce the pressure of front-line risk control. At the same time, the solution fills the gap in information technology for risk prevention and control beforehand, and moves the risk prevention and control line to the forefront of sales. This solution is the first to transform risk prevention and control from investigation and supervision to service self-control and active prevention and control, which completely changes the traditional way of risk prevention and control.

S120, taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file;

Among them, the file to be detected is taken out from the detection queue as the target file. The file type of the target file can be a single type or a mixed type, such as a mixture of text and pictures, or a mixture of audio and text. It can be known that different detection algorithms can be used for different types of target files to detect sensitive words.

S130, performing sensitive information detection on the target file based on the detection algorithm;

Among them, the detection algorithm can be used to detect sensitive information in the target disk file, such as detecting sensitive words that have appeared, or detecting other sensitive information, such as detecting sensitive image elements, or sensitive audio information, etc.

In this solution, specifically, the text content of different types of target files can be extracted, such as extracting the text content of video or audio files, and comparing it with pre-determined sensitive words to determine whether sensitive words are contained. If so, sensitive information is detected. For pictures, the text content included in the pictures can be extracted and compared with pre-determined sensitive words.

S140. Determine a detection report format according to the file type of the target file, and fill in the sensitive information detection result of the target file based on the detection report format to obtain a detection report of the target file.

Different detection report formats may be determined for different file types. It is understandable that the detection report format may be used to better present the sensitive information detection result of the current target file.

After determining the detection report format, the sensitive information detection results can be filled into the detection report format accordingly to obtain a detection report for the target file.

For example, it can intelligently identify the voice, image, and text content in video files, automatically detect sensitive information about insurance sales, and issue audio content detection reports and image content detection reports for video files respectively. In the video content detection report, the number of video frames where sensitive words appear, the order of video frames, the specific time when the video frames appear in the video file, and the sensitive words detected on the video frames are accurately marked, and the sensitive words detected on the video frames are highlighted. In the audio content detection report, it is clearly noted that each audio segment where sensitive words are detected appears at the start and end time of the entire video file, and the corresponding transcript is provided. At the same time, the detected sensitive words are accurately and highlighted in the transcript. This solution greatly improves the efficiency and quality of video file detection.

In the technical solution, optionally, according to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain the detection report of the target file, including:

If the file type of the target file is a text file, the sensitive information detection result of the target file is filled into a text detection report format; wherein the text detection report format includes the amount of sensitive information and the paragraph position where the sensitive information is located;

If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information;

If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information;

If the file type of the target file is an audio file, the sensitive information detection result of the target file is filled into an audio detection report format; wherein the audio detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence where the sensitive information is located, and the start time point and the end time point of the sensitive information;

If the file type of the target file is a video file, the sensitive information detection result of the target file is filled into a video detection report format; wherein the video detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence containing the sensitive information, the start time point and the end time point of the sensitive information, and the associated video frames of the sensitive information.

It can be understood that for text files, the detected sensitive information can be directly filled into the detection report, including the amount of sensitive information and the paragraph location where the sensitive information is located, or the location of the sensitive information can be directly highlighted.

For picture files, the text content or image content included in the picture files can be detected to determine whether there is sensitive information in the text content or image content. If so, it can be framed in the form of a rectangular frame, or the background can be highlighted, etc. The location of the sensitive information can also be specifically output, such as the pixel coordinates of the sensitive information.

For audio files, voice information can be extracted and converted into text information, so as to identify whether the text information contains sensitive information, the amount of sensitive information, the text content of the sensitive information, the text content of the sentence containing the sensitive information, and the start and end time points of the sensitive information.

For video files, voice information of different characters can be extracted and converted into text information, so as to identify whether the text information contains sensitive information, as well as the amount of sensitive information, the text content of the sensitive information, the text content of the sentence containing the sensitive information, the start time and end time of the sensitive information, and the associated video frames of the sensitive information.

This arrangement of the present scheme allows for the review of sensitive information in corresponding ways for different types of target files, thereby improving the wide applicability of the sensitive information detection system. At the same time, the formatting of the detection report for the target file in this scheme can better reflect the relevant content of the sensitive information contained in the target file, which can help improve the efficient positioning of the location of the sensitive information by viewers and improve the display effect of the detection report.

