Document convention risk identification method, device and equipment and storage medium

Abstract

The invention relates to the technical field of image processing and the technical field of artificial intelligence, and discloses a document consignment risk identification method, device and equipment and a storage medium, and the method comprises the steps: obtaining a consignment text and a document image of insurance underwriting; performing structured decomposition on the special text to obtain target text information in the special text; performing calibration identification on the document image to obtain target document information in the document image; determining a special risk level according to the target text information and the target document information to obtain a risk level result; and when the risk level result meets the alarm condition, carrying out alarm notification on insurance underwriting. According to the invention, the identification accuracy and the identification efficiency of the document special risk can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the fields of image processing technology, artificial intelligence technology and financial technology, and in particular to a document-specific risk identification method, device, electronic device and computer-readable storage medium. Background Art

[0002] Identification and classification of special risks in documents is often a very important issue in the insurance industry. The insurance policies and special terms in the insurance industry are complex and diverse, and fixed special terms are difficult to meet the personalized needs of customers. When documents allow the entry of free editable special terms, there may be a risk of inconsistency between the special terms and the policy content in terms of customer information, subject information, terms information and liability information. The consistency of special terms and policy content is extremely important to both customers and insurance companies. When a claim occurs later, it can reduce the communication costs of both parties. Therefore, it is necessary to strengthen the review during the issuance of the policy to ensure that the special terms and policy content are consistent.

[0003] The existing document review process is highly dependent on manual labor. Although it can handle complex semantics and special situations, the special clauses are very long and manual review is prone to errors. In addition, as the number of insurance policies increases, the review time and cost also increase. Therefore, how to improve the recognition accuracy and efficiency of special risk in documents has become an urgent problem to be solved. Summary of the invention

[0004] The present invention provides a method, device and computer-readable storage medium for identifying document-specific risks, the main purpose of which is to solve the problems of insufficient accuracy and low efficiency in identifying document-specific risks.

[0005] In a first aspect, to achieve the above-mentioned purpose, the present invention provides a method for identifying document-specific risks, comprising:

[0006] Obtain insurance coverage contract text and document images;

[0007] Performing structural decomposition on the special text to obtain target text information in the special text;

[0008] Calibrate and identify the document image to obtain target document information in the document image;

[0009] Determine the special risk level according to the target text information and the target document information to obtain a risk level result;

[0010] When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting.

[0011] In a second aspect, the present invention further provides a document-specific risk identification device, the device comprising:

[0012] The acquisition module is used to obtain the special text and document images of insurance underwriting;

[0013] A structured decomposition module, used for performing structured decomposition on the special text to obtain target text information in the special text;

[0014] A calibration and recognition module, used to calibrate and recognize the document image to obtain target document information in the document image;

[0015] A risk level module, used to determine the special risk level according to the target text information and the target document information, and obtain a risk level result;

[0016] The alarm module is used to issue an alarm notification for insurance underwriting when the risk level result meets the alarm condition.

[0017] In a third aspect, the present invention further provides an electronic device, the electronic device comprising:

[0018] at least one processor; and,

[0019] a memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned method for identifying document-specific risks.

[0021] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein at least one computer program is stored in the computer-readable storage medium, and the at least one computer program is executed by a processor in an electronic device to implement the above-mentioned method for identifying document-specific risks.

[0022] In the embodiment of the present invention, the special text is obtained by querying the official website, which ensures the timeliness and authority of the special text, and the document image is obtained by intelligent shooting tools, etc., which ensures the high quality of the obtained document image, which is convenient for the subsequent recognition and processing of the document image; the special text is structurally decomposed and the document image is calibrated and recognized, which can quickly process a large number of insurance policies and special clauses texts, significantly reduce manual review time and cost, and greatly improve the overall recognition efficiency; the target text information and the target document information are compared through an intelligent rule engine to determine the special risk level, reduce errors and omissions in human review, ensure the consistency and compliance of documents and special clauses, and improve the review efficiency; according to the risk level result, an alarm notification is issued to the insurance underwriting, so as to realize timely and accurate identification and alarm of risks in the insurance underwriting process, and improve the recognition efficiency and review efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0024] Figure 1 It is a schematic diagram of an application environment of a document-specific risk identification method in one embodiment of the present invention;

[0025] Figure 2 A schematic diagram of a process flow of a document-specific risk identification method provided in one embodiment of the present invention;

[0026] Figure 3 A schematic diagram of a module of a document-specific risk identification device provided by an embodiment of the present invention;

[0027] Figure 4 A schematic diagram of the structure of an electronic device of a document-specific risk identification device provided by an embodiment of the present invention;

[0028] Figure 5 It is another structural schematic diagram of an electronic device of a document-specific risk identification device provided by an embodiment of the present invention.

