Sensitive information detection method and system

Abstract

The invention discloses a sensitive information detection method and system. The method comprises the following steps: acquiring to-be-detected text information; performing detection processing on the to-be-detected text information to obtain target text detection information; the target text detection information comprises first target detection information, and / or second target detection information, and / or third target detection information; the first target detection information comprises L pieces of first target detection word information; the second target detection information comprises M pieces of second target detection word information; the third target detection information comprises N pieces of third target detection word information; and performing optimization processing on the target text detection information to obtain target detection result information.

Description

Technical Field

[0001] The present invention relates to the technical field of information detection, and particularly to a sensitive information detection method and system. Background Art

[0002] In the digital age, the management and protection of sensitive information are crucial. However, with the explosive growth of its types and quantities, sensitive information detection technologies are facing challenges. On the one hand, the improper dissemination of sensitive information will bring many risks, and enterprises and organizations also need to accurately identify sensitive information. On the other hand, existing sensitive information detection technologies have limitations, such as poor quality of sensitive data sets, insufficient flexibility of traditional rule detection methods, high computational costs and poor adaptability of deep learning models, etc. Therefore, there is a need to provide a sensitive information detection method and system to improve the accuracy and efficiency of sensitive information identification, and enhance the accuracy and comprehensiveness of sensitive information recognition. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a sensitive information detection method and system, which is beneficial to improving the accuracy and efficiency of sensitive information identification, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0004] To solve the above technical problem, in the first aspect, an embodiment of the present invention discloses a sensitive information detection method, which includes:

[0005] Obtain the text information to be detected;

[0006] Perform detection processing on the text information to be detected to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L first target detection word information; the second target detection information includes M second target detection word information; the third target detection information includes N third target detection word information;

[0007] Perform optimization processing on the target text detection information to obtain target detection result information.

[0008] In the second aspect, an embodiment of the present invention discloses a sensitive information detection system, which includes:

[0009] An acquisition module, configured to obtain the text information to be detected;

[0010] A second processing module, configured to detect and process the to-be-detected text information to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L pieces of first target detection word information; the second target detection information includes M pieces of second target detection word information; the third target detection information includes N pieces of third target detection word information;

[0011] A third processing module, configured to optimize the target text detection information to obtain target detection result information.

[0012] A third aspect of the present invention discloses another sensitive information detection system, the system includes:

[0013] A memory storing executable program code;

[0014] A processor coupled to the memory;

[0015] The processor calls the executable program code stored in the memory and executes some or all of the steps in the sensitive information detection method disclosed in the first aspect of the embodiments of the present invention.

[0016] A fourth aspect of the present invention discloses a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute some or all of the steps in the sensitive information detection method disclosed in the first aspect of the embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0018] Figure 1 is a scenario schematic diagram of the sensitive information detection system provided by the embodiments of the present invention;

[0019] Figure 2 is a flowchart of a sensitive information detection method disclosed in the embodiments of the present invention;

[0020] Figure 3 is a structural schematic diagram of a sensitive information detection system disclosed in the embodiments of the present invention;

[0021] Figure 4 is a structural schematic diagram of another sensitive information detection system disclosed in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0023] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices.

[0024] Referring to "embodiments" herein means that a particular feature, structure, or characteristic described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0025] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description". Any embodiment described as "exemplary" in this application is not necessarily to be construed as more preferred or advantageous than other embodiments. In order to enable any person skilled in the art to implement and use this application, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those of ordinary skill in the art can recognize that this application can be implemented without using these specific details. In other instances, well-known structures and processes will not be elaborated in detail to avoid unnecessary details from obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed in this application.

[0026] It should be noted that since the method of the embodiments of this application is executed in a computer device, the processing objects of each computer device exist in the form of data or information. For example, time, which is actually time information. It can be understood that in subsequent embodiments, if dimensions, quantities, positions, etc. are mentioned, they are all corresponding data existences for the computer device to process, and specific details are not elaborated here.

[0027] It should be noted that a brief description of the artificial intelligence-related technologies that may be involved in this application is provided. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. Artificial intelligence also studies the design principles and implementation methods of various intelligent machines to enable the machines to have the functions of perception, reasoning, and decision-making.

[0028] Artificial intelligence technology is an interdisciplinary subject with a wide range of fields involved, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0029] Computer Vision Technology (CV) Computer vision is a science that studies how to enable machines to "see". Further, it refers to using cameras and computers to replace human eyes for tasks such as object recognition and measurement in machine vision, and further performing graphic processing to make the images processed by the computer more suitable for human eyes to observe or be transmitted to instruments for detection. As a scientific discipline, computer vision studies related theories and technologies and attempts to establish artificial intelligence systems that can obtain information from images or multi-dimensional data. Computer vision technology usually includes technologies such as image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, and also includes common biometric recognition technologies such as face recognition and fingerprint recognition.