The technical solution provided by this embodiment, in response to a sensitive information detection request, obtains a file to be detected, and puts the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file; takes out a target file from the detection queue, and determines a detection algorithm according to the file type of the target file; performs sensitive information detection on the target file based on the detection algorithm; determines a detection report format according to the file type of the target file, and fills in the sensitive information detection result of the target file based on the detection report format to obtain a detection report of the target file. This solution adopts different detection algorithms for files to be detected of different file types, obtains the detection results and fills them into the corresponding detection report, and can detect and display the sensitive information of the file to be detected, so that users can watch and understand the sensitive information problems in the file to be detected. The present invention reduces manual participation, saves processing links, replaces manual work with machines, saves manual processing costs, reduces errors, improves the efficiency of sensitive information verification, and greatly shortens the processing time of the business.

In the above technical solutions, optionally, after putting the to-be-detected file into the detection queue, the method further includes:

Switching the real-time status of the sensitive information detection request from an upload status to a pending detection status;

After taking out the target file from the detection queue and determining the detection algorithm according to the file type of the target file, the method further includes:

Switching the real-time status of the sensitive information detection request from a pending detection status to a detecting status;

After obtaining the detection report of the target file, the method further includes:

Switching the real-time status of the sensitive information detection request from a detecting status to a detection completed status;

The method further comprises:

In response to the status query instruction of the sensitive information detection request, the real-time status of the sensitive information detection request is read and displayed.

Among them, the status query instruction of the sensitive information detection request can be issued manually by the user, or it can be automatically issued by the system when the user enters the corresponding interface of the sensitive information detection system. For example, when entering the display interface of the file to be detected, the real-time status of the file to be detected can be displayed at its corresponding position.

Through such a setting, this solution can help sales personnel or users to more intuitively obtain the detection status of the previously input files to be tested for status synchronization, better provide sales personnel or users with an intuitive reflection of the detection status of the files to be tested, and improve the user experience of the sensitive information detection system.

In the above technical solutions, optionally, after obtaining the detection report of the target file, the method further includes:

In response to a request to view the detection report of the target file, the detection report is obtained and displayed; and upon receiving an instruction to view details, the detected sensitive information is displayed; or upon receiving an instruction to view the source file, the target file is displayed.

Among them, the request to view the test report can be issued when the user clicks the text link "View Details" of the test report. By clicking the text link, the content and location of specific sensitive information in the test report can be entered. At this time, the salesperson or user can also click the "View Source File" button to view the input file to be tested, that is, the target file. It can be used to view the context of the information appearing in the source file, etc., so that the user can better understand under what circumstances to avoid sensitive words.

This scheme is configured to provide users with a wider range of operations after obtaining the detection report, making the user's operations on the sensitive information detection system more flexible, helping to grasp the context of sensitive information, pay attention to corresponding avoidance methods and other information.

In each of the above technical solutions, optionally, performing sensitive information detection on the target file based on the detection algorithm includes:

Converting the target file into text information based on a detection algorithm;

The text information is matched with a preset sensitive word library for sensitive words.

It is understandable that if the target file is a text file, the text information can be directly obtained, and if the target file is a file in other forms such as a picture, audio, or video, the text information can be obtained. For example, a series of mainstream AI technologies such as media content analysis, video recognition, speech recognition, speech transcription, image recognition, and OCR recognition can be used.

This scheme is set up in this way to realize intelligent recognition of whether the voice, image, and text content contained in multimedia files (video files, audio files) and non-multimedia files (common documents such as WORD, PPT, PDF, excel, TXT, and various formats of pictures such as png, jpg, jpeg, bmp, etc.) involve sensitive information on insurance sales, accurately realize independent risk prevention and intelligent risk investigation in the insurance field, and move the risk prevention and control line to the forefront of sales to achieve source control. This scheme enables more insurance risk control application scenarios by creating an AI capability engine to achieve capability output. A series of AI technologies such as media content analysis, video recognition, voice recognition, voice transcription, image recognition, and OCR recognition are implemented based on the Tencent Cloud AI interface. Through the Tencent Cloud media processing, OCR recognition, and object storage interface, they are integrated into this application and used as AI technology engine support.

In the above technical solutions, optionally, the method further includes:

In response to a query request for a sensitive word library, sensitive words in the sensitive word library are displayed according to pre-classified classification results.

In this solution, the sensitive information detection system can be provided with a sensitive word library, which can be used to search for sensitive words. For example, a user can enter the "word library query" page, first, to view all the words contained in the system's sensitive word library, second, to select a type and view the words contained in different types, and third, to manually enter a word to search and detect whether the word is a sensitive word.