[0029] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, and to fully understand and implement how the present disclosure applies technical means to solve technical problems and achieve the corresponding technical effects, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only embodiments of a part of the present disclosure, not all of the embodiments. The embodiments of the present disclosure and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0032] The embodiment of the present application provides a method for identifying risks of special documents. The execution subject of the method for identifying risks of special documents includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the device provided by the embodiment of the present application. In other words, the method for identifying risks of special documents can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0033] The document-specific risk identification method provided by the embodiment of the present invention can be applied in the following aspects: Figure 1In an application environment, the client communicates with the server through a network. The server can obtain the special contract text and the document image of the insurance underwriting through the client, perform structural decomposition on the special contract text to obtain the target text information in the special contract text, perform calibration and recognition on the document image to obtain the target document information in the document image, determine the special contract risk level according to the target text information and the target document information, obtain the risk level result, and when the risk level result meets the alarm condition, issue an alarm notification for the insurance underwriting, and feed the risk level result back to the client. In the present invention, for the identification of special risks of documents, in the embodiment of the present invention, the special text is obtained by querying the official website, which ensures the timeliness and authority of the special text, and the document image is obtained by intelligent shooting tools, etc., which ensures the high quality of the obtained document image, which is convenient for the subsequent identification and processing of the document image; the special text is structurally decomposed and the document image is calibrated and identified, which can quickly process a large number of insurance policies and special clauses, significantly reduce the time and cost of manual review, and greatly improve the overall recognition efficiency; the target text information and the target document information are compared by the intelligent rule engine, so as to determine the special risk level, reduce the errors and omissions of manual review, ensure the consistency and compliance of the document and the special clause, and improve the review efficiency; according to the risk level result, the insurance underwriting is warned and notified, so as to realize the timely and accurate identification and warning of the risks in the insurance underwriting process, and improve the recognition efficiency and review efficiency. Among them, the client can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0034] Reference Figure 2 The figure is a flow chart of a method for identifying document-specific risks provided by one embodiment of the present invention.

[0035] In this embodiment, the document-specific risk identification method includes:

[0036] S1. Obtain the special contract text and document image of the insurance underwriting.

[0037] In the embodiment of the present invention, the insurance underwriting refers to the process by which the insurance company reviews, verifies and selects risks for insurance applications submitted by units or individuals willing to purchase insurance, and makes a decision on whether to accept and how to accept the insurance; the insurance underwriting needs to properly handle the pressure from the government's compulsory insurance, the contradiction between sales promotion and underwriting, and handle adverse risks. For excessively adverse risks, insurance should be refused. For risks higher than general, the premium rate should be increased or improvement suggestions should be made to the insured so that they can be covered after improvement.

[0038] In an embodiment of the present invention, the special text of insurance underwriting refers to the laws and regulations that the insured needs to use in insurance underwriting, that is, a text specially agreed upon to fulfill special obligations; the special text includes the insured's customer information, subject information, clause name and liability information.

[0039] In an embodiment of the present invention, the insurance underwriting document image refers to the certification material required by the insured in the insurance underwriting. The insurance document is the insurance company's proof of commitment to the insured, and is also a contract between the insurance company and the insured, which specifically stipulates the rights and obligations of both parties; when the insured goods suffer losses, the insurance document is the basis for the insured's claim and is also the main basis for the insurance company to settle claims; the document image includes the insured's customer information, subject information, clause name and liability information.