[0030] Single-modal information is data of only one type, such as one of the data information types like text, image, audio, video, electromagnetic signals, etc. Multi-modal information is data information that includes at least two types of single-modal information. Further, multi-modal information is applicable to complex tasks that require integrating multiple information sources, such as sentiment analysis, robot interaction, autonomous driving, etc. By integrating information of multiple modalities, higher performance and accuracy can usually be achieved in tasks.

[0031] Large models refer to artificial neural network models with a very large number of parameters. In the field of artificial intelligence, large models usually refer to models with hundreds of millions to trillions of parameters. The models usually need to be trained on large-scale datasets and require a large amount of computing resources for optimization and adjustment. Large models are usually used to solve complex natural language processing, computer vision, speech recognition and other tasks. Generative AI is a type of AI that can create new content and ideas, including conversations, stories, images, videos and music. In the embodiments of this application, the large model can be large-scale language models such as ChatGPT, BERT, XLNet, Zhipu Model, Claude, Moonshot AI Model, ChatGLM Model, Tongwen Qianyi Model, MiniMax Model, Spark Model, Llama Model, 360GPT Model, Qwen Model, Baichuan Model, Lark Model, vivoLM Model and Wenxin Yiyan, and the embodiments of this application do not make any limitations.

[0032] The embodiments of this application provide a sensitive information detection method, system, computer device and computer-readable storage medium, which will be described in detail below.

[0033] Please refer to Figure 1 , Figure 1 , which is a schematic diagram of the scenario of the sensitive information detection system provided by the embodiments of this application. The sensitive information detection system may include a computer device 100, and the sensitive information detection system is integrated in the computer device 100, such as Figure 1 the computer device in

[0034] In the embodiments of this application, the computer device 100 is mainly used to obtain the text information to be detected;

[0035] perform detection processing on the text information to be detected to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L first target detection word information; the second target detection information includes M second target detection word information; the third target detection information includes N third target detection word information;

[0036] perform optimization processing on the target text detection information to obtain target detection result information.

[0037] It can improve the accuracy and efficiency of sensitive information recognition, and enhance the accuracy and comprehensiveness of sensitive information recognition.

[0038] In the embodiments of the present application, the computer device 100 may be an independent server or a server network or server cluster composed of servers. For example, the computer device 100 described in the embodiments of the present application includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. Among them, the cloud server is composed of a large number of computers or network servers based on cloud computing.

[0039] It can be understood that the computer device 100 used in the embodiments of the present application may be a device that includes both receiving and transmitting hardware, that is, a device having receiving and transmitting hardware capable of performing two-way communication on a two-way communication link. Such devices may include: cellular or other communication devices, which have a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. Specifically, the computer device 100 may specifically be a desktop terminal or a mobile terminal, and the computer device 100 may specifically also be one of a mobile phone, a tablet computer, a laptop computer, etc.

[0040] Those skilled in the art can understand that Figure 1 the application environment shown is only one application scenario of the solution of the present application and does not constitute a limitation on the application scenario of the solution of the present application. Other application environments may also include more or fewer computer devices than Figure 1 shown. For example Figure 1 only 1 computer device is shown in . It can be understood that the sensitive information detection system may also include one or more other services, which are not specifically limited here.

[0041] In addition, as Figure 1 shown, the sensitive information detection system may also include a memory 200 for storing data, such as image data, location information, etc.

[0042] It should be noted that Figure 1 the scenario schematic diagram of the sensitive information detection system shown is only an example. The sensitive information detection system and scenario described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art know that with the evolution of the sensitive information detection system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0043] The present invention discloses a sensitive information detection method and system, which are beneficial to improving the accuracy and efficiency of sensitive information recognition and enhancing the accuracy and comprehensiveness of sensitive information recognition. The following will be described in detail respectively.

[0044] Embodiment 1

[0045] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of a sensitive information detection method disclosed in an embodiment of the present invention. Among them, Figure 2 the described sensitive information detection method is applied in a management system, such as a local server or a cloud server for management, etc., and the embodiments of the present invention do not make limitations. As Figure 2 shown, the sensitive information detection method may include the following operations:

[0046] 101. Obtain the text information to be detected.

[0047] 102. Perform detection processing on the text information to be detected to obtain target text detection information.

[0048] In the embodiments of the present invention, the target text detection information includes first target detection information, and / or second target detection information, and / or third target detection information; the first target detection information includes L first target detection word information; the second target detection information includes M second target detection word information; the third target detection information includes N third target detection word information.

[0049] 103. Perform optimization processing on the target text detection information to obtain target detection result information.