Through such a setting, this solution can help sales personnel to know in advance whether the words they want to use are sensitive words, and avoid sensitive words that are likely to appear in sales information, etc. Therefore, the compliance of the text or picture content formulated by the sales personnel can be improved.

In another embodiment, optionally, the method further comprises:

In response to a request to report sensitive words, receiving words to be entered;

Send out vocabulary entry review requests and receive vocabulary entry review results;

If the review result is that the entry is allowed, the category to which the word to be entered belongs is determined, and the word to be entered is entered into the sensitive word library according to the category;

If the review result is that entry is not allowed, the reporting process of sensitive words will end.

In this solution, the sensitive information detection system can enter the "Sensitive Word Management" page, where you can first report words: select the type of words to be added, enter the keywords to be reported, and support reporting multiple words at the same time, with each word separated by a comma; second, you can view the review status of the words you reported; if you are a review administrator, third, you can view a list of reported words; fourth, you can review each word one by one: if the review is passed, select the type of words to be added, and click the "Approved" button; if the review is not passed, fill in the reason for the failure, and click the "Failed" button.

Through such a setting, this solution can allow sales personnel or users to participate in the process of collecting sensitive words. At the same time, through the review of auditors, the entry of sensitive words in the use of the open source sensitive word library can be made more standardized, thereby optimizing the performance of the sensitive information detection system.

Figure 2 is a schematic diagram of the interactive process of the sensitive information detection system provided in Example 1 of the present application. As shown in Figure 2, the user can upload the file to be detected, and the sensitive information detection system can return the sensitive information detection result by automatically calling the interface, using media content analysis, video recognition, voice recognition, voice transcription, image recognition, OCR recognition and other means, and feedback the detection report to the user.

FIG3 is a schematic diagram of the upload detection process of the sensitive information detection system provided in Example 1 of the present application, and FIG4 is a schematic diagram of the input detection process of the sensitive information detection system provided in Example 1 of the present application. As shown in FIG3 and FIG4, the uploaded file and the input file can be detected respectively, and the detection report can be fed back. For example, enter the sensitive information detection system, select "Upload Detection", select the file to be detected, and after the system automatically uploads the file, it will automatically jump to the "My Detection" interface to view the file detection progress and report details; or enter the "Input Detection" page, enter the text content, click "Start Detection", and after the system automatically uploads, it will automatically jump to the "My Detection" interface to view the input text detection progress and report details.

Figure 5 is a schematic diagram of the historical detection list interaction process of the sensitive information detection system provided in Example 1 of the present application. As shown in Figure 5, the sensitive information detection system can enter the "My Detection" page to view one's own statistics, a list of all uploaded detection files, and the corresponding detection report.

Figure 6 is a schematic diagram of the vocabulary query process of the sensitive information detection system provided in Example 1 of the present application. As shown in Figure 6, the sensitive information detection system can enter the "vocabulary query" page. First, all the words contained in the system's sensitive vocabulary can be viewed. Second, the type can be selected to view the words contained in different types respectively. Third, a word can be manually entered to search and detect whether the word is a sensitive word.

FIG7 is a schematic diagram of the sensitive word addition process of the sensitive information detection system provided in the first embodiment of the present application, and FIG8 is a schematic diagram of the newly added sensitive word review process of the sensitive information detection system provided in the first embodiment of the present application. As shown in FIG7 and FIG8, the sensitive information detection system can enter the "Sensitive Word Management" page, first, it can report words: select the type of words to be added, enter the keywords to be reported, support multiple words to be reported at the same time, and each word is separated by a comma; second, it can view the review status of the words reported by itself; if it is a review administrator, third, it can view the list of reported words; fourth, it can review each word one by one: if the review is passed, select the type of words to be added, click the "Approved" button, if the review is not passed, fill in the reason for the failure, and click the "Failed" button.