[0040] Furthermore, during the insurance underwriting process, the insurance underwriting should be carried out in a timely manner to ensure that the insured can obtain compensation in a timely manner and greatly improve the quotation timeliness. It should also follow the principles of fairness, equity and openness to ensure the objectivity and accuracy of the underwriting results. All underwriting activities must be carried out in accordance with the law and must not violate the provisions of relevant laws and regulations.

[0041] The special contract text of the present invention can be connected to an official website such as the official website of the insurance company where the insured is located through the HTTP protocol through programming languages ​​such as Python and Java, and an HTTP request is sent to obtain data, thereby obtaining the special contract text of the insurance underwriting.

[0042] The present invention can use intelligent shooting tools to shoot the documents required for insurance underwriting, such as using a mobile phone to shoot the insured's documents and materials, such as insurance policies, insurance certificates, etc.

[0043] Furthermore, a scanner may be used to scan the document into an electronic image, and the electronic image in the scanner may be accessed through a computer platform to obtain the document image.

[0044] In the embodiment of the present invention, the special text is obtained by querying the official website, which ensures the timeliness and authority of the special text, and the document image is obtained by intelligent shooting tools, which ensures the high quality of the obtained document image, which facilitates the subsequent recognition and processing of the document image.

[0045] S2. Structurally decompose the special text to obtain target text information in the special text.

[0046] In the embodiments of the present invention, the structured decomposition refers to converting unstructured special text into a structured data format, and decomposing the special text into multiple structured texts according to certain rules or algorithms, so as to more easily extract, analyze and use the special text information; the target text information includes customer information (name of the applicant, ID number of the applicant, name of the insured, ID number of the insured, name of the beneficiary, ID number of the beneficiary), subject information (subject type, subject location), clause name and liability information in the special text.

[0047] In the embodiments of the present invention, the structured decomposition of the special text to obtain the target text information in the special text includes:

[0048] Cleaning and standardizing the special text to obtain a standardized special text;

[0049] Using a preset text classification algorithm to classify the standardized special text to obtain the target text information in the special text.

[0050] Specifically, the text cleaning of the special text includes removing irrelevant characters, handling spelling mistakes, removing stop words, etc. Removing irrelevant characters means deleting special characters, punctuation marks, and numbers in the special text to remove noise; handling spelling mistakes means correcting spelling mistakes in the special text to ensure the accuracy of the special text; removing stop words refers to words that frequently appear in the special text but contribute little to the meaning of the text, such as "de", "le", etc. Removing stop words can reduce the size of the special text.

[0051] Furthermore, the standardization processing of the special text is vocabulary standardization and special text segmentation. By replacing synonyms or near-synonyms with unified vocabulary, the complexity brought by vocabulary diversity can be reduced; the special text segmentation refers to splitting long texts into smaller paragraphs or sentences for subsequent processing.

[0052] The text classification algorithm of the present invention can be a support vector machine algorithm. The basic idea of the support vector machine algorithm is to find an optimal hyperplane in the feature space to separate samples of different categories. This hyperplane is determined by support vectors, which are the sample points closest to the hyperplane. The support vector machine attempts to find a hyperplane such that sample points of different categories are as far away from the hyperplane as possible, while sample points of the same category are as close to it as possible.

[0053] In the embodiments of the present invention, the use of the text classification algorithm to classify the standardized special text to obtain the target text information in the special text includes:

[0054] Performing text vectorization on the standardized special text to obtain a numerical vector text corresponding to the standardized special text;

[0055] Using a preset support vector machine algorithm to divide the numerical vector text into a plurality of classified special text blocks;

[0056] Preset keywords are extracted based on the classified special text blocks to obtain target text information in the standardized special text.

[0057] In an embodiment of the present invention, the keywords include customer information in the special contract text, namely, the name of the insured, the insured's ID number, the name of the insured, the insured's ID number, the name of the beneficiary, and the beneficiary's ID number; the subject information, namely, the type of subject, the location of the subject; the name of the clause and liability information, etc.

[0058] In detail, text vectorization technology such as word embedding is used to represent the text as a continuous numerical vector, which can capture the semantic relationship between texts, thereby obtaining numerical vector text so that the computer can process it; the numerical vector text is classified using a preset support vector machine algorithm to obtain the category to which the text belongs, and the text is divided into corresponding classified special text blocks, where the "text block" can be a single text item or a collection of multiple similar texts; preset keywords are extracted from the classified standardized special text through keyword extraction technology such as frequency-inverse document frequency technology to ensure that the extracted keyword information is the target text information.