[0050] In the embodiments of the present invention, the target order is used to indicate picking the target goods out of the warehouse.

[0051] It should be noted that the above L, M, and N are positive integers not less than 0, and the embodiments of the present invention do not make limitations.

[0052] It should be noted that the sensitive information detection method of the present application can achieve efficient and accurate sensitive information detection, that is, by combining rule-based rapid screening and the deep semantic analysis ability of the deep learning model, improvements are made to the pain points in the prior art. By constructing a high-quality sensitive data set and integrating a rule engine and deep learning, the system can quickly adapt to changing contexts and emerging sensitive information types, while significantly reducing the computational resource consumption in large-scale data processing, providing an efficient and flexible solution for sensitive information detection in complex contexts, and the embodiments of the present invention do not make limitations. Further, through various forms of sensitive word detection, the high accuracy of rule recognition can be combined with the high recall rate of deep learning recognition to ensure comprehensive coverage, and the embodiments of the present invention do not make limitations.

[0053] It should be noted that the above text information to be detected can be obtained through web crawlers or obtained from text input by users, and the embodiments of the present invention do not make limitations.

[0054] It should be noted that the above target detection result information can be used to delete the text content of the text information to be detected, so as to optimize and improve the text, thereby avoiding problems such as violations caused by sensitive words in the text expression, or discovering the problem of illegal text expressions. The embodiments of the present invention do not make limitations in this regard.

[0055] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0056] In an optional embodiment, the above step of detecting and processing the text information to be detected to obtain the target text detection information includes:

[0057] In response to the user's selection operation, obtain the detection type information;

[0058] Determine whether the detection type information is the first detection type to obtain a type judgment result;

[0059] When the type judgment result is yes, perform rule-based detection processing on the text information to be detected to obtain the first target detection information and the second target detection information;

[0060] Use the target information detection model to detect and process the text information to be detected to obtain the third target detection information; the target information detection model includes the first detection model, the second detection model, and the third detection model;

[0061] When the type judgment result is no, perform rule-based detection processing on the text information to be detected to obtain the first target detection information and the second target detection information.

[0062] It should be noted that the above first detection type indicates that the target information detection model is to be used to detect and process the text information to be detected. That is, before processing the text of the sensitive word to be detected, the user's opinion should be asked first, and the user performs a selection operation. If the user chooses to use the deep model, then the text will enter the recognition link based on the target information detection model; if the user chooses not to use the target information detection model, the detection step based on the target information detection model will be skipped, and the subsequent process will be directly entered. After the above two specific detection methods in the rule-based detection branch and the target information detection model (if selected), the detection results obtained from each branch will be merged, and the sensitive information discovered by different detection methods will be summarized together to ensure comprehensive coverage of sensitive word detection.

[0063] It should be noted that the training sample set for the above target information detection model is obtained through the following method:

[0064] (1) Construct a sensitive word library: After data cleaning and classification, construct a custom sensitive word library.

[0065] (2) Construct a training sample set (labeled data set and test data set): Based on the sensitive word library combination, generate labeled data and test data respectively, where the test data should contain a certain proportion of positive and negative examples.

[0066] Furthermore, constructing the sensitive word library generally consists of the following steps:

[0067] (a) Data collection and cleaning: Use the collected sensitive words for data cleaning, remove duplicate, invalid and noisy data to ensure the accuracy and integrity of the data.

[0068] (b) Sensitive word classification: Classify the sensitive words to ensure that each category has a clear definition and standard, which is used to construct the word library of sensitive words and generate sentences containing sensitive words.

[0069] (c) Sentence creation and evaluation: Create sentences according to each classification, requiring that the sentences contain sensitive words of multiple classifications and are used as training and test data. This helps to improve the model's recognition ability for different contexts and expressions.

[0070] Furthermore, the process of the target information detection model is as follows:

[0071] 1. Training data preparation

[0072] (A) It is necessary to prepare a labeled JSON data file, such as output.json, and this file should be organized in the following format:

[0073]

[0074]

[0075] text: Represents the text to be labeled.

[0076] label: Labeling information, which contains the index range of the category and the corresponding entity.

[0077] 2. Training data preprocessing

[0078] (B) Expand the labeled data

[0079] Use a script to preprocess the JSON data and convert it into a flattened format for subsequent processing. After the script is executed, a corresponding train.json file will be generated.

[0080] (C) Prepare the data directory

[0081] Create two subfolders in the data directory, ori_data (for storing the original data file train.json) and ner_data (for storing the preprocessed training file)

[0082] (D) Generate training data

[0083] Executing the data processing script will generate the files required for model training, dev.txt, labels.txt, and train.txt, from train.json in the ori_data folder.

[0084] 3. Training

[0085] (E) Create a directory to save the model (such as checkpoint / sensitive).