The technical solution provided in this application is committed to promoting the intelligentization of insurance risk governance and constantly exploring the application of new technologies such as artificial intelligence in the field of insurance risk management. The application is mainly based on a series of mainstream AI technologies such as media content analysis, video recognition, voice recognition, voice transcription, image recognition, OCR recognition, etc., to realize intelligent recognition of multimedia files (video files, audio files) and non-multimedia files (common documents such as WORD, PPT, PDF, excel, TXT, png, jpg, jpeg, bmp and other formats of pictures, etc.) The voice, image, and text content contained in the content, whether it involves sensitive information on insurance sales, accurately realizes risk prevention self-checking and risk investigation intelligent check in the insurance field, and advances the risk prevention and control line to the forefront of sales to achieve source control; pioneers the transformation of risk prevention and control from investigation and supervision to service self-control; completely changes the traditional risk management working method, and changes passive disposal to active prevention and control; pioneers the "co-construction, sharing, and common use" model of sensitive word library, establishes a sales risk prevention and control brain, and keeps pace with the times; by creating an AI capability engine, realizes capability output, and empowers more insurance risk control application scenarios. A series of AI technologies such as media content analysis, video recognition, speech recognition, speech transcription, image recognition, OCR recognition, etc. are implemented based on Tencent Cloud AI interface. They are integrated into this application through Tencent Cloud media processing, OCR recognition, and object storage interface as AI technology engine support.

This solution pioneered the "co-construction, sharing, and use" model for sensitive word libraries, pioneered the full process of digital management of sensitive word libraries, and established a sales risk prevention and control brain.

This solution is the first to create a dedicated insurance sales sensitive word library. The insurance sales sensitive word library is established based on regulatory, industry, and company-related documents, and each sensitive word is classified and organized according to different categories. In addition, the system has established a set of standard and complete digital management processes for sensitive word libraries. In this process, the system divides users into two roles: ordinary users and approval administrators. In the review stage, ordinary users and approval administrators can review the sensitive words currently adopted and used in the system at any time. In the reporting stage, ordinary users can report sensitive words missing in the word library at any time, but they need to be approved by the approval administrator. Only sensitive words that have been reviewed and approved by the approval administrator can be officially adopted and used. At the same time, the approval administrator can directly enter the sensitive words reported by himself into the library, and once they are entered into the library, they will be officially adopted and used. In the approval stage, ordinary users can view the review results and rejection reasons in real time; the review administrator can make an application for approval or rejection of the reported sensitive words. If the operation is "agree to adopt", it is necessary to check the category to which the word belongs before it can be entered into the library. If the operation is "reject application", the review administrator must also fill in the rejection reason and reason.

The entire process of digital management of sensitive word libraries covers reporting, approval, and review. While realizing digital management, it introduces the Internet thinking of "openness, co-construction, and sharing" to ensure that the content continues to follow changes in national regulatory requirements and keep pace with the times.

In addition, this solution uses new AI technologies such as media content analysis and video recognition to detect sensitive information in video files.

This solution intelligently identifies the voice, image, and text content in video files, automatically detects sensitive information about insurance sales, and issues audio content detection reports and picture content detection reports for video files. In the picture content detection report, the number of video frames where sensitive words appear, the order of video frames, the specific time when the video frames appear in the video file, and the sensitive words detected on the video frames are accurately marked, and the sensitive words detected on the video frames are highlighted. In the audio content detection report, the start and end time of each audio segment where sensitive words are detected in the entire video file are clearly noted, and the corresponding transcript is provided. At the same time, the detected sensitive words are accurately and highlighted in the transcript. Greatly improve the efficiency and quality of video file detection.

Secondly, this solution uses technologies such as speech recognition and transcription to detect sensitive information in audio files.

This solution automatically identifies the content of audio files as text content and performs intelligent detection on the text content. The solution issues two audio file detection reports: one is to feedback the detection results of each audio file where sensitive words are detected, including the start and end time of the audio file in the entire audio file, provide the corresponding transcript, and accurately and highlight the detected sensitive words in the transcript; the other is to feedback a complete detection report, provide the full text transcript of the entire audio file, and highlight the detected sensitive words for insurance sales in the transcript.

In addition, this solution uses AI technologies such as image recognition and OCR recognition to detect sensitive information in non-multimedia files.

This solution supports the detection of sales sensitive words in various common file types such as WORD, PPT, TXT, EXCEL, WPS, PDF, and various formats of pictures such as png, jpg, jpeg, bmp, etc. The specific page number where sensitive words are detected will be clearly indicated in the test report, and the sensitive words detected in the text and picture parts of the page will be noted separately, and the detected sensitive words will be highlighted to help users quickly lock in sensitive words.

In addition, this solution records each user's detection and issues a personal detection behavior analysis report for each user. Users can clearly understand their use of the system, the detection of illegal words, and the top five sensitive words in their personal files, helping users understand their word usage habits and avoid using these sensitive words repeatedly in future work. At the same time, the system retains the uploaded source files for easy verification and use afterwards.

Finally, this solution creates a sensitive information detection capability engine to realize a new model of scientific and technological services.

