Underwriting Questionnaire Problem Processing Method, Device, Computer Equipment and Storage Medium

Abstract

The present application provides a method, apparatus, computer device, and storage medium for processing underwriting questionnaire questions, relating to the field of information processing technology. The method includes: obtaining a newly added underwriting questionnaire and determining a target disease name in the newly added underwriting questionnaire; screening out first underwriting questions corresponding to the target disease name in the newly added underwriting questionnaire, and searching for first general questions corresponding to the target disease name in the general underwriting questionnaire; classifying the first underwriting questions and the first general questions based on a trained classification model to obtain a classification result; and judging whether to update the first general questions in the general underwriting questionnaire according to the classification result. The present application can avoid problems such as omission and mapping errors, improve the work efficiency of business personnel, ensure that the general underwriting questionnaire remains a complete and comprehensive questionnaire after a new insurance company or insurance product is added, and ensure its applicability to the newly added insurance company or insurance product.

Description

Technical Field

[0001] The present application relates to the field of information processing technologies, and in particular, to a method for processing underwriting questionnaire questions, an apparatus for processing underwriting questionnaire questions, a computer device, and a readable storage medium. Background Art

[0002] For the same disease, the number of questions and the order of questions in the intelligent underwriting questionnaires designed by different insurance companies, different insurance types, and different insurance products for this disease may be different. If you want to know which insurance types or insurance products a customer can pass the underwriting for and which insurance types or insurance products the customer will be rejected for, it is necessary for the customer to complete the underwriting questionnaire for each product to draw a conclusion, which is inefficient and likely to cause customer loss.

[0003] In the related art, a method for integrating a complete and comprehensive set of underwriting questions and answers and underwriting conclusions is obtained based on the underwriting questions and answers at all levels of diseases for each insurance type, that is, a general underwriting questionnaire. After the customer answers the questions once, the intelligent underwriting conclusions of each company can be obtained. The general underwriting questionnaire integrated by the above method may be applicable to existing insurance companies and insurance products. However, for an Internet insurance brokerage company, the insurance companies and insurance products it connects to are constantly increasing, and the complete and comprehensive questionnaire obtained above may not be applicable to new insurance companies and insurance products. Summary of the Invention

[0004] In view of this, the present application provides a method for processing underwriting questionnaire questions, an apparatus for processing underwriting questionnaire questions, a computer device, and a readable storage medium, which realizes assisting business personnel in matching and mapping a new underwriting questionnaire with a general underwriting questionnaire and updating the general underwriting questionnaire according to the new underwriting questionnaire.

[0005] In a first aspect, an embodiment of the present application provides a method for processing underwriting questionnaire questions, including: obtaining a new underwriting questionnaire, and determining a target disease name in the new underwriting questionnaire; screening out first underwriting questions corresponding to the target disease name in the new underwriting questionnaire, and searching for first general questions corresponding to the target disease name in the general underwriting questionnaire; classifying the first underwriting questions and the first general questions based on a trained classification model to obtain a classification result; and judging whether to update the first general questions in the general underwriting questionnaire according to the classification result.

[0006] According to the above method for processing underwriting questionnaire questions in the embodiment of the present application, the following additional technical features may also be provided:

[0007] In the above technical solution, optionally, based on the trained classification model, classify the first underwriting question and the first general question to obtain a classification result, including: performing format conversion on the first underwriting question and the first general question to generate input data; performing binary classification on the input data based on the trained classification model to obtain a classification result.

[0008] In any of the above technical solutions, optionally, performing format conversion on the first underwriting question and the first general question to generate input data includes: performing format conversion on the first underwriting question and the first general question to generate first input data and second input data; wherein, the first input data is [CLS]sentenceA[SEP]sentenceB[SEP], and the second input data is [CLS]sentenceB[SEP]sentenceA[SEP], sentenceA represents the first underwriting question, and sentenceB represents the first general question.

[0009] In any of the above technical solutions, optionally, according to the classification result, determine whether to update the first general question in the general underwriting questionnaire, including: if the classification result corresponding to the first input data is 1 and the classification result of the second input data is 1, or if the classification result corresponding to the first input data is 0 and the classification result of the second input data is 1, then there is no need to update the first general question in the general underwriting questionnaire; if the classification result corresponding to the first input data is 1 and the classification result of the second input data is 0, then it is necessary to update the first general question in the general underwriting questionnaire.

