Business processing intelligent recommendation method, device and equipment and storage medium thereof

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to a business processing intelligent recommendation method and device, equipment and a storage medium thereof. Preprocessing the to-be-processed business data to obtain preprocessed data; inputting the preprocessed data into the intelligent recommendation model after learning training; taking the preprocessed data as a service type calculation basis, and calculating the service type of the to-be-processed service data by adopting a feature comprehensive calculation network in the intelligent recommendation model; and processing scheme recommendation is carried out on the to-be-processed business data by combining recommendation nodes which are respectively learned and trained based on different business types in the intelligent recommendation model. The method is applied to an insurance claim settlement service classification intelligent processing recommendation scene, intelligent processing scheme recommendation can be carried out for claim settlement service types corresponding to different insurance types, and an insurance institution is assisted to better carry out claim settlement service classification shunting processing.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence technology, and is applied to the intelligent processing recommendation scene of insurance claim business classification, such as intelligent recommendation scenes of car insurance claim, health insurance claim, crop insurance claim, investment claim, etc., and relates to a business processing intelligent recommendation method, device, equipment and storage medium thereof. BACKGROUND

[0002] In the traditional claim service, the customer usually needs to fill in the tedious claim application form, submit a large number of proof materials, and the claim process is relatively template, and the audit time is long, which undoubtedly reduces the claim business processing efficiency.

[0003] With the continuous increase of insurance types, the types of claim business have also gradually increased. Since insurance agencies or underwriting companies are involved in different types of claims, there are more or less differences in the details of the claims involved in different types of claims. This has led to a fixed claim process or template that cannot be applied to all types of claims. Therefore, an intelligent business processing recommendation method is needed to recommend different processing schemes for different claim cases to improve claim efficiency. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a business processing intelligent recommendation method, device, equipment and storage medium thereof, so as to recommend different processing schemes for different claim cases to improve claim efficiency.

[0005] In a first aspect, the embodiments of the present application provide a business processing intelligent recommendation method, which adopts the technical scheme as follows:

[0006] A business processing intelligent recommendation method includes the following steps:

[0007] Receiving to-be-processed business data;

[0008] Preprocessing the to-be-processed business data to obtain preprocessed data;

[0009] Inputting the preprocessed data into a learning-trained intelligent recommendation model;

[0010] Taking the preprocessed data as the basis for calculating the business type, using the feature comprehensive calculation network in the intelligent recommendation model to calculate the business type of the to-be-processed business data;

[0011] Combining the recommendation nodes learned and trained based on different business types in the intelligent recommendation model, the to-be-processed business data is recommended for processing scheme, wherein different recommendation nodes are processing entry nodes of different business processing paths corresponding to different business types.

[0012] In a second aspect, the embodiments of the present application further provide a business processing intelligent recommendation device, which adopts the technical scheme as follows:

[0013] A business processing intelligent recommendation device comprises:

[0014] A data receiving module is configured to receive to-be-processed business data.

[0015] A data preprocessing module is configured to preprocess the to-be-processed business data to obtain preprocessed data.

[0016] A model input module is configured to input the preprocessed data into a learned intelligent recommendation model.

[0017] A business type calculation module is configured to take the preprocessed data as a basis for business type calculation, and calculate a business type of the to-be-processed business data by using a feature comprehensive calculation network in the intelligent recommendation model.

[0018] A processing scheme recommendation module is configured to recommend a processing scheme for the to-be-processed business data in combination with recommendation nodes learned and trained based on different business types in the intelligent recommendation model, wherein different recommendation nodes are processing cut-in nodes of business processing paths corresponding to different business types.

[0019] In a third aspect, the embodiments of the present application further provide a computer device, which adopts the technical scheme as follows:

[0020] A computer device comprises a memory and a processor, the memory stores computer readable instructions, and the processor implements the steps of the business processing intelligent recommendation method as described above when executing the computer readable instructions.

[0021] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which adopts the technical scheme as follows:

[0022] A computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by a processor to implement the steps of the business processing intelligent recommendation method as described above.