[0086] (F) Modify the dataset path and name in the training script.

[0087] (G) Execute the training script to start model training. After training is completed, the model file will be saved in the checkpoint directory.

[0088] It should be noted that the above rule-based detection and processing of the text information to be detected is a rule-based rapid screening mechanism, and the design of a multi-level rule engine for rule-based detection and processing supports the customization and configuration of rules and sensitive word libraries, which is not limited in the embodiments of the present invention. Furthermore, the rule-based detection and processing has a clear recognition capability for personal sensitive information with fixed formats or obvious characteristics and sensitive word libraries, and has a low false alarm rate, which is not limited in the embodiments of the present invention.

[0089] It should be noted that the above-mentioned detection and processing of the text information to be detected using the target information detection model can handle implicit expressions with complex semantics and difficult-to-define rules, and identify potential sensitive information, which is not limited in the embodiments of the present invention.

[0090] It can be seen that implementing the sensitive information detection method described in the embodiment of the present invention is conducive to improving the accuracy and efficiency of sensitive information identification, and improving the accuracy and comprehensiveness of sensitive information identification.

[0091] In another optional embodiment, performing rule-based detection processing on the text information to be detected to obtain first target detection information and second target detection information includes:

[0092] Using regular expressions to perform detection processing on the text information to be detected, to obtain first target detection information;

[0093] The text information to be detected is subjected to sensitive information matching detection processing to obtain second target detection information.

[0094] It should be noted that the above detection and processing of the text information to be detected using regular expressions are based on rich, diverse, and accurate regular expression configurations, which can quickly locate and mark various sensitive information with extremely high accuracy, effectively ensuring information security and compliance, meeting the stringent requirements for sensitive information recognition in various scenarios, and providing a comprehensive, reliable, and efficient sensitive information protection barrier. This is not limited in the embodiments of the present invention. Further, the regular expressions for characterizing and matching various types of sensitive information that may be involved are set in the following manner:

[0095] Regular expression configuration: Configure through a yaml format file, and it also supports direct modification in the system. The style of the configuration file is as follows:

[0096]

[0097]

[0098] Modify regular expressions: Modify in the system using the json style.

[0099] Add regular expressions: When adding rules in the system, just input the corresponding content, and the system will automatically convert it to the yaml format and write it into the configuration file, simplifying the addition process and reducing the operation difficulty.

[0100] Add a sensitive word library: Select the function of importing a sensitive word library to add a sensitive word library. It supports txt format files, and each line of content corresponds to a sensitive word. When importing, the word library will be imported into the rules of the corresponding category. If the category does not exist, it will be automatically created; if the category already exists, it will be merged with the existing category after removing duplicates to ensure the accuracy and integrity of the word library. In addition, the system also supports editing operations on the word library.

[0101] It should be noted that the above matching and detection processing of sensitive information for the text information to be detected quickly filters out a large amount of non-sensitive information through the vectors corresponding to the preset sensitive words, greatly improving the detection speed of sensitive words, and solving the problem of the lack of fast preprocessing means in the prior art. This is not limited in the embodiments of the present invention.

[0102] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0103] In another optional embodiment, the matching and detection processing of sensitive information for the text information to be detected to obtain the second target detection information includes:

[0104] Perform word segmentation processing on the text information to be detected to obtain the first processed text information; the first processed text information includes several first text word information;

[0105] For any first text word information in the first processed text information, use an association calculation model to calculate and process the sensitive word vector information in the first text word information and the sensitive word symbol information, and obtain the word symbol association value information corresponding to the first text word information; the word symbol association value information includes a plurality of word symbol association values;

[0106] Among them, the association calculation model is:

[0107]

[0108] Among them, CFGLZ represents the word symbol association value; WBC and YFC respectively represent the first text word information and the sensitive word vector information; qz1 and qz2 respectively represent the first weight coefficient and the second weight information;

[0109] Judge whether there is a word symbol association value greater than or equal to the association threshold in the word symbol association value information, and obtain the association value judgment result;

[0110] When the association value judgment result is yes, determine the first text word information as a second target detection word information;

[0111] When the association value judgment result is no, end the detection process corresponding to the first text word information.

[0112] It should be noted that the above first weight coefficient and second weight information can be set by the user or preset by the system, and their values are numerical values between 0 and 1, which are not limited in the embodiments of the present invention.

[0113] It should be noted that the sensitive word vector information in the above sensitive word symbol information represents the data after vectorization of the sensitive words preset in the sensitive word library of the system, and its vectorization can be based on the LSTM model or a large model, which is not limited in the embodiments of the present invention.

[0114] It should be noted that the above association threshold can be set by the user or preset by the system, and its value is a numerical value between 0.5 and 2, which is not limited in the embodiments of the present invention.