This solution is based on Tencent Cloud media processing, OCR recognition, object storage interface and other technologies to build a technical system for sensitive information detection of multimedia files and non-multimedia files in the China Life system. A set of file sensitive information detection APIs that support video, audio, common office file types and pictures have been developed. The API detects sensitive words in files and returns the detection results. The file sensitive information detection API is open to the entire China Life insurance system, realizes the output of detection capabilities, supports rapid docking with front-end applications, enables more detection application scenarios, improves the efficiency of technology applications, and realizes a new model of technology services. At present, the solution has successfully empowered China Life's education and training related systems and realized intelligent prevention and control of education and training courseware.

Compared with the traditional sensitive information review method, this solution has the following advantages:

1. Disruptive innovation, moving the risk prevention and control line forward to the forefront of sales.

This solution realizes self-inspection of risk prevention and achieves the purpose of self-control. Insurance company sales staff, office staff or managers can use the Zhichacha system to self-check whether the voice, image and text of relevant documents contain sales-sensitive words before sending WeChat, Douyin, Xiaohongshu, Weibo, or after completing promotional videos, training PPT and other materials, strengthen the self-regulation of insurance practitioners, and provide strong support for the realization of self-inspection in advance and the joint prevention of insurance sales misleading risks by all employees. Prevent the occurrence of sales risks from the source and completely subvert the traditional working model.

2. Empowered by technology, we are the first to explore the technical system and application practice of file sensitive information detection in the insurance industry, filling the gap in the application of sensitive information detection in the company's system.

This solution realizes intelligent risk screening and achieves the purpose of intelligent prevention and control. The system supports the detection of sensitive words for insurance sales in various types of video files, audio files, common office files, and various types of pictures. At the same time, the detection content includes the voice, image, and text parts contained in the file. The detection content is comprehensive and has no blind spots. At the same time, corresponding test reports are issued for different types of files. In various test reports, you can clearly understand whether the voice, image, and text parts of this file contain sales sensitive words, which sensitive words are contained, and highlight them to help users quickly lock in sensitive words and facilitate users to quickly make corrections. This solution is the first to make insurance risk prevention and control work information-based, intelligent, and tool-based, completely subverting the working mode of manual review and improving quality and efficiency.

3. The first full process of digital management of sensitive vocabulary.

This solution brings together multiple documents related to supervision, industry, and companies, and creates the first exclusive insurance sales sensitive word library. As of now, it has officially been put into use, including non-compliant descriptions such as suspected speculation, suspension, limited sales, and panic buying, suspected non-compliant descriptions such as confusing the nature of the product or confusing the product development entity, suspected non-compliant descriptions of giving or promising to give the policyholder, the insured, or the beneficiary benefits beyond the insurance contract, suspected non-standard terms for recruiting salesmen, and suspected absolute descriptions that violate the "Advertising Law". It provides users with a unified, standard, and digital insurance sales sensitive word library, which is convenient for users to access and use anytime, anywhere. At the same time, the system has established a complete digital management process for sensitive word libraries, covering reporting, approval, and review, etc., to achieve digital management while ensuring that the content continues to follow up on changes in national regulatory requirements and keep pace with the times.

4. By building an AI capability engine, we can achieve capability output and enable more insurance risk control application scenarios.

This solution opens the file detection capability interface to the entire China Life Insurance system. The API detects sensitive words in files and returns the detection results. By providing an API, the development cost and time cost of developers are reduced, the security and stability of the system are improved, the consistency of data is guaranteed, and room for improvement is reserved for the expansion of new functions of the system. This solution creates an AI capability engine for sensitive information detection, realizes the empowerment output of AI detection capabilities, supports rapid access to front-end applications, enables intelligent application scenarios in various fields of business management, smart risk control, and other fields of the company, and improves the level of technological empowerment. Currently, this solution has successfully empowered the education and training field and realized the intelligent prevention and control of education and training courseware.

5. Comprehensively improve sales risk prevention and control levels to meet diverse office needs.

This solution enables users to conduct self-checking and intelligent checking. Users can upload files to be tested, and the system will issue a test report after the intelligent test. Users can view the detailed test report and quickly understand the situation of sensitive words in the file. At the same time, the system helps users quickly lock sensitive words, which is convenient for users to make quick corrections. Users can also query the situation of sensitive words at any time and implement the reporting, approval, and rejection of sensitive words. At the same time, the solution provides two versions, which support mobile and PC use respectively. The two versions have the same functions, operations, and data sharing to meet the needs of diversified offices.