[0023] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0024] The business processing intelligent recommendation method provided in the embodiments of the present application comprises the following steps: receiving to-be-processed business data; preprocessing the to-be-processed business data to obtain preprocessed data; inputting the preprocessed data into a learned intelligent recommendation model; calculating the business type of the to-be-processed business data by using a feature comprehensive calculation network in the intelligent recommendation model based on the preprocessed data; and recommending a processing scheme for the to-be-processed business data by combining the recommendation nodes learned and trained based on different business types in the intelligent recommendation model. The method can be applied to the intelligent processing recommendation scene of insurance claim business classification, and can recommend intelligent processing schemes for different claim business types corresponding to different insurance types, thereby assisting insurance institutions to better perform claim business classification and shunting processing. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the schemes in the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0026] Figure 1 is an exemplary system architecture diagram to which the present application can be applied;

[0027] Figure 2 is a flowchart of one embodiment of a business processing intelligent recommendation method according to the present application;

[0028] Figure 3 is a flowchart of one specific embodiment of step 202 shown in Figure 2

[0029] Figure 4 is a flowchart of one specific embodiment of learning and training of an intelligent recommendation model in the business processing intelligent recommendation method provided in the present application;

[0030] Figure 5 is a flowchart of one specific embodiment of step 402 shown in Figure 4

[0031] Figure 6 is a flowchart of one specific embodiment of data quantity preposition judgment of historical claim case data of different business types in the business processing intelligent recommendation method provided in the present application;

[0032] Figure 7 is a flowchart of one specific embodiment of step 403 shown in Figure 4

[0033] Figure 8 is a flowchart of one specific embodiment of step 404 shown in Figure 7 ​​​A flow chart of one specific embodiment of the illustrated step 701.

[0034] Figure 9 is Figure 4 A flow chart of one specific embodiment of the illustrated step 404.

[0035] Figure 10 is a structural schematic diagram of one embodiment of a business processing intelligent recommendation device according to the present application;

[0036] Figure 11 is a structural schematic diagram of one embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description herein is for describing particular embodiments only and is not intended to be limiting of the application; the description and the drawings are to be regarded as illustrative in nature and are not intended to limit the application; the terminology used in the description and the claims of the present application and the above description of the drawings includes the terms specifically mentioned above, as well as their derivatives.

[0038] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase "in an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily all directed to the same embodiment, or to a single alternative embodiment.

[0039] In order for those skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings below.

[0040] As Figure 1 illustrated, the system architecture 100 can include a terminal device 101, a network 102 and a server 103, the terminal device 101 can be a notebook computer 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is a medium for providing a communication link between the terminal device 101 and the server 103. The network 102 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0041] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0042] The terminal device 101 can be various electronic devices with display screens and supporting web browsing, in addition to the notebook computer 1011, the tablet computer 1012 or the mobile phone 1013, the terminal device 101 can also be an electronic book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer and a desktop computer, etc.

[0043] The server 103 can be a server providing various services, such as a background server supporting the page displayed on the terminal device 101.

[0044] It should be noted that the business processing intelligent recommendation method provided by the embodiments of the present application is generally executed by a server, and accordingly, the business processing intelligent recommendation device is generally arranged in the server.

[0045] It should be understood that Figure 1 The number of terminal devices, networks and servers in

[0046] With reference to Figure 2 , a flow chart of one embodiment of a business processing intelligent recommendation method according to the present application is shown. The business processing intelligent recommendation method comprises the following steps:

[0047] Step 201, receiving to-be-processed business data.

[0048] In this embodiment, the to-be-processed business data includes to-be-processed claim settlement business data, and the claim settlement business data includes claim settlement audit data, audit feedback data, contract term agreement data at the time of signing an insurance contract, claim settlement processing flow data after the audit passes, and the like. The claim settlement audit data includes case-related matter data, for example, vehicle accident-related data in a vehicle insurance claim settlement scenario, related diagnosis, and hospital-prescribed medical treatment expense receipts in a medical insurance claim settlement scenario. The audit feedback data includes liability compensation data and compensation amount borne by an insurer. The claim settlement processing estimation flow data after the audit passes includes task estimation processing nodes during actual claim settlement and bank card number data for fund transfer.

[0049] It should be understood that the business processing intelligent recommendation method described in this embodiment is applied to an insurance claim settlement business type intelligent processing recommendation scenario, for example, an intelligent recommendation scenario of vehicle insurance claim settlement, health insurance claim settlement, crop insurance claim settlement, and investment claim settlement. Since an insurance institution or an insurer is involved in different types of claim settlement, the details of claim settlement involved in different types of insurance also have more or less differences, which leads to a fixed claim settlement process or template that cannot be applied to all types of claim settlement. Therefore, the present application aims to provide a business processing intelligent recommendation method that can recommend intelligent processing solutions for different types of claim settlement business types corresponding to different types of insurance, thereby assisting the insurance institution to better process claim settlement business classification and distribution.

[0050] Step 202, preprocessing the to-be-processed business data to obtain preprocessed data.

[0051] In this embodiment, the preprocessing includes cleaning, standardizing, and feature encoding the original claim settlement business data, so that the subsequent intelligent recommendation model can quickly perform artificial intelligence processing based on the feature encoding.

[0052] Step 203, inputting the preprocessed data into the learned intelligent recommendation model.