[0115] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and improving the accuracy and comprehensiveness of sensitive information recognition.

[0116] In another optional embodiment, use a target information detection model to detect and process the text information to be detected, and obtain third target detection information, including:

[0117] Perform text preprocessing on the text information to be detected to obtain first text processing information;

[0118] Perform bidirectional encoding processing on the first text processing information using the first detection model to obtain the second text processing information;

[0119] Perform bidirectional feature extraction processing on the second text processing information using the second detection model to obtain the third text processing information; the second detection model includes a first detection sub-model, a second detection sub-model, and a third detection sub-model;

[0120] Perform detection and recognition processing on the third text processing information using the third detection model to obtain the third target detection information.

[0121] It should be noted that the above text preprocessing of the text information to be detected includes data cleaning and word segmentation processing, etc., which are not limited in the embodiments of the present invention.

[0122] It should be noted that the above bidirectional encoding processing of the first text processing information using the first detection model is through context-aware embedding of sensitive word texts, understanding the subtle differences of words in the context, capturing the complex semantic relationships in the text, solving the deficiencies in the prior art when dealing with polysemy and complex contexts, and improving the detection accuracy. Its implementation can be based on a large model or can be implemented based on models such as BERT, which are not limited in the embodiments of the present invention. Further, the first detection model mainly undertakes the task of in-depth semantic understanding and feature extraction of sensitive word texts, deeply mining the semantic representations of words and sentences. Through bidirectional encoding of the input sensitive word texts, fully considering the context information before and after the words, high-quality word vectors containing multi-dimensional information such as semantics and grammar are generated. These word vectors provide a strong semantic feature basis for subsequent sensitive information judgment, enabling the target information detection model to more accurately understand the semantic content of sensitive word texts, thereby effectively distinguishing normal texts from texts containing sensitive information. For example, when identifying sensitive texts involving implicit semantic expressions, the first detection model can accurately capture the potential sensitive intentions therein by virtue of its in-depth semantic understanding ability, greatly improving the accuracy and recall rate of sensitive information recognition and enhancing the overall recognition effect of the system, which are not limited in the embodiments of the present invention.

[0123] It should be noted that the above-mentioned two-way feature extraction and processing of the second text processing information by the second detection model is achieved by capturing the long-distance dependence relationship in the sensitive word text, processing the information flow in the long text, enhancing the context understanding ability, and solving the problems that cannot be compared by the unidirectional LSTM model in the prior art. The embodiments of the present invention are not limited thereto. Further, when the second detection model processes the sensitive word text sequence in sensitive information recognition, during the forward processing, each word is processed in sequence according to the natural order of the sensitive word text to accurately capture the information flow from the beginning to the end; during the reverse processing, the words are processed from the end to the beginning to obtain the reverse information. By integrating the information in both the forward and reverse directions, the second detection model can generate a comprehensive feature representation containing rich context information for each word. This feature representation is extremely crucial in the sequence labeling task of this application because sensitive information recognition often requires considering the order of the text sequence and the context dependence relationship. For example, when identifying a text containing sensitive vocabulary but whose sensitivity depends on the context, the feature representation generated by the second detection model can provide strong support for accurate judgment, effectively reducing misjudgment or missed judgment caused by insufficient context understanding, and improving the stability and reliability of the system's recognition of sensitive information in complex text sequences. The embodiments of the present invention are not limited thereto.

[0124] It should be noted that the above-mentioned detection and recognition processing of the third text processing information by the third detection model takes into account the correlation between sensitive word tags. Through the method of sequence labeling, it ensures that the dependence relationship between sensitive word tags is effectively mapped, improves the accuracy of sensitive word detection, and solves the functions that common independent classifiers in the prior art do not have. The embodiments of the present invention are not limited thereto. Further, in the sequence labeling link of sensitive information recognition, each input sequence element needs to be assigned an accurate label, and there are usually specific constraint relationships between these labels. The third detection model can learn the transition probabilities between these labels. By constructing a probability model, it calculates the conditional probabilities of all possible label sequences, and then determines the label sequence with the highest probability as the output. The third detection model can also use the constraints between labels to correct the errors in single label prediction. For example, in the face of some text fragments that are prone to ambiguity, individual models may give incorrect label predictions, but the third detection model can correct them according to the constraint relationship between labels, making the final label sequence more in line with the language rules and the requirements of the sensitive information recognition task, thereby effectively improving the accuracy of the sequence labeling task, ensuring that the entire sensitive information recognition system can output more accurate and reliable recognition results when processing text sequences, reducing the risks of false alarms and missed detections, and improving the overall performance and credibility of the system. The embodiments of the present invention are not limited thereto.