Compared with the current manual detection method, the technical advantages of the technical solution provided by this application are:

This proposal applies to disclose a method and system for proactive self-control of risks. In accordance with the idea of "preparing for a rainy day and preventing problems before they occur", the proposal develops the "Zhi Cha Cha" system to realize self-checking of risk prevention. Before sending WeChat, Douyin, Xiaohongshu, Weibo, or after completing promotional videos, training PPTs and other materials, users can use the Zhi Cha Cha system to autonomously and proactively check whether the audio, text, and pictures of relevant documents contain sales-sensitive words. Prevent the occurrence of sales risks at the source, change passive disposal to active prevention and control, form a closed loop of risk control covering before, during, and after the event, and effectively reduce the pressure of front-line risk control. At the same time, the proposal fills the gap in information technology for risk prevention and control beforehand, and moves the risk prevention and control line of defense to the forefront of sales. This proposal pioneered the transformation of risk prevention and control from investigation and supervision to service self-control and proactive prevention and control, which completely changed the traditional way of risk prevention and control.

This proposal discloses an intelligent risk prevention and control method and system. The proposal focuses on studying and exploring how to implement new technologies such as artificial intelligence in the field of insurance risk management, and promote the intelligent, digital and tool-based management of insurance risks. This proposal launches the "Smart Check" system based on AI technologies such as Tencent Cloud media processing, OCR recognition, and object storage interface. The system can help users intelligently check whether the audio, text, and picture content of video files, audio files, common office files, and various types of pictures contain sales-sensitive words; corresponding test reports are issued for different types of files. All types of test reports can clearly understand whether the voice, image, and text parts of this file contain sales-sensitive words, which sensitive words are contained, and highlight them to help users quickly lock sensitive words and facilitate users to quickly rectify them. The system breaks the dilemma of pure manual operation. By providing simple, fast, and convenient intelligent inspection, review, analysis and other tools, it reduces the difficulty of operation and the threshold for personnel to use, so that everyone can learn and understand it once they use it, and everyone can use it, which completely subverts the working mode of manual review and improves quality and efficiency. It helps build an intelligent line of defense for sales risk prevention and control, and at the same time fills the gap in sensitive information detection applications in the industry. It is an invention and innovation in the field of technical services.

This proposal discloses a risk prevention and control management method and system. The proposal pioneered a sales risk sensitive word library, which was established based on regulatory, industry, and company-related documents, and each sensitive word was classified and sorted according to different categories. In addition, a set of standard and complete online digital sensitive word library management processes was established, covering reporting, approval, and review, to achieve closed-loop digital management of sensitive words. The management of the sensitive word library adopts the "co-construction, sharing, and common use" model, introduces the open, shared, and win-win Internet thinking, breaks the restrictions of lines, positions, and regions, and all system users can participate in the construction of sensitive words and become the builders, users, and ultimate beneficiaries of the system. The digital sensitive word library management process ensures that the content of the word library continues to follow up on changes in national regulatory requirements and keeps pace with the times. At the same time, by providing a standard and unified sensitive word library, it ensures the real-time, effectiveness, and consistency of the implementation of various sales risk prevention and control policies, which is a subversive innovation in the risk prevention and control management model.

This proposal discloses a method and system for outputting risk prevention and control capabilities. This proposal is based on Tencent Cloud media processing, OCR recognition, object storage interface and other technologies to build a technical system for sensitive information detection of multimedia files and non-multimedia files in the China Life system, and develops a set of file sensitive information detection APIs that support video files, audio files, common office files, and various types of pictures. The API detects sensitive words in files and returns the detection results. The file sensitive information detection API is open to the entire China Life life insurance system, realizes the output of risk prevention and control capabilities, supports rapid access to front-end applications, and enables the company's various business management, education and training, smart risk control and other intelligent risk control application scenarios in more fields, improves the level of technological empowerment, improves the efficiency of technological applications, and realizes a new model of technological services.

Embodiment 2

FIG9 is a schematic diagram of the structure of a sensitive information detection device provided in Example 2 of the present application. As shown in FIG9 , the device includes:

The to-be-detected file acquisition module 910 is used to respond to the sensitive information detection request, acquire the to-be-detected file, and put the to-be-detected file into the detection queue; wherein the file type of the to-be-detected file includes one or more of a text file, a picture file, an audio file, and a video file;

A detection algorithm determination module 920, configured to take out a target file from the detection queue and determine a detection algorithm according to a file type of the target file;

A detection module 930, configured to perform sensitive information detection on the target file based on the detection algorithm;

The detection report generation module 940 is used to determine the detection report format according to the file type of the target file, and fill in the sensitive information detection result of the target file based on the detection report format to obtain the detection report of the target file.