[0010] In any of the above technical solutions, optionally, after determining whether to update the first general question in the general underwriting questionnaire according to the classification result, it further includes: if it is determined that the first general question in the general underwriting questionnaire needs to be updated, then send a prompt message, and the prompt message is used to prompt the business personnel to update the first general question in the general underwriting questionnaire.

[0011] In any of the above technical solutions, optionally, before obtaining the new underwriting questionnaire, it further includes: obtaining a sample underwriting questionnaire and determining the sample diseases in the sample underwriting questionnaire; matching and annotating the second underwriting questions corresponding to the sample diseases in the sample underwriting questionnaire with the second general questions corresponding to the sample diseases in the general underwriting questionnaire to generate annotation data; performing format conversion on the annotation data to generate training sample data; inputting the training sample data into the BERT sentence pair classification model for model training to construct a classification model.

[0012] In any of the above technical solutions, optionally, the format of the training sample data is "[CLS]sentence1[SEP]sentence2[SEP]", and the label is 1 or 0. Wherein, if sentence1 represents the second underwriting question, then sentence2 represents the second general question; if sentence1 represents the second general question, then sentence2 represents the second underwriting question.

[0013] In a second aspect, an underwriting questionnaire question processing device provided by an embodiment of the present application includes: a questionnaire acquisition module, configured to acquire a new underwriting questionnaire and determine a target disease name in the new underwriting questionnaire; a question acquisition module, configured to screen out the first underwriting questions corresponding to the target disease name in the new underwriting questionnaire, and search for the first general questions corresponding to the target disease name in the general underwriting questionnaire; a classification module, configured to classify the first underwriting questions and the first general questions based on a trained classification model to obtain a classification result; a judgment module, configured to judge whether to update the first general questions in the general underwriting questionnaire according to the classification result.

[0014] In a third aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory. The memory stores a program or instruction that can run on the processor. When the program or instruction is executed by the processor, the steps of the method in the first aspect are implemented.

[0015] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method in the first aspect are implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is configured to run a program or instruction to implement the method in the first aspect.

[0017] In a sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the method in the first aspect.

[0018] In the embodiments of the present application, for the newly added underwriting questionnaire, first select the target disease name from it, and screen out the questions at all levels corresponding to the target disease name (that is, the first underwriting questions), and then find the first general question corresponding to the target disease name from the general underwriting questionnaire. Further, after converting the formats of the first underwriting questions and the first general questions, input them into the trained classification model to obtain a binary classification result, which is used to indicate whether the semantics of the first underwriting questions and the semantics of the first general questions can be inferred from each other, thereby realizing the matching mapping between the newly added underwriting questionnaire and the general underwriting questionnaire. Finally, according to this classification result, it is judged whether it is necessary to update the first general question in the general underwriting questionnaire according to the first underwriting question.

[0019] In the embodiments of the present application, after integrating a set of general underwriting questionnaires based on the underwriting questions and answers at all levels of diseases in multiple different underwriting questionnaires, in order to assist business personnel in matching and mapping the newly added underwriting questionnaire with the general underwriting questionnaire and updating the general underwriting questionnaire according to the newly added underwriting questionnaire, the above-mentioned question matching method for underwriting questionnaires is proposed, which can avoid problems such as omission and mapping errors, improve the work efficiency of business personnel, ensure that the general underwriting questionnaire is still a complete and comprehensive questionnaire after adding a new insurance company or insurance product, and ensure its applicability to the newly added insurance company or insurance product.

[0020] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are hereinafter specifically exemplified. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0022] Figure 1 A flowchart showing the method for processing underwriting questionnaire questions in the embodiments of the present application is shown;

[0023] Figure 2 A schematic diagram showing the prediction of underwriting questionnaire questions in the embodiments of the present application is shown;

[0024] Figure 3 A schematic diagram showing the training of underwriting questionnaire questions in the embodiments of the present application is shown;

[0025] Figure 4 A block diagram showing the structure of the underwriting questionnaire question processing device in the embodiments of the present application is shown;

[0026] Figure 5The structural block diagram of the computer device according to the embodiment of the present application is shown. Detailed implementation manners

[0027] Next, the technical solutions in the embodiments of the present application will be clearly described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0028] The terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same category, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / " generally means that the associated objects before and after are in an "or" relationship.