[0125] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0126] In an alternative embodiment, a second detection model is used to perform bidirectional feature extraction processing on second text processing information to obtain third text processing information, including:

[0127] Using a first detection sub-model to perform bidirectional semantic feature extraction on the second text processing information to obtain first sub-text processing information; the first sub-text processing information represents the bidirectional semantic features of forward propagation and backward propagation in the text;

[0128] Using a second detection sub-model to perform forward and backward data splicing processing on the first sub-text processing information to obtain second sub-text processing information;

[0129] Using a third detection sub-model to perform sequence annotation processing on the second sub-text processing information to obtain third text processing information.

[0130] It should be noted that the above-mentioned use of the first detection sub-model to perform bidirectional semantic feature extraction on the second text processing information may be to use a bidirectional text information capture unit module (such as constructed based on a convolutional layer or a neural network) to capture the word vector information by unidirectionally transmitting information respectively to obtain the hidden layer state, so as to realize the extraction of forward and backward semantic features. The embodiments of the present invention are not limited thereto.

[0131] It should be noted that the above-mentioned use of the second detection sub-model (such as a unit module constructed based on a splicing operation) to perform forward and backward data splicing processing on the first sub-text processing information may be to use the second detection sub-model to perform forward and backward splicing of context information to realize a complete and rich representation of the forward and backward information of the text sequence. The embodiments of the present invention are not limited thereto.

[0132] It should be noted that the above-mentioned use of the third detection sub-model (such as constructed based on a softmax activation function) to perform sequence annotation processing on the second sub-text processing information is to project the second sub-text processing information into a numerical interval of text labels with clear meaning expressions, and then identify whether the text word is a sensitive word. The embodiments of the present invention are not limited thereto.

[0133] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0134] In another alternative embodiment, optimizing the target text detection information to obtain target detection result information, including:

[0135] Perform a union operation on the target text detection information to obtain optimized text detection information;

[0136] Detect whether text modification information input by the user is received within the optimized time range to obtain detection result information;

[0137] When the detection result information is yes, exclude the intersection information between the optimized text detection information and the detection result information from the optimized text detection information to obtain the target detection result information;

[0138] When the detection result information is no, determine the optimized text detection information as the target detection result information.

[0139] It should be noted that the above union operation on the target text detection information is to take the union of the sensitive word information detected by rule-based sensitive word detection and the target information detection model to ensure that no potential sensitive information is missed. The embodiments of the present invention do not make any limitations.

[0140] It should be noted that the above optimized time range is the time within 1 time unit after obtaining the optimized text detection information. Further, the above time unit can be one of 3 minutes, 5 minutes, or 10 minutes. The embodiments of the present invention do not make any limitations. Further, the time space given by 1 optimized time range, that is, it leaves enough time for the user to modify and check, and there will be no long waiting time, thus ensuring both the further optimization accuracy of sensitive words and the detection efficiency. The embodiments of the present invention do not make any limitations.

[0141] It should be noted that the above text modification information represents the non-sensitive words selected by the user, that is, the text words that need to be excluded from the target text detection information, so as to further accurately optimize the text representation acceptable to the user and avoid misjudgment of sensitive words caused by overly strict sensitive word detection, and further improve the detection accuracy and efficiency of sensitive words. The embodiments of the present invention do not make any limitations.

[0142] It can be seen that implementing the sensitive information detection method described in the embodiments of the present invention is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0143] Embodiment 2

[0144] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a sensitive information detection system disclosed in the embodiments of the present invention. Among them, Figure 3 the described system can be applied to a management system, such as a local server or a cloud server for management, etc. The embodiments of the present invention do not make any limitations. As Figure 3 shown, the system may include:

[0145] An acquisition module 201, configured to acquire text information to be detected;

[0146] A second processing module 202, configured to perform detection processing on the text information to be detected to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L pieces of first target detection word information; the second target detection information includes M pieces of second target detection word information; the third target detection information includes N pieces of third target detection word information;

[0147] A third processing module 203, configured to perform optimization processing on the target text detection information to obtain target detection result information.

[0148] It can be seen that implementing Figure 3 the described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0149] In another alternative embodiment, as Figure 3 shown, performing detection processing on the text information to be detected to obtain target text detection information includes:

[0150] Responding to a user's selection operation to obtain detection type information;

[0151] Judging whether the detection type information is a first detection type to obtain a type judgment result;

[0152] When the type judgment result is yes, performing rule-based detection processing on the text information to be detected to obtain first target detection information and second target detection information;

[0153] Using a target information detection model to perform detection processing on the text information to be detected to obtain third target detection information; the target information detection model includes a first detection model, a second detection model, and a third detection model;

[0154] When the type judgment result is no, performing rule-based detection processing on the text information to be detected to obtain first target detection information and second target detection information.