This device can execute the sensitive information detection method provided in the above embodiments, and has corresponding functional units and beneficial effects, which will not be described in detail here.

Embodiment 3

The embodiment of the present application further provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to perform a sensitive information detection method when executed by a computer processor, the method comprising:

In response to a sensitive information detection request, obtain a file to be detected, and put the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file;

Taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file;

Performing sensitive information detection on the target file based on the detection algorithm;

According to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain a detection report of the target file.

Storage medium - any of various types of memory electronic devices or storage electronic devices. The term "storage medium" is intended to include: installation media, such as CD-ROM, floppy disk or tape device; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disk or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the computer system in which the program is executed, or may be located in a different second computer system, which is connected to the computer system via a network (such as the Internet). The second computer system can provide program instructions to the computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations (for example, in different computer systems connected by a network). The storage medium may store program instructions (for example, embodied as a computer program) that can be executed by one or more processors.

Of course, the storage medium containing computer-executable instructions provided in an embodiment of the present application, whose computer-executable instructions are not limited to the sensitive information detection operations described above, can also execute related operations in the sensitive information detection method provided in any embodiment of the present application.

Embodiment 4

The embodiment of the present application further provides a sensitive information detection system, wherein the sensitive information detection system is used to execute a sensitive information detection method, the method comprising:

In response to a sensitive information detection request, obtain a file to be detected, and put the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file;

Taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file;

Performing sensitive information detection on the target file based on the detection algorithm;

According to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain a detection report of the target file.

By providing the above-mentioned sensitive information detection system, this embodiment can adopt different detection algorithms for files to be detected of different file types, fill in the corresponding detection report after obtaining the detection results, and detect and display the sensitive information of the files to be detected, so that users can view and understand the sensitive information problems in the files to be detected. The present invention reduces manual participation, saves processing links, replaces manual work with machines, saves manual processing costs, reduces errors, improves the efficiency of sensitive information verification, and greatly shortens the processing time of the business.

Embodiment 5

The embodiment of the present application provides an electronic device. FIG10 is a schematic diagram of the structure of an electronic device provided in the fifth embodiment of the present application. As shown in FIG10 , the present embodiment provides an electronic device 1000, which includes: one or more processors 1020; a storage device 1010, which is used to store one or more programs, and when the one or more programs are run by the one or more processors 1020, the one or more processors 1020 implement the sensitive information detection method provided in the embodiment of the present application, and the method includes:

In response to a sensitive information detection request, obtain a file to be detected, and put the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file;

Taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file;

Performing sensitive information detection on the target file based on the detection algorithm;

According to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain a detection report of the target file.

The electronic device 1000 shown in FIG. 10 is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present application.

As shown in FIG10 , the electronic device 1000 includes a processor 1020, a storage device 1010, an input device 1030, and an output device 1040; the number of processors 1020 in the electronic device may be one or more, and FIG10 takes one processor 1020 as an example; the processor 1020, the storage device 1010, the input device 1030, and the output device 1040 in the electronic device may be connected via a bus or other means, and FIG10 takes the connection via a bus 1050 as an example.

The storage device 1010, as a computer-readable storage medium, can be used to store software programs, computer-executable programs and module units, such as program instructions corresponding to the sensitive information detection method in the embodiment of the present application.

The storage device 1010 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the storage device 1010 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the storage device 1010 may further include a memory remotely arranged relative to the processor 1020, and these remote memories may be connected via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

The input device 1030 may be used to receive input numbers, character information or voice information, and generate key signal inputs related to user settings and function control of the electronic device. The output device 1040 may include electronic devices such as a display screen and a speaker.

The electronic device provided in the embodiment of the present application can detect and display the sensitive information of the file to be detected by using different detection algorithms for the files of different file types, and fill the detection results into the corresponding detection report after obtaining them, so as to facilitate the user to view and understand the sensitive information problems existing in the file to be detected. The present invention reduces manual participation, saves processing links, replaces manual work with machines, saves manual processing costs, reduces errors, improves the efficiency of checking sensitive information, and greatly shortens the processing time of the business.