[0053] In this embodiment, the learned intelligent recommendation model can recommend intelligent processing solutions for different types of claim settlement business types corresponding to different types of insurance.

[0054] Step 204, calculating the business type of the to-be-processed business data by using the feature comprehensive calculation network in the intelligent recommendation model based on the preprocessed data.

[0055] In this embodiment, the business type is a preset claim settlement business type according to different types of insurance, for example: car insurance claim settlement, health insurance claim settlement, crop insurance claim settlement, investment claim settlement, etc., and health insurance claim settlement can be further divided into more detailed claim settlement business types according to different disease types or insurance types.

[0056] Since the claim settlement business data corresponding to different claim settlement business types have certain differences, the data features hit by different claim settlement business types also have certain differences, so the feature code corresponding to the claim settlement business data to be processed is taken as the basis for business type calculation, thereby identifying the business type corresponding to the claim settlement business data to be processed.

[0057] By taking the preprocessed data as the basis for business type calculation, that is, taking the feature code corresponding to the claim settlement business data to be processed as the basis for business type calculation, the feature comprehensive calculation network in the intelligent recommendation model is used to calculate the business type of the business data to be processed, thereby realizing automatic and intelligent prediction of the business type.

[0058] In step 205, the recommendation nodes learned and trained based on different business types in the intelligent recommendation model are combined to recommend a processing scheme for the business data to be processed.

[0059] Different recommendation nodes are processing entry nodes of business processing paths corresponding to different business types.

[0060] In this embodiment, the method comprises the following steps: receiving business data to be processed; preprocessing the business data to be processed to obtain preprocessed data; inputting the preprocessed data into the learned intelligent recommendation model; taking the preprocessed data as the basis for business type calculation, and using the feature comprehensive calculation network in the intelligent recommendation model to calculate the business type of the business data to be processed; and combining the recommendation nodes learned and trained based on different business types in the intelligent recommendation model to recommend a processing scheme for the business data to be processed. When this method is applied to the intelligent processing recommendation scene of insurance claim settlement business classification, intelligent processing scheme recommendation can be performed for different claim settlement business types corresponding to different types of insurance, thereby assisting insurance institutions to better classify and process claim settlement business.

[0061] With reference to Figure 3 , Figure 3 is Figure 2 a flowchart of one specific embodiment of step 202, comprising the following steps:

[0062] In step 301, a data cleaning algorithm is used to remove noise data and correct error values from the business data to be processed.

[0063] Step 302, for the to-be-processed business data after removing noise data and correcting error values, uniform formatting processing is performed by using a standardization technique to obtain to-be-processed business data with uniform data formats;

[0064] Step 303, the to-be-processed business data with uniform data formats is input into a preset feature extraction network to perform feature coding and extract business feature data codes corresponding to the to-be-processed business data.

[0065] Through the cleaning, standardization and feature coding of the claim settlement business data, subsequent intelligent recommendation models can perform rapid artificial intelligence processing according to the feature codes.

[0066] With reference to Figure 4 In some optional implementations, before step 203, the method further includes a step of learning and training an intelligent recommendation model, Figure 4 is a flowchart of a specific embodiment of the method of learning and training an intelligent recommendation model in the business processing intelligent recommendation method described in the present application, including the following steps:

[0067] Step 401, a batch of historical claim settlement case data is collected, wherein the historical claim settlement case data all contain claim settlement processing schemes;

[0068] Specifically, the batch of historical claim settlement case data refers to historical claim settlement case data of all business types stored by insurance agencies or underwriting companies, and also includes typical claim settlement case data corresponding to different business types provided by third-party platforms.

[0069] Step 402, the batch of historical claim settlement case data is labeled to obtain historical claim settlement case data of different business types and claim settlement processing schemes corresponding to different business types;

[0070] Specifically, when performing labeling, automatic labeling can be performed in combination with a keyword matching method, for example, when a car insurance claim settlement business is performed, keywords such as "car" and "vehicle" are definitely included; when a health insurance claim settlement business is performed, keywords such as "disease name" and "patient" are definitely included.

[0071] Step 403, data augmentation is performed on the historical claim settlement case data of different business types to obtain augmented claim settlement case data sets corresponding to each business type;

[0072] The historical claim case data of different business types is expanded to avoid that the claim case data of certain insurance types is insufficient to support the learning and training of the model. Therefore, data augmentation is performed to ensure that the intelligent recommendation model after learning and training is trained under sufficient data, thereby improving the credibility of the model in actual use.

[0073] In step 404, the expanded claim case data set corresponding to all business types is input into the intelligent recommendation model to be trained to learn the claim processing scheme corresponding to the claim case data of different business types, and a pre-trained intelligent recommendation model is obtained.