[0059] In an embodiment of the present invention, the special text is structured and decomposed, so that the special text information is more orderly and easy to understand. In particular, when processing a large amount of text data, the processing time can be greatly shortened, and misunderstandings or omissions caused by manual reading and understanding of the text can be avoided, thereby ensuring the accuracy and reliability of the extracted target text information, improving the utilization rate of the text information, and providing support for subsequent processing or analysis.

[0060] S3. Calibrate and identify the document image to obtain target document information in the document image.

[0061] In an embodiment of the present invention, the calibration and recognition includes calibration processing and optical character recognition of the document image, wherein the calibration processing includes grayscale and normalization processing to improve the accuracy of subsequent recognition; the optical character recognition refers to a process in which an electronic device (such as a scanner or a digital camera) checks characters printed on paper, determines their shapes by detecting dark and light patterns, and then translates the shapes into computer text using a character recognition method; for example, for printed characters, an optical method is used to convert the text in a paper document into a black and white dot matrix image file, and the text in the image is converted into a text format through recognition software for further editing and processing by a word processing software.

[0062] In an embodiment of the present invention, the target document information includes customer information, subject information, clause information and liability information; wherein the customer information includes the name of the policyholder, the policyholder ID number, the name of the insured, the insured ID number, the name of the beneficiary, and the beneficiary ID number; the subject information includes the type of subject and the location of subject; the clause name includes the name of the main insurance clause and the name of the supplementary insurance clause.

[0063] In the embodiment of the present invention, the step of calibrating and identifying the document image to obtain target document information in the document image includes:

[0064] Gray-scaling and normalizing the document image to obtain a calibration document;

[0065] Using a text detection algorithm to locate text on the calibration document to obtain a text area in the calibration document;

[0066] Optical character recognition is performed on the text area to obtain target document information in the document image.

[0067] In the embodiment of the present invention, optical character recognition refers to identifying the document content related to insurance underwriting in the calibration document. In the insurance field, it refers to identifying the customer information, target information, clause information and liability information in the calibration document.

[0068] In detail, the grayscale processing is the process of converting a color image into a grayscale image. In a color image, each pixel is composed of three color channels: red, green, and blue, while each pixel in a grayscale image has only one brightness value, and the brightness value is usually obtained by calculating the average or weighted average of the three color channels of red, green, and blue in the color image. The grayscale processing can reduce the amount of image data, simplify subsequent processing steps, and retain sufficient image information for subsequent processing.

[0069] Furthermore, normalization is the process of converting image data into values ​​within a specific range (usually 0 to 1 or -1 to 1). For document images, normalization can eliminate image brightness differences caused by factors such as uneven lighting and equipment differences, thereby improving the accuracy of subsequent text detection and recognition.

[0070] Furthermore, the text detection algorithm is a technology for locating text areas in an image. In a document image, the text usually has a specific format and layout, such as a title, field name, data, etc. These features can be used by the text detection algorithm to locate the text area and identify the rectangular area containing the text in the image (usually called a text box or text block). After locating the text on the calibration document using the text detection algorithm, one or more text areas can be obtained, and the text area contains all possible text information in the image.

[0071] In an embodiment of the present invention, the document image is grayed and normalized to obtain a calibration document, and the text of the calibration document is located to obtain the text area in the calibration document, which can reduce interference from non-text areas and improve subsequent recognition accuracy.

[0072] In the embodiment of the present invention, performing optical character recognition on the text area to obtain target document information in the document image includes:

[0073] Extracting character features in the text area as characters to be recognized, and matching the characters to be recognized with a preset character template library to obtain target characters;

[0074] The text area containing the target characters is taken as the target document area, and the target document information in the target document area is extracted.

[0075] In an embodiment of the present invention, character features, such as edges, pixel distribution, etc., are extracted from a text area by using a one-hot encoding technique for characters, the extracted character features are matched with a preset character template library, the most similar character features, i.e., the target characters, are found, the text area containing the target characters is used as a target document area, and the target document information in the target document area is extracted using a preset text extraction technique, thereby achieving more accurate recognition.