[0155] It can be seen that implementing Figure 3 the described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0156] In yet another alternative embodiment, as Figure 3 shown, performing rule-based detection processing on the text information to be detected to obtain first target detection information and second target detection information includes:

[0157] The text information to be detected is detected and processed using regular expressions to obtain the first target detection information;

[0158] The text information to be detected is subjected to sensitive information matching detection and processing to obtain the second target detection information.

[0159] It can be seen that implementing Figure 3 the described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0160] In another alternative embodiment, as Figure 3 shown, the text information to be detected is subjected to sensitive information matching detection and processing to obtain the second target detection information, including:

[0161] The text information to be detected is segmented to obtain the first processed text information; the first processed text information includes a number of first text word information;

[0162] For any first text word information in the first processed text information, the association calculation model is used to calculate and process the first text word information and the sensitive word vector information in the sensitive word character information to obtain the word character association value information corresponding to the first text word information; the word character association value information includes a number of word character association values;

[0163] Among them, the association calculation model is:

[0164]

[0165] Among them, CFGLZ represents the word character association value; WBC and YFC respectively represent the first text word information and the sensitive word vector information; qz1 and qz2 respectively represent the first weight coefficient and the second weight information;

[0166] It is determined whether there is a word character association value greater than or equal to the association threshold in the word character association value information to obtain the association value judgment result;

[0167] When the association value judgment result is yes, the first text word information is determined as a second target detection word information;

[0168] When the association value judgment result is no, the detection process corresponding to the first text word information is ended.

[0169] It can be seen that implementing Figure 3 the described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0170] In another alternative embodiment, as Figure 3As shown, the target information detection model is used to detect and process the text information to be detected, and the third target detection information is obtained, including:

[0171] Perform text preprocessing on the text information to be detected to obtain the first text processing information;

[0172] Use the first detection model to perform bidirectional encoding processing on the first text processing information to obtain the second text processing information;

[0173] Use the second detection model to perform bidirectional feature extraction processing on the second text processing information to obtain the third text processing information; The second detection model includes a first detection sub-model, a second detection sub-model, and a third detection sub-model;

[0174] Use the third detection model to perform detection and recognition processing on the third text processing information to obtain the third target detection information.

[0175] It can be seen that implementing Figure 3 The described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and improving the accuracy and comprehensiveness of sensitive information recognition.

[0176] In another alternative embodiment, as Figure 3 shown, use the second detection model to perform bidirectional feature extraction processing on the second text processing information to obtain the third text processing information, including:

[0177] Use the first detection sub-model to perform bidirectional semantic feature extraction on the second text processing information to obtain the first sub-text processing information; The first sub-text processing information represents the bidirectional semantic features of forward propagation and backward propagation in the text;

[0178] Use the second detection sub-model to perform front-back bidirectional data splicing processing on the first sub-text processing information to obtain the second sub-text processing information;

[0179] Use the third detection sub-model to perform sequence annotation processing on the second sub-text processing information to obtain the third text processing information.

[0180] It can be seen that implementing Figure 3 The described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and improving the accuracy and comprehensiveness of sensitive information recognition.

[0181] In another alternative embodiment, as Figure 3 shown, perform optimization processing on the target text detection information to obtain the target detection result information, including:

[0182] Perform union processing on the target text detection information to obtain the optimized text detection information;

[0183] Detect whether text modification information input by the user is received within the optimized time range to obtain detection result information;

[0184] When the detection result information is yes, remove the intersection information between the optimized text detection information and the detection result information from the optimized text detection information to obtain target detection result information;

[0185] When the detection result information is no, determine the optimized text detection information as the target detection result information.

[0186] It can be seen that implementing Figure 3 the described sensitive information detection system is beneficial to improving the accuracy and efficiency of sensitive information recognition, and enhancing the accuracy and comprehensiveness of sensitive information recognition.

[0187] Embodiment III

[0188] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of another sensitive information detection system disclosed in an embodiment of the present invention. Among them, Figure 4 the described system can be applied to a management system, such as a local server or a cloud server for management, etc., which is not limited in the embodiments of the present invention. As Figure 4 shown, the system may include:

[0189] A memory 301 storing executable program code;

[0190] A processor 302 coupled to the memory 301;

[0191] The processor 302 invokes the executable program code stored in the memory 301 to execute the steps in the sensitive information detection method described in Embodiment I.

[0192] Embodiment IV

[0193] An embodiment of the present invention discloses a computer-readable storage medium that stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps in the sensitive information detection method described in Embodiment I.

[0194] Embodiment V

[0195] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the sensitive information detection method described in Embodiment I.

[0196] The system embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0197] Through the specific descriptions of the above embodiments, those skilled in the art can clearly understand that each implementation can be achieved by means of software plus a necessary general hardware platform, and of course, it can also be achieved by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium that can be used to carry or store data.