[0074] By learning the claim processing scheme corresponding to the claim case data of different business types, different claim processing schemes can be allocated to different claim business types in subsequent processes.

[0075] In the embodiment, before step 402 is performed, the method further includes: preprocessing the batch of historical claim case data to obtain preprocessed data; specifically, using a data cleaning algorithm to remove noise data and correct error values; using standardization technology to perform unified formatting processing on the batch of historical claim case data after removing noise data and correcting error values, to obtain batch of historical claim case data with unified data format; inputting the batch of historical claim case data with unified data format into a pre-set feature extraction network to perform feature encoding and extracting business feature data code corresponding to the batch of historical claim case data.

[0076] With reference to Figure 5 , Figure 5 is Figure 4 a flowchart of one specific embodiment of step 402, including the following steps:

[0077] In step 501, a keyword recognition technology is used to identify the business type field in all historical claim case data, wherein the different business types include different claim types preset according to different insurance types.

[0078] Specifically, the keyword recognition technology can use a natural language-based text recognition technology to identify the business type field in all historical claim case data.

[0079] In step 502, according to the business type field identification result, all historical claim case data is processed by automatic claim type annotation to obtain the corresponding relationship between the claim type and the historical claim case data.

[0080] Specifically, a loop processing mode can be adopted to automatically label the types of all historical claim case data one by one until the labeling is completed and the loop labeling step is stopped.

[0081] In step 503, the claim processing solutions corresponding to all historical claim case data are identified respectively.

[0082] In step 504, the claim processing solutions corresponding to different claim types are automatically labeled according to the correspondence between the claim types and the historical claim case data and the claim processing solutions corresponding to all historical claim case data respectively.

[0083] By labeling the claim processing solutions corresponding to different claim types, all claim processing solutions corresponding to the same claim type can be sorted, so that the optimal claim processing solution or the most commonly used claim processing solution corresponding to the same claim type can be obtained.

[0084] With reference to Figure 6 In some optional implementations, before step 403, the method further includes a step of performing data quantity pre-judgment on historical claim case data of different business types, Figure 6 is a flowchart of one specific embodiment of the method for performing data quantity pre-judgment on historical claim case data of different business types in the business processing intelligent recommendation method described in the present application, including the following steps:

[0085] In step 601, the data quantity of historical claim case data of different business types is counted to obtain a statistical result.

[0086] In step 602, whether the historical claim case data of different business types reaches a preset data quantity threshold is judged according to the statistical result.

[0087] In step 603, if the historical claim case data of the current business type does not reach the preset data quantity threshold, the historical claim case data of the current business type is filtered out.

[0088] In addition, if the historical claim case data of the current business type reaches the preset data quantity threshold, step 404 is directly performed to learn the claim processing solutions corresponding to the historical claim case data of the current business type.

[0089] By judging whether the historical claim case data of different business types reaches the preset data quantity threshold, the historical claim case data of the target business type is expanded to ensure that the training data of each business type for model learning and training is sufficient enough, so as to avoid that the model is too inclined to a specific business type.

[0090] With reference to Figure 7, Figure 7 is Figure 4 a flow chart of one specific embodiment of step 403, comprising the following steps:

[0091] Step 701, using a preset generative adversarial network to expand the historical claim case data of the current business type;

[0092] Specifically, the generative adversarial network includes a generator and a discriminator.

[0093] Step 702, until the total number of expanded historical claim case data of the current business type reaches the data threshold, stop expanding the historical claim case data of the current business type, and obtain the expanded claim case data set corresponding to the current business type.

[0094] Continuing to refer to Figure 8 , Figure 8 is Figure 7 a flow chart of one specific embodiment of step 701, comprising the following steps:

[0095] Step 801, obtaining the business feature data code corresponding to the historical claim case data of the current business type respectively;

[0096] Step 802, identifying the common business feature data code contained in the historical claim case data of the current business type from the business feature data code;

[0097] Step 803, decoding the common business feature data code to obtain the common claim case data of the historical claim case data of the current business type;

[0098] By identifying the common business feature data code contained in the historical claim case data of the current business type, and decoding the common business feature data code, the common claim case data of the historical claim case data of the current business type is obtained. In order to ignore the non-common feature data under the same case type in the subsequent, only use the common feature data and common feature data code under the same case type for data expansion processing, which avoids the complexity of expansion while fully utilizing important common data.

[0099] Step 804, inputting the common business feature data code into the generator to simulate the generation of claim case data, and obtaining the simulation generation result;

[0100] Step 805, inputting the simulation generation result into the discriminator, and calculating the difference value between the simulation generation result and the common claim case data;

[0101] Step 806, until the difference value is lower than the preset difference threshold, then expand the data based on the simulation generation result to expand the target number of claim case data, wherein the target number is the difference between the data number threshold and the data number of the historical claim case data of the current business type.