[0076] Furthermore, the one-hot encoding technique of the characters is a data preprocessing method commonly used to process categorical variables, especially in the fields of machine learning and data mining; this method converts each possible value of the categorical variable into a new binary feature (also called a one-hot feature), thereby constructing a vector containing only 0 and 1 to represent the original data, and the one-hot encoding technique of the characters is mainly used to process categorical data or unordered discrete features. One-hot encoding represents each category by assigning an independent binary vector to the category, the length of the vector is equal to the total number of categories, each category corresponds to a position in the vector, the value at the position is 1, and the values ​​at the remaining positions are all 0.

[0077] The present invention extracts the character features in the text area as the characters to be recognized, and one-hot encoding can be used to represent the characters in the document image. Each character is mapped to a vector of a fixed length, eliminating the erroneous numerical relationship that may exist between documents, especially algorithms involving distance calculation (such as the K nearest neighbor algorithm). One-hot encoding converts classified data into binary form, allowing these algorithms to process better, thereby allowing computers to understand and process more easily, thereby quickly obtaining the target document.

[0078] Furthermore, the character template library is a collection of various character features, each character template is designed for the shape and structure of a specific character, and can be created manually or generated by scanning known character samples. In OCR (optical character recognition) technology, feature extraction is performed on the document image to be recognized. These features may include edges, contours, corner points, etc., which are used to describe the shape of the character. These features are compared and matched with the templates in the character template library. The similarity between the character to be recognized and the template character is calculated by methods such as Euclidean distance or correlation coefficient, and the template most similar to the character to be recognized is found, thereby realizing character recognition.

[0079] In the embodiment of the present invention, the document image is calibrated and recognized to eliminate problems such as distortion, uneven brightness, color deviation, etc. in the image, thereby improving the overall quality of the image. The target document information corresponding to the document image is extracted through optical character recognition technology, thereby improving recognition accuracy and underwriting efficiency.

[0080] S4. Determine the special risk level according to the target text information and the target document information to obtain a risk level result.

[0081] In the embodiment of the present invention, the special risk level refers to the level divided after risk assessment of specific risks or specific objects (such as special merchants, insurance customers, etc.); the special risk level includes high risk, medium risk, low risk and no risk.

[0082] In the embodiment of the present invention, determining the special risk level according to the target text information and the target document information to obtain the risk level result includes:

[0083] Perform risk type identification on the target text information and the target document information to obtain a risk type information record;

[0084] The risk level of the risk type information record is determined according to a predefined risk level determination rule to obtain a risk level result.

[0085] In the embodiment of the present invention, the predefined risk level determination rules are specifically shown in Table 1 below:

[0086] Table 1

[0087]

[0088] In an embodiment of the present invention, risk type identification is performed on the customer information, subject information, clause name and liability information in the target text information and the target document information, that is, the target text information and the target document information are compared to obtain the inconsistency rate of each target information, and the risk level is determined according to the inconsistency rate in accordance with predefined risk level determination rules to obtain a risk level result.

[0089] In detail, for example, Enterprise A purchases an engineering insurance policy from an insurance company, and the insured is "Let's Play Together, Guangdong Branch", but the special agreement states that the insured is "Let's Play Together, Guangdong Branch". Then, through comparison, it is found that the inconsistency rate of the insured, i.e. the customer information, is 38.46%, and the special agreement is determined to be a medium-risk special agreement.

[0090] Furthermore, for example, if Company B purchases a property all-risk insurance policy from an insurance company, two clauses have been added to the policy, namely the warehouse property declaration clause and the replacement value clause, and then two clauses are extended through a special agreement, namely the abnormal event extension clause and the total theft clause. After comparison, it is found that there are two inconsistent clauses in the special agreement, and the special agreement is determined to be a medium-risk special agreement.