[0198] Finally, it should be noted that: what is disclosed in a sensitive information detection method and system disclosed in an embodiment of the present invention is only a preferred embodiment of the present invention, only for explaining the technical solution of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; 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 present invention.

Claims

1. A sensitive information detection method, characterized in that: The method comprises: Get the text information to be detected; Performing detection processing on the text information to be detected to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L first target detection word information; the second target detection information includes M second target detection word information; the third target detection information includes N third target detection word information; The target text detection information is optimized to obtain target detection result information.

2. The sensitive information detection method according to claim 1, characterized in that: The detecting and processing the text information to be detected to obtain target text detection information includes: In response to a selection operation by a user, obtaining detection type information; Determine whether the detection type information is a first detection type, and obtain a type determination result; When the type judgment result is yes, performing rule-based detection processing on the text information to be detected to obtain the first target detection information and the second target detection information; Using a target information detection model to detect the text information to be detected to obtain the third target detection information; the target information detection model includes a first detection model, a second detection model and a third detection model; When the type judgment result is no, the text information to be detected is subjected to rule-based detection processing to obtain the first target detection information and the second target detection information.

3. The sensitive information detection method according to claim 2, characterized in that: The performing rule-based detection processing on the text information to be detected to obtain the first target detection information and the second target detection information includes: Performing detection processing on the text information to be detected using a regular expression to obtain the first target detection information; The text information to be detected is subjected to sensitive information matching detection processing to obtain the second target detection information.

4. The sensitive information detection method according to claim 3, characterized in that: The performing sensitive information matching detection processing on the text information to be detected to obtain the second target detection information includes: Performing word segmentation processing on the text information to be detected to obtain first processed text information; the first processed text information includes a plurality of first text word information; For any of the first text word information in the first processed text information, use the association calculation model to calculate and process the first text word information and the sensitive word vector information in the sensitive word information to obtain word association value information corresponding to the first text word information; the word association value information includes a plurality of word association values; Wherein, the association calculation model is: Among them, CFGLZ represents the word token association value; WBC and YFC represent the first text word information and the sensitive word token vector information respectively; qz1 and qz2 represent the first weight coefficient and the second weight information respectively; Determine whether there is a word-token association value greater than or equal to an association threshold in the word-token association value information, and obtain an association value determination result; When the association value judgment result is yes, determining the first text word information as the second target detection word information; When the result of the association value judgment is no, the detection process corresponding to the first text word information is terminated.

5. The sensitive information detection method according to claim 2, characterized in that: The detecting and processing the to-be-detected text information by using the target information detection model to obtain the third target detection information includes: Performing text preprocessing on the text information to be detected to obtain first text processing information; Using the first detection model to perform bidirectional encoding processing on the first text processing information to obtain second text processing information; Using the second detection model to perform bidirectional feature extraction processing on the second text processing information to obtain third text processing information; the second detection model includes a first detection sub-model, a second detection sub-model and a third detection sub-model; The third detection model is used to perform detection and recognition processing on the third text processing information to obtain the third target detection information.

6. The sensitive information detection method according to claim 5, characterized in that: The using the second detection model to perform bidirectional feature extraction processing on the second text processing information to obtain third text processing information includes: Using the first detection sub-model to perform bidirectional semantic feature extraction on the second text processing information to obtain first sub-text processing information; the first sub-text processing information represents the bidirectional semantic features of forward propagation and backward propagation in the text; Using the second detection sub-model to perform forward and backward bidirectional data splicing processing on the first sub-text processing information to obtain second sub-text processing information; The second sub-text processing information is sequence labeled using the third detection sub-model to obtain third text processing information.

7. The sensitive information detection method according to claim 1, characterized in that: The optimizing process of the target text detection information to obtain target detection result information includes: Performing a union process on the target text detection information to obtain optimized text detection information; Detect whether text modification information input by the user is received within the optimized time range, and obtain detection result information; When the detection result information is yes, the intersection information of the optimized text detection information and the detection result information is removed from the optimized text detection information to obtain the target detection result information; When the detection result information is negative, the optimized text detection information is determined as the target detection result information.

8. A sensitive information detection system, characterized in that: The system comprises: An acquisition module is used to acquire text information to be detected; A second processing module is used to perform detection processing on the text information to be detected to obtain target text detection information; the target text detection information includes first target detection information, and / or, second target detection information, and / or, third target detection information; the first target detection information includes L first target detection word information; the second target detection information includes M second target detection word information; the third target detection information includes N third target detection word information; The third processing module is used to optimize the target text detection information to obtain target detection result information.

9. A sensitive information detection system, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the sensitive information detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which, when called, are used to execute the sensitive information detection method as described in any one of claims 1-7.

Images