[0029] Since the underwriting questionnaires corresponding to different insurance types and different products may be different, in order to enable customers to obtain intelligent underwriting conclusions for each product by only taking one questionnaire, business personnel integrated different underwriting questionnaires to obtain a general underwriting questionnaire. However, with the increase in the number of cooperating insurance companies and insurance products, it is necessary to iteratively update the general underwriting questionnaire and map the newly added underwriting questionnaire to the general questionnaire. In the related art, manual matching and mapping of the newly added underwriting questionnaire and the general underwriting questionnaire, and updating the general underwriting questionnaire are used. However, relying solely on manual matching and mapping is inefficient and may have omissions.

[0030] In addition, the number of questions finally integrated for the first-level questions about a certain disease may be 1, 2, or even 3, that is, one-to-many. If the questions corresponding to the diseases in the newly added underwriting questionnaire are matched with the questions corresponding to the diseases in the general underwriting questionnaire by means of similarity calculation, it is impossible to determine how many similar questions should be returned for one question.

[0031] In the field of NLP (Natural Language Processing), text entailment is defined as a directed inference relationship between a pair of texts, where the antecedent of the entailment is denoted as T (Text), and the consequent of the entailment is denoted as H (Hypothesis). If the semantics of H can be inferred from the semantics of T, then it is said that T entails H, denoted as T→H. Strict text retelling is equivalent to bidirectional entailment. For example:

[0032] T: xx is the founder of yy company

[0033] H: xx founded yy company

[0034] The process of matching the questions in the newly added underwriting questionnaire with those in the general underwriting questionnaire can be approximated as a text paraphrasing task and modeled as a sentence pair binary classification task based on BERT.

[0035] The embodiments of the present application propose an underwriting questionnaire question matching scheme based on BERT sentence pair classification, which improves the work efficiency of business personnel and avoids problems such as omission and mapping errors. The following will, in conjunction with the accompanying drawings, elaborate on the underwriting questionnaire question processing method, underwriting questionnaire question processing device, computer device, and readable storage medium provided by the embodiments of the present application through specific embodiments and their application scenarios.

[0036] The embodiments of the present application provide an underwriting questionnaire question processing method, as Figure 1 shown, the method includes:

[0037] Step 101, obtain the newly added underwriting questionnaire and determine the target disease name in the newly added underwriting questionnaire;

[0038] Step 102, screen out the first underwriting questions corresponding to the target disease name in the newly added underwriting questionnaire, and search for the first general questions corresponding to the target disease name in the general underwriting questionnaire;

[0039] Step 103, classify the first underwriting questions and the first general questions based on the trained classification model to obtain a classification result;

[0040] Step 104, determine whether to update the first general questions in the general underwriting questionnaire according to the classification result.

[0041] In this embodiment, for the newly added underwriting questionnaire, first select the target disease name from it, screen out the questions at all levels corresponding to the target disease name (that is, the first underwriting questions), and then find the first general questions corresponding to the target disease name from the general underwriting questionnaire.

[0042] Furthermore, after formatting the first underwriting questions and the first general questions, input them into the trained classification model to obtain a binary classification result, which is used to indicate whether the semantics of the first underwriting questions and the semantics of the first general questions can be inferred from each other, thereby realizing the matching and mapping between the newly added underwriting questionnaire and the general underwriting questionnaire. Finally, determine whether to update the first general questions in the general underwriting questionnaire according to the first underwriting questions based on the classification result.