[0091] Furthermore, for example, Enterprise C purchases a property all-risk insurance policy from an insurance company, and expands the exclusions through a special agreement, which states that "the insurance company must pay compensation if the following situations occur: 1. Earthquakes, tsunamis and their secondary disasters; 2. Nuclear radiation, nuclear fission, nuclear fusion, nuclear pollution and other radioactive pollution; 3. Theft, robbery; ..." After comparison, it is found that there are multiple liability inconsistencies in the special agreement, and the special agreement is determined to be a high-risk special agreement.

[0092] In the embodiment of the present invention, by integrating the target text information and the target document information, a comprehensive and in-depth analysis of the risks of the contracted object can be performed, thereby determining the contracted risk level, improving the accuracy of risk assessment, and ensuring the accuracy and reliability of the risk level results.

[0093] S5. When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting.

[0094] In the embodiment of the present invention, risk warnings are issued to relevant risk departments and insured persons for each insurance underwriting in the form of message notifications according to the risk level results, and the risk warnings are recorded.

[0095] In the embodiment of the present invention, when the risk level result meets the warning condition, issuing a warning notification for insurance underwriting includes:

[0096] Determine whether the risk level result meets the preset alarm condition;

[0097] If the risk level result meets the warning condition, a risk warning corresponding to the risk level result is issued and the risk warning is recorded;

[0098] If the risk level result does not meet the alarm condition, no alarm notification is issued.

[0099] In an embodiment of the present invention, the risk level results that meet the alarm conditions include high risk, medium risk, and low risk. The obtained risk level results are compared with the preset alarm conditions. If the risk level results meet the alarm conditions, the next alarm notification is required. If the risk level results meet the alarm conditions, an alarm is generated and the alarm notification is sent to relevant underwriting personnel through a preset communication channel (such as SMS, email, internal system, etc.). In addition to issuing an alarm notification, the alarm information will also be recorded in a preset risk management system or database. The alarm information record includes key information such as alarm time, alarm content, processing personnel, processing results, etc., which is helpful for subsequent risk analysis, auditing and tracing.

[0100] Specifically, if the special risk level is high risk, the insurance ban process will be initiated; if the special risk level is medium risk, the insurance underwriting will be transferred to manual review, and the underwriter will be informed of the underwriting results and the inconsistencies.

[0101] If the underwriter determines that there is still a need for underwriting, he or she may apply for flexibility, and the underwriter at the headquarters will make underwriting recommendations based on comprehensive considerations; if the special risk level is low risk, the insurance underwriting will be transferred to manual review, and the underwriter will be informed of the underwriting results and the inconsistencies. If the underwriter determines that there is still a need for underwriting, the underwriter will record the reasons for underwriting; if the special risk level is no risk, normal underwriting will be allowed.

[0102] In an embodiment of the present invention, when the risk level result meets the alarm condition, an alarm notification is issued for insurance underwriting, which can achieve timely and accurate identification and alarm of risks in the insurance underwriting process, help improve the risk management capabilities of the insurance company, reduce losses caused by potential risks, and also help improve customers' trust and satisfaction with the insurance company, thereby improving the accuracy and efficiency of identifying special risks of documents.

[0103] In the embodiment of the present invention, the special contract text is obtained by querying the official website, which ensures the timeliness and authority of the special contract text, and the document image is obtained by intelligent shooting tools, etc., which ensures the high quality of the obtained document image, which is convenient for the subsequent recognition and processing of the document image; the structural decomposition of the special contract text and the calibration and recognition of the document image can quickly process a large number of insurance policies and special clauses, significantly reduce the time and cost of manual review, and greatly improve the overall recognition efficiency; the target text information and the target document information are compared by the intelligent rule engine to determine the special risk level, reduce the errors and omissions of manual review, ensure the consistency and compliance of the document and special clauses, and improve the review efficiency; according to the risk level result, the insurance underwriting is warned and notified, so as to realize the timely and accurate recognition and warning of the risks in the insurance underwriting process, and improve the recognition efficiency and review efficiency. It should be understood that the size of the sequence number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0104] In the embodiment of the present disclosure, a document-specific risk identification device is provided, and the document-specific risk identification device corresponds one-to-one with the document-specific risk identification method in the above embodiment. Figure 3 As shown, the document-specific risk identification device 100 can be installed in an electronic device. According to the functions to be implemented, the document-specific risk identification device 100 includes a collection module 101, a structured decomposition module 102, a calibration identification module 103, a risk level module 104, and an alarm module 105. The functional modules are described in detail as follows:

[0105] The acquisition module 101 is used to obtain the special contract text and document image of the insurance underwriting;

[0106] A structured decomposition module 102 is used to perform structured decomposition on the special text to obtain target text information in the special text;

[0107] A calibration and recognition module 103 is used to calibrate and recognize the document image to obtain target document information in the document image;

[0108] The risk level module 104 is used to determine the special risk level according to the target text information and the target document information to obtain a risk level result;

[0109] The alarm module 105 is used to issue an alarm notification for insurance underwriting when the risk level result meets the alarm condition.

[0110] In one embodiment, the structured decomposition module 102 is specifically configured to:

[0111] Cleaning and standardizing the special contract text to obtain a standardized special contract text;

[0112] The standardized special text is classified using a preset text classification algorithm to obtain target text information in the special text.

[0113] In one embodiment, the structured decomposition module 102 is specifically configured to:

[0114] Performing text vectorization on the standardized special text to obtain a numerical vector text corresponding to the standardized special text;

[0115] Using a preset support vector machine algorithm to divide the numerical vector text into a plurality of classified special text blocks;

[0116] Preset keywords are extracted based on the classified special text blocks to obtain target text information in the standardized special text.

[0117] In one embodiment, the calibration identification module 103 is specifically configured to:

[0118] Gray-scaling and normalizing the document image to obtain a calibration document;

[0119] Using a text detection algorithm to locate text on the calibration document to obtain a text area in the calibration document;

[0120] Optical character recognition is performed on the text area to obtain target document information in the document image.

[0121] In one embodiment, the calibration identification module 103 is specifically configured to:

[0122] Extracting character features in the text area as characters to be recognized, and matching the characters to be recognized with a preset character template library to obtain target characters;

[0123] The text area containing the target characters is taken as the target document area, and the target document information in the target document area is extracted.

[0124] In one embodiment, the risk level module 104 is specifically configured to:

[0125] Perform risk type identification on the target text information and the target document information to obtain a risk type information record;

[0126] The risk level of the risk type information record is determined according to a predefined risk level determination rule to obtain a risk level result.

[0127] In one embodiment, the alarm module 105 is specifically used to:

[0128] Determine whether the risk level result meets the preset alarm condition;

[0129] If the risk level result meets the warning condition, a risk warning corresponding to the risk level result is issued and the risk warning is recorded;

[0130] If the risk level result does not meet the alarm condition, no alarm notification is issued.

[0131] In the present invention, for the identification of special risks of documents, the special text is obtained by querying the official website, which ensures the timeliness and authority of the special text, and the document image is obtained by intelligent shooting tools, etc., which ensures the high quality of the obtained document image, which is convenient for the subsequent identification and processing of the document image; the special text is structurally decomposed and the document image is calibrated and identified, which can quickly process a large number of insurance policies and special clauses texts, significantly reduce the time and cost of manual review, and greatly improve the overall recognition efficiency; the target text information and the target document information are compared by the intelligent rule engine to determine the special risk level, reduce the errors and omissions of human review, ensure the consistency and compliance of the documents and special clauses, and improve the review efficiency; according to the risk level result, the insurance underwriting is warned and notified, so as to realize the timely and accurate identification and warning of the risks in the insurance underwriting process, and improve the identification efficiency and review efficiency. The specific definition of the document special risk identification device can refer to the definition of the document special risk method above, which will not be repeated here. Each module in the above-mentioned document special risk identification device can be fully or partially realized by software, hardware and their combination. The above modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above modules.

[0132] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client through a network connection. When the computer program is executed by the processor, it realizes the functions or steps on the service side of a document special risk identification method.

[0133] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, it realizes the functions or steps on the client side of a document special risk identification method.

[0134] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0135] Obtain insurance coverage contract text and document images;

[0136] Performing structural decomposition on the special text to obtain target text information in the special text;

[0137] Calibrate and identify the document image to obtain target document information in the document image;

[0138] Determine the special risk level according to the target text information and the target document information to obtain a risk level result;

[0139] When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting.