[0102] With reference to Figure 9 , Figure 9 Is Figure 4 a flowchart of one specific embodiment of step 404, including the following steps:

[0103] Step 901, identifying the claim processing scheme corresponding to different claim types respectively;

[0104] Step 902, arranging the claim processing nodes of the claim processing scheme corresponding to the same claim type, and according to the execution relationship of the claim processing nodes, summarizing the claim processing path corresponding to different claim types;

[0105] Step 903, setting the first processing node in the claim processing path corresponding to different claim types as the corresponding processing cut-in node;

[0106] Step 904, deploying the processing cut-in node of the claim processing path corresponding to different claim types to the intelligent recommendation model as a recommended node, to obtain the pre-trained intelligent recommendation model.

[0107] By setting the first processing node in the claim processing path corresponding to different claim types as the corresponding processing cut-in node, and deploying the processing cut-in node of the claim processing path corresponding to different claim types to the intelligent recommendation model as a recommended node, so as to facilitate subsequent provision of the processing cut-in node to the to-be-processed business data after determining the business type of the to-be-processed business data, and automatically performing process-based claim processing according to the claim processing path.

[0108] In this embodiment, by receiving to-be-processed business data; pre-processing the to-be-processed business data to obtain pre-processed data; inputting the pre-processed data into the learned intelligent recommendation model; using the pre-processed data as the basis for calculating the business type, using the feature comprehensive calculation network in the intelligent recommendation model to calculate the business type of the to-be-processed business data; and combining the recommended nodes learned and trained based on different business types in the intelligent recommendation model to recommend a processing scheme for the to-be-processed business data. Applying this method to the intelligent processing recommendation scene of insurance claim business classification can intelligently recommend processing schemes for different claim business types corresponding to different insurance types, and assist insurance institutions in better classifying and distributing claim business.

[0109] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence (Artificial Intelligence, AI) is to use 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.

[0110] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The artificial intelligence software technology mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc. several directions.

[0111] In the embodiment, the to-be-processed business data is received; the to-be-processed business data is preprocessed to obtain preprocessed data; the preprocessed data is input into the intelligent recommendation model after learning and training; the business type of the to-be-processed business data is calculated by the intelligent recommendation model based on the preprocessed data as the business type calculation basis; and the recommendation node learned and trained based on different business types in the intelligent recommendation model is combined to recommend a processing scheme for the to-be-processed business data. The method is applied to the intelligent processing recommendation scene of insurance claim business classification, which can recommend intelligent processing schemes for different claim business types corresponding to different insurance types, and assist insurance agencies to better classify and process claims.

[0112] Further reference Figure 10 , as an implementation of the method shown in Figure 2 , the present application provides an embodiment of a business processing intelligent recommendation device. The device embodiment corresponds to the method embodiment shown in Figure 2 , and the device can be applied to various electronic devices.

[0113] As shown in Figure 10 , the business processing intelligent recommendation device 10 described in the embodiment includes a data receiving module 10a, a data preprocessing module 10b, a model input module 10c, a business type calculation module 10d, and a processing scheme recommendation module 10e. Among them:

[0114] The data receiving module 10a is configured to receive to-be-processed business data.

[0115] The data preprocessing module 10b is configured to preprocess the to-be-processed business data to obtain preprocessed data.

[0116] The model input module 10c is configured to input the preprocessed data into the intelligent recommendation model after learning and training.

[0117] a business type calculation module 10d configured to calculate the business type of the to-be-processed business data by using a feature comprehensive calculation network in the intelligent recommendation model based on the pre-processed data;

[0118] a processing scheme recommendation module 10e configured to recommend a processing scheme for the to-be-processed business data by combining different recommendation nodes learned and trained based on different business types in the intelligent recommendation model, wherein each recommendation node corresponds to a processing entry node of a business processing path of a different business type.

[0119] The present application receives to-be-processed business data, pre-processes the to-be-processed business data to obtain pre-processed data, inputs the pre-processed data into a learned intelligent recommendation model, calculates the business type of the to-be-processed business data by using a feature comprehensive calculation network in the intelligent recommendation model based on the pre-processed data, and recommends a processing scheme for the to-be-processed business data by combining different recommendation nodes learned and trained based on different business types in the intelligent recommendation model. The method can be applied to an insurance claim business type intelligent processing recommendation scenario, can recommend intelligent processing schemes for different claim business types corresponding to different insurance types, and can assist insurance institutions in better classifying and distributing claim business.