[0043] In the embodiment of the present application, after integrating a set of general underwriting questionnaires based on the underwriting questions and answers at all levels of diseases in multiple different underwriting questionnaires, in order to assist business personnel in matching and mapping the new underwriting questionnaire with the general underwriting questionnaire and updating the general underwriting questionnaire according to the new underwriting questionnaire, the above-mentioned question matching method for underwriting questionnaires is proposed, which can avoid problems such as omission and mapping errors, improve the work efficiency of business personnel, ensure that the general underwriting questionnaire remains a complete and comprehensive questionnaire after adding a new insurance company or insurance product, and ensure its applicability to the new insurance company or insurance product.

[0044] It should be noted that when it is determined that the first general question in the general underwriting questionnaire needs to be updated, a prompt message can be sent to prompt the business personnel to update the first general question in the general underwriting questionnaire, or the first general question in the general underwriting questionnaire can also be automatically updated according to the first underwriting question.

[0045] In an embodiment of the present application, the overall process of the underwriting questionnaire question processing method is divided into the following two stages:

[0046] 1. Training stage

[0047] 1.1 Prepare labeled data

[0048] Select the sample underwriting questionnaire corresponding to a specific product. For the sample diseases therein, match and label each level of questions (i.e., the second underwriting questions) corresponding to them with the second general questions corresponding to the same sample diseases in the general underwriting questionnaire.

[0049] As Figure 2 shown, the second underwriting questions include: New_Q1, New_Q2, and the second general questions include: Common_Q1, Common_Q2, Common_Q3, Common_Q4, Common_Q5. "New_Q1 ↔ Common_Q3" means that Common_Q3 can be derived from New_Q1, and vice versa. The relationship between New_Q1 and Common_Q3 is bidirectional implication, that is, the two are completely matched; "New_Q2 → Common_Q4" means that Common_Q4 can be derived from New_Q2, but Common_Q4 cannot be derived from New_Q2 vice versa, indicating that New_Q2 may incorporate other questions in addition to Common_Q4, and the scope of Common_Q4 is smaller. Business personnel need to consider whether to update the content of Common_Q4 according to New_Q2 to make it more generalizable.

[0050] 1.2 Construct training sample data

[0051] Since the BERT sentence pair classification model is adopted, it is necessary to process the labeled data into the input format acceptable to the BERT sentence pair classification model (specifically as shown in Figure 2 ), that is, "[CLS]sentence1[SEP]sentence2[SEP], the label is 1 or 0". Since the results of "whether sentence2 can be deduced from sentence1" and "whether sentence1 can be deduced from sentence2" are different, two labels need to be generated for each question. The label 1 means that sentence2 can be deduced from sentence1, and the label 0 means that sentence2 cannot be deduced from sentence1.

[0052] 1.3 Training the classification model

[0053] As shown in Figure 2 , input the training sample data constructed above into the BERT sentence pair classification model. Based on the difference between the classification result output by the BERT sentence pair classification model and the manually labeled label, fine-tune the model parameters until the loss no longer decreases, and then save the model parameters for use in the prediction stage.

[0054] 2. Prediction stage

[0055] 2.1 Constructing input data

[0056] As shown in Figure 3 , for the newly added underwriting questionnaire, first select the target disease name (for example, disease 1), screen out its corresponding questions at all levels (that is, the first underwriting question), and then find the first general underwriting question corresponding to disease 1 from the general underwriting questionnaire, and combine them into the format described in 1.2.

[0057] 2.2 Classification

[0058] Based on the trained BERT sentence pair classification model, perform binary classification on the input data "[CLS]sentenceA[SEP]sentenceB[SEP]" and "[CLS]sentenceB[SEP]sentenceA[SEP]", and the result is 0 or 1, where sentenceA refers to the first underwriting question and sentenceB refers to the first general question.

[0059] If the classification results of both "[CLS]sentenceA[SEP]sentenceB[SEP]" and "[CLS]sentenceB[SEP]sentenceA[SEP]" are 1, it means that the first underwriting question and the first general question can be deduced from each other and are completely equivalent, that is, there is no need to update the first general question in the general underwriting questionnaire.

[0060] If the classification results of "[CLS]sentenceA[SEP]sentenceB[SEP]" and "[CLS]sentenceB[SEP]sentenceA[SEP]" are 0 and 1 respectively, it means that sentenceB cannot be deduced from sentenceA, but sentenceB can be deduced from sentenceA, and sentenceB has stronger generalization ability, that is, the first general question in the general underwriting questionnaire has a wider coverage. SentenceA can still be mapped to sentenceB, that is, it is not necessary to update the first general question in the general underwriting questionnaire.