[0140] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and apparatuses can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0141] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0142] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0143] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0144] In some implementations of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.

[0145] The readable storage medium of the present invention stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can achieve:

[0146] Obtain insurance coverage contract text and document images;

[0147] Performing structural decomposition on the special text to obtain target text information in the special text;

[0148] Calibrate and identify the document image to obtain target document information in the document image;

[0149] Determine the special risk level according to the target text information and the target document information to obtain a risk level result;

[0150] When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting.

[0151] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0152] The computer-readable storage medium may also store at least one computer executable program / instruction, which may be, for example, a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.

[0153] In addition, the computer device may also include (but not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (eg, keyboard, mouse, speaker, etc.), etc.

[0154] The processor may communicate with external devices via an I / O bus via a wired or wireless network.

[0155] In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and / or method steps in the embodiments described in the present technology.

[0156] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0157] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0158] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0159] It should be noted that in the present disclosure, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, an element limited by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0160] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

[0161] It should be noted that if software tools or components other than those of our company appear in the embodiments of the present application, they are only used for illustrative purposes and do not represent actual use.

Claims

1. A document-specific risk identification method, characterized in that: The method comprises: Obtain insurance coverage contract text and document images; Performing structural decomposition on the special text to obtain target text information in the special text; Calibrate and identify the document image to obtain target document information in the document image; Determine the special risk level according to the target text information and the target document information to obtain a risk level result; When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting.

2. A document-specific risk identification method as claimed in claim 1, characterized in that: The structural decomposition of the special text to obtain target text information in the special text includes: Cleaning and standardizing the special contract text to obtain a standardized special contract text; The standardized special text is classified using a preset text classification algorithm to obtain target text information in the special text.

3. A document-specific risk identification method as claimed in claim 2, characterized in that: The step of classifying the standardized special text using a text classification algorithm to obtain target text information in the special text includes: Performing text vectorization on the standardized special text to obtain a numerical vector text corresponding to the standardized special text; Using a preset support vector machine algorithm to divide the numerical vector text into a plurality of classified special text blocks; Preset keywords are extracted based on the classified special text blocks to obtain target text information in the standardized special text.

4. A document-specific risk identification method as claimed in claim 1, characterized in that: The step of calibrating and identifying the document image to obtain target document information in the document image includes: Gray-scaling and normalizing the document image to obtain a calibration document; Using a text detection algorithm to locate text on the calibration document to obtain a text area in the calibration document; Optical character recognition is performed on the text area to obtain target document information in the document image.

5. A document-specific risk identification method as claimed in claim 4, characterized in that: The performing optical character recognition on the text area to obtain target document information in the document image includes: Extracting character features in the text area as characters to be recognized, and matching the characters to be recognized with a preset character template library to obtain target characters; The text area containing the target characters is taken as the target document area, and the target document information in the target document area is extracted.

6. A document-specific risk identification method as claimed in claim 1, characterized in that: Determining the special risk level according to the target text information and the target document information to obtain the risk level result includes: Perform risk type identification on the target text information and the target document information to obtain a risk type information record; The risk level of the risk type information record is determined according to a predefined risk level determination rule to obtain a risk level result.

7. A document-specific risk identification method as claimed in claim 1, characterized in that: When the risk level result meets the warning condition, an alarm notification is issued for insurance underwriting, including: Determine whether the risk level result meets the preset alarm condition; If the risk level result meets the warning condition, a risk warning corresponding to the risk level result is issued and the risk warning is recorded; If the risk level result does not meet the alarm condition, no alarm notification is issued.

8. A document-specific risk identification device, characterized in that: The device comprises: The acquisition module is used to obtain the special text and document images of insurance underwriting; A structured decomposition module, used for performing structured decomposition on the special text to obtain target text information in the special text; A calibration and recognition module, used to calibrate and recognize the document image to obtain target document information in the document image; A risk level module, used to determine the special risk level according to the target text information and the target document information, and obtain a risk level result; The alarm module is used to issue an alarm notification for insurance underwriting when the risk level result meets the alarm condition.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute a document-specific risk identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a document-specific risk identification method as described in any one of claims 1 to 7 is implemented.

Images