[0120] In the embodiment, the business processing intelligent recommendation device 10 further includes a batch acquisition module, a case labeling module, a data expansion module, and a model learning module. Specifically,

[0121] The batch acquisition module is configured to acquire batch historical claim case data, wherein the historical claim case data all include claim processing schemes.

[0122] The case labeling module is configured to label the batch historical claim case data to obtain historical claim case data of different business types and claim processing schemes corresponding to the different business types.

[0123] The data expansion module is configured to expand the historical claim case data of different business types to obtain expanded claim case data sets corresponding to each business type.

[0124] The model learning module is configured to input the expanded claim case data sets corresponding to all business types into a to-be-trained intelligent recommendation model to learn claim processing schemes corresponding to the claim case data of different business types respectively, and obtain a pre-trained intelligent recommendation model.

[0125] In the embodiment, the case labeling module comprises a keyword recognition unit, a first labeling unit, a claim processing scheme recognition unit and a second labeling unit. Wherein:

[0126] The keyword recognition unit is configured to recognize the business type field in all historical claim case data by using keyword recognition technology, wherein the different business types include different claim types preset according to different insurance types;

[0127] The first labeling unit is configured to automatically label the historical claim case data according to the business type field recognition result, so as to obtain the corresponding relationship between the claim type and the historical claim case data;

[0128] The claim processing scheme recognition unit is configured to recognize the claim processing scheme corresponding to all historical claim case data respectively;

[0129] The second labeling unit is configured to automatically label the claim processing scheme corresponding to different claim types according to the corresponding relationship between the claim type and the historical claim case data, and the claim processing scheme corresponding to all historical claim case data respectively.

[0130] In the embodiment, the business processing intelligent recommendation device 900 further comprises a data quantity statistical module, a case quantity judgment module and a branch processing module after judgment. Wherein:

[0131] The data quantity statistical module is configured to count the data quantity of the historical claim case data of different business types, so as to obtain a statistical result;

[0132] The case quantity judgment module is configured to judge whether the historical claim case data of different business types reaches a preset data quantity threshold according to the statistical result;

[0133] The branch processing module after judgment is configured to filter out the historical claim case data of the current business type if the historical claim case data of the current business type does not reach the preset data quantity threshold.

[0134] In the embodiment, the data expansion module comprises a business feature data coding acquisition unit, a common business feature data coding recognition unit, a decoding unit, a simulation generation unit, a difference value calculation unit and a case data expansion unit. Wherein:

[0135] The business feature data coding acquisition unit is configured to acquire the business feature data coding corresponding to the historical claim case data of the current business type respectively;

[0136] The common service feature data code identification unit is configured to identify, from the service feature data code, common service feature data code contained in the historical claim case data of the current service type;

[0137] The de-encoding unit is configured to de-encode the common service feature data code to obtain the claim case data common to the historical claim case data of the current service type;

[0138] The simulation generation unit is configured to input the common service feature data code into the generator to generate simulated claim case data, and obtain a simulation generation result;

[0139] The difference value calculation unit is configured to input the simulation generation result into the discriminator to calculate a difference value between the simulation generation result and the common claim case data;

[0140] The case data expansion unit is configured to expand data based on the simulation generation result to expand the target number of claim case data until the difference value is lower than a preset difference threshold, wherein the target number is a difference between the data quantity threshold and the data quantity of the historical claim case data of the current service type.

[0141] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by computer readable instructions instructing related hardware, and the computer readable instructions can be stored in a computer readable storage medium. When the program is executed, the processes of the above-mentioned embodiments can be included. The storage medium can be a non-volatile storage medium such as a magnetic disc, an optical disc, a read-only memory (ROM), or a random access memory (RAM).

[0142] It should be understood that although each step in the flowchart of the accompanying drawings is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0143] To solve the above technical problems, the embodiments of the present application further provide a computer device. For details, please refer to Figure 11 , Figure 11A basic structure block diagram of a computer device for the embodiment is shown in FIG. 1.

[0144] The computer device 11 includes a memory 11a, a processor 11b, and a network interface 11c, which are communicatively connected via a system bus. It should be noted that only the computer device 11 with the memory 11a, the processor 11b, and the network interface 11c is shown in the figure, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art can understand that the computer device herein is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0145] The computer device can be a desktop computer, a notebook computer, a palm computer, a cloud server, or the like. The computer device can interact with a user through a keyboard, a mouse, a remote controller, a touchpad, a voice control device, or the like.

[0146] The memory 11a includes at least one type of readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, or the like. In some embodiments, the memory 11a can be an internal storage unit of the computer device 11, such as a hard disk or a memory of the computer device 11. In other embodiments, the memory 11a can also be an external storage device of the computer device 11, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, or the like. Of course, the memory 11a can also include both the internal storage unit and the external storage device of the computer device 11. In the embodiment, the memory 11a is generally used to store an operating system and various application software installed in the computer device 11, such as computer readable instructions of a business processing intelligent recommendation method, etc. In addition, the memory 11a can also be used to temporarily store various data that have been output or will be output.