The sensitive information detection device, medium and system provided in the above embodiments can run the sensitive information detection method provided in any embodiment of the present application, and have the corresponding functional modules and beneficial effects of running the method. For technical details not described in detail in the above embodiments, please refer to the sensitive information detection method provided in any embodiment of the present application.

Those skilled in the art will appreciate that the embodiments of the present application provide 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 combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

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 produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

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

The memory may include non-permanent storage 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.

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 tape 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.

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.

The above is only an 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 changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (10)

1. A sensitive information detection method, characterized in that the method comprises: In response to a sensitive information detection request, obtain a file to be detected, and put the file to be detected into a detection queue; wherein the file type of the file to be detected includes one or more of a text file, a picture file, an audio file, and a video file; Taking out a target file from the detection queue, and determining a detection algorithm according to a file type of the target file; Performing sensitive information detection on the target file based on the detection algorithm; According to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain a detection report of the target file. 2. The method according to claim 1, characterized in that after putting the file to be detected into the detection queue, it also includes: Switching the real-time status of the sensitive information detection request from an upload status to a pending detection status; After taking out the target file from the detection queue and determining the detection algorithm according to the file type of the target file, the method further includes: Switching the real-time status of the sensitive information detection request from a pending detection status to a detecting status; After obtaining the detection report of the target file, the method further includes: Switching the real-time status of the sensitive information detection request from a detecting status to a detection completed status; The method further comprises: In response to the status query instruction of the sensitive information detection request, the real-time status of the sensitive information detection request is read and displayed. 3. The method according to claim 1 is characterized in that, according to the file type of the target file, a detection report format is determined, and based on the detection report format, the sensitive information detection result of the target file is filled in to obtain the detection report of the target file, comprising: If the file type of the target file is a text file, the sensitive information detection result of the target file is filled into a text detection report format; wherein the text detection report format includes the amount of sensitive information and the paragraph position where the sensitive information is located; If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information; If the file type of the target file is an image file, the sensitive information detection result of the target file is filled into an image detection report format; wherein the image detection report format includes the amount of sensitive information, the text content of the sensitive information, and the pixel location of the sensitive information; If the file type of the target file is an audio file, the sensitive information detection result of the target file is filled into an audio detection report format; wherein the audio detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence where the sensitive information is located, and the start time point and the end time point of the sensitive information; If the file type of the target file is a video file, the sensitive information detection result of the target file is filled into a video detection report format; wherein the video detection report format includes the amount of sensitive information, the text content of the sensitive information, the text content of the sentence containing the sensitive information, the start time point and the end time point of the sensitive information, and the associated video frames of the sensitive information. 4. The method according to claim 1, characterized in that after obtaining the detection report of the target file, it also includes: In response to a request to view the detection report of the target file, the detection report is obtained and displayed; and upon receiving an instruction to view details, the detected sensitive information is displayed; or upon receiving an instruction to view the source file, the target file is displayed. 5. The method according to claim 1, characterized in that the sensitive information detection of the target file based on the detection algorithm comprises: Converting the target file into text information based on a detection algorithm; The text information is matched with a preset sensitive word library for sensitive words. 6. The method according to claim 5, characterized in that the method further comprises: In response to a query request for a sensitive word library, sensitive words in the sensitive word library are displayed according to pre-classified classification results. 7. The method according to claim 5, characterized in that the method further comprises: In response to a request to report sensitive words, receiving words to be entered; Send out vocabulary entry review requests and receive vocabulary entry review results; If the review result is that the entry is allowed, the category to which the word to be entered belongs is determined, and the word to be entered is entered into the sensitive word library according to the category; If the review result is that entry is not allowed, the reporting process of sensitive words will end. 8. A sensitive information detection device, characterized in that the device comprises: A module for acquiring files to be detected, used to respond to a sensitive information detection request, acquire files to be detected, and put the files to be detected into a detection queue; wherein the file types of the files to be detected include one or more of text files, picture files, audio files, and video files; A detection algorithm determination module, used to take out a target file from the detection queue and determine a detection algorithm according to the file type of the target file; A detection module, used to perform sensitive information detection on the target file based on the detection algorithm; The detection report generation module is used to determine the detection report format according to the file type of the target file, and fill in the sensitive information detection result of the target file based on the detection report format to obtain the detection report of the target file. 9. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the sensitive information detection method as described in any one of claims 1 to 7 is implemented. 10. A sensitive information detection system, characterized in that the sensitive information detection system is used to execute the sensitive information detection method as described in any one of claims 1-7.