[0147] The processor 11b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 11b is generally used to control the overall operation of the computer device 11. In the present embodiment, the processor 11b is configured to execute computer readable instructions stored in the memory 11a or process data, such as computer readable instructions of the business processing intelligent recommendation method.

[0148] The network interface 11c may include a wireless network interface or a wired network interface, and is generally used to establish a communication connection between the computer device 11 and other electronic devices.

[0149] The computer device provided in the present embodiment belongs to the field of artificial intelligence technology, and is applied to an intelligent processing recommendation scenario of insurance claim business classification, such as intelligent recommendation scenarios of car insurance claim, health insurance claim, crop insurance claim, investment claim, etc. The present application receives business data to be processed, pre-processes the business data to be processed to obtain pre-processed data, inputs the pre-processed data into a learned intelligent recommendation model, calculates the business type of the business data to be processed by using a feature comprehensive calculation network in the intelligent recommendation model based on the pre-processed data as the basis for business type calculation, and combines the recommendation nodes learned and trained based on different business types in the intelligent recommendation model to recommend a processing scheme for the business data to be processed. The method is applied to the intelligent processing recommendation scenario of insurance claim business classification, and can recommend intelligent processing schemes for different claim business types corresponding to different insurance types, thereby assisting insurance institutions to better classify and process claim business.

[0150] The present application also provides another embodiment, that is, a computer readable storage medium storing computer readable instructions, which can be executed by a processor to make the processor execute the steps of a business processing intelligent recommendation method as described above.

[0151] The computer readable storage medium provided in the embodiment belongs to the technical field of artificial intelligence, and is applied to an intelligent processing recommendation scene of insurance claim business classification, such as intelligent recommendation scenes of vehicle insurance claim, health insurance claim, crop insurance claim, investment claim, and the like. The application receives to-be-processed business data, pre-processes the to-be-processed business data to obtain pre-processed data, inputs the pre-processed data into a learned intelligent recommendation model, calculates a business type of the to-be-processed business data by using a feature comprehensive calculation network in the intelligent recommendation model based on the pre-processed data as a calculation basis of the business type, and combines a recommendation node learned and trained based on different business types in the intelligent recommendation model to recommend a processing scheme for the to-be-processed business data. The method is applied to the intelligent processing recommendation scene of insurance claim business classification, can recommend intelligent processing schemes for different claim business types corresponding to different insurance types, and assists an insurance institution to better perform claim business classification and shunting processing.

[0152] Through the description of the above implementation manners, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and a necessary general hardware platform, and of course, can also be realized by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the application can be embodied in the form of a software product, the computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, an optical disk), and includes a plurality of instructions to make a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device) execute the methods described in the various embodiments of the application.

[0153] Obviously, the above-described embodiments are only some of the embodiments of the application, not all the embodiments, and the preferred embodiments of the application are given in the drawings, but do not limit the patent scope of the application. The application can be implemented in many different forms, and conversely, the purpose of providing these embodiments is to make the disclosure of the application more thorough and comprehensive. Although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing specific embodiments, or equivalently replace some technical features. Any equivalent structure made by using the contents of the specification and drawings, directly or indirectly applied to other related technical fields, is also within the patent protection scope of the application. The non-company software tools or components appearing in the embodiments of the application are only illustrative, not representing actual use.

Claims

1. A business processing intelligent recommendation method, characterized in that, Includes the following steps: Receive pending business data; The business data to be processed is preprocessed to obtain preprocessed data; The preprocessed data is then input into the trained intelligent recommendation model; Using the preprocessed data as the basis for calculating the business type, the feature integration calculation network in the intelligent recommendation model is used to calculate the business type of the business data to be processed. By combining the recommendation nodes trained based on different business types in the intelligent recommendation model, a processing solution is recommended for the business data to be processed. Here, different recommendation nodes are the processing entry nodes of the business processing path corresponding to different business types.

2. The intelligent recommendation method for business processing according to claim 1, characterized in that, The step of preprocessing the business data to be processed to obtain preprocessed data specifically includes: Data cleaning algorithms are used to remove noise and correct errors in the business data to be processed. After removing noise and correcting errors, the business data to be processed is formatted using standardization techniques to obtain business data with a unified data format. The standardized business data to be processed is input into a preset feature extraction network for feature encoding, thereby extracting the business feature data code corresponding to the business data to be processed.

3. The intelligent recommendation method for business processing according to claim 1, characterized in that, Before performing the step of inputting the preprocessed data into the pre-trained intelligent recommendation model, the method further includes: Collect batches of historical claims data, which include claims processing solutions. The batch of historical claims data is labeled to obtain historical claims data for different business types and corresponding claims processing solutions for different business types; The historical claims data for the different business types are augmented to obtain augmented claims datasets for each business type. The expanded claims dataset corresponding to all business types is input into the intelligent recommendation model to be trained, so as to learn the claims processing schemes corresponding to the claims data of different business types and obtain the pre-trained intelligent recommendation model.

4. The intelligent recommendation method for business processing according to claim 3, characterized in that, The step of annotating the batch of historical claims data to obtain historical claims data for different business types and corresponding claims processing solutions for different business types specifically includes: Keyword recognition technology is used to identify business type fields in all historical claims data, wherein the different business types include different claims types preset according to different insurance types; Based on the business type field identification results, all historical claims data are automatically labeled with claims type one by one to obtain the correspondence between claims type and historical claims data; Identify the corresponding claim processing plan for each of the historical claim cases; Based on the correspondence between claim types and historical claim data, and the claim processing solutions corresponding to each of the historical claim data, the system automatically marks the claim processing solutions corresponding to different claim types.

5. The intelligent recommendation method for business processing according to claim 3, characterized in that, Before performing the step of augmenting the historical claims data for the different business types to obtain the augmented claims dataset for each business type, the method further includes: The number of data entries in historical claims cases for different business types was counted to obtain the statistical results. Based on the statistical results, determine whether the historical claims data for different business types have reached the preset data entry threshold; If the historical claims data for the current business type has not reached the preset data count threshold, then the historical claims data for the current business type will be filtered out. The step of augmenting the historical claims data for different business types to obtain augmented claims datasets for each business type specifically includes: A pre-defined generative adversarial network is used to expand and generate historical claims data for the current business type. The process of expanding the historical claims data for the current business type will continue until the total number of data entries reaches the threshold value. This will stop the process of expanding the historical claims data for the current business type and result in the expanded claims dataset for the current business type.

6. The intelligent recommendation method for business processing according to claim 3, characterized in that, The step of inputting the expanded claims case dataset corresponding to all business types into the intelligent recommendation model to be trained, so as to learn the claims processing schemes corresponding to the claims case data of different business types and obtain the pre-trained intelligent recommendation model, specifically includes: Identify the corresponding claim processing solutions for different claim types; Organize the claim processing nodes for the same claim type and summarize the claim processing paths for different claim types based on the order of execution of the claim processing nodes. Set the first processing node in the claims processing path corresponding to different claims types as the corresponding processing entry node; The processing entry nodes of the claims processing paths corresponding to different claims types are deployed as recommendation nodes into the intelligent recommendation model to obtain the pre-trained intelligent recommendation model.

7. The intelligent recommendation method for business processing according to claim 5, characterized in that, The generative adversarial network (GAN) comprises a generator and a discriminator. The step of using a preset GAN to expand and generate historical claims data for the current business type specifically includes: Obtain the business characteristic data codes corresponding to the historical claims data of the current business type; From the business feature data codes, identify the common business feature data codes contained in the historical claims data of the current business type; The common business feature data encoding is reverse encoded to obtain the claim case data common to the historical claim case data of the current business type; The common business characteristic data is encoded and input into the generator to simulate and generate claims case data, and the simulation generation result is obtained. The simulation results are input into the discriminator to calculate the difference between the simulation results and the common claims data. Until the difference value is lower than the preset difference threshold, the data is expanded based on the simulation result to expand the data to a target number of claims cases, wherein the target number is the difference between the data entry threshold and the number of data entries in the historical claims data of the current business type.

8. A business processing intelligent recommendation device, characterized in that, The intelligent recommendation device for business processing is used to implement the steps of the intelligent recommendation method for business processing as described in any one of claims 1 to 7, and the intelligent recommendation device for business processing includes: The data receiving module is used to receive business data to be processed. The data preprocessing module is used to preprocess the business data to be processed to obtain preprocessed data; The model input module is used to input preprocessed data into the trained intelligent recommendation model; The business type calculation module is used to calculate the business type of the business data to be processed by using the preprocessed data as the basis for business type calculation and employing the feature comprehensive calculation network in the intelligent recommendation model. The processing solution recommendation module is used to combine the recommendation nodes learned and trained based on different business types in the intelligent recommendation model to recommend processing solutions for the business data to be processed. The different recommendation nodes are the processing entry nodes of the business processing paths corresponding to different business types.

9. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the intelligent recommendation method for business processing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the intelligent recommendation method for business processing as described in any one of claims 1 to 7.

Images