[0061] If the classification results of "[CLS]sentenceA[SEP]sentenceB[SEP]" and "[CLS]sentenceB[SEP]sentenceA[SEP]" are 1 and 0 respectively, it means that sentenceB can be deduced from sentenceA, and sentenceA cannot be deduced from sentenceB. Then it is necessary to update the first general question in the general underwriting questionnaire, so as to send a prompt message to prompt the business personnel to consider whether it is necessary to update the first general question in the general underwriting questionnaire to make its generalization ability stronger, or map sentenceA to multiple first general questions.

[0062] In the embodiment of the present application, a sentence pair classification method based on BERT is used to match and map the new questions in the new underwriting questionnaire with the general questions in the general underwriting questionnaire, and judge whether it is necessary to update the general underwriting questionnaire questions based on the results of the sentence pair classification method based on BERT, which can avoid problems such as omission and mapping errors, improve the work efficiency of business personnel, ensure that the general underwriting questionnaire is still a complete and comprehensive questionnaire after adding a new insurance company or insurance product, and ensure that it is applicable to the new insurance company or insurance product.

[0063] As a specific implementation of the above underwriting questionnaire question processing method, the embodiment of the present application provides an underwriting questionnaire question processing device. As Figure 4 shown, the underwriting questionnaire question processing device 400 includes: a questionnaire acquisition module 401, a question acquisition module 402, a classification module 403, and a judgment module 404.

[0064] Among them, the questionnaire acquisition module 401 is used to acquire a new underwriting questionnaire and determine the target disease name in the new underwriting questionnaire; the question acquisition module 402 is used to screen out the first underwriting questions corresponding to the target disease name in the new underwriting questionnaire and search for the first general questions corresponding to the target disease name in the general underwriting questionnaire; the classification module 403 is used to classify the first underwriting questions and the first general questions based on the trained classification model to obtain a classification result; the judgment module 404 is used to judge whether to update the first general questions in the general underwriting questionnaire according to the classification result.

[0065] In this embodiment, for the newly added underwriting questionnaire, first select the target disease name from it, screen out the questions at all levels corresponding to the target disease name (that is, the first underwriting questions), and then find the first general questions corresponding to the target disease name from the general underwriting questionnaire. Further, after converting the formats of the first underwriting questions and the first general questions, input them into the trained classification model to obtain a binary classification result, which is used to indicate whether the semantics of the first underwriting questions and the semantics of the first general questions can be inferred from each other, thereby realizing the matching mapping between the new underwriting questionnaire and the general underwriting questionnaire. Finally, judge whether it is necessary to update the first general questions in the general underwriting questionnaire according to the first underwriting questions according to the classification result.

[0066] In the embodiment of the present application, after integrating a set of general underwriting questionnaires based on the underwriting questions and answers at all levels of diseases in multiple different underwriting questionnaires, in order to assist business personnel in matching and mapping the new underwriting questionnaire with the general underwriting questionnaire and updating the general underwriting questionnaire according to the new underwriting questionnaire, the above-mentioned question matching method for underwriting questionnaires is proposed, which can avoid problems such as omission and mapping errors, improve the work efficiency of business personnel, ensure that the general underwriting questionnaire is still a complete and comprehensive questionnaire after adding a new insurance company or insurance product, and ensure that it is applicable to the new insurance company or insurance product.

[0067] Further, the classification module 403 is specifically used for: converting the formats of the first underwriting questions and the first general questions to generate input data; performing binary classification on the input data based on the trained classification model to obtain a classification result.

[0068] Further, the classification module 403 is specifically used for: converting the formats of the first underwriting questions and the first general questions to generate first input data and second input data; where the first input data is [CLS]sentenceA[SEP]sentenceB[SEP], and the second input data is [CLS]sentenceB[SEP]sentenceA[SEP], sentenceA represents the first underwriting question, and sentenceB represents the first general question.

[0069] Further, the judgment module 404 is specifically configured to: if the classification result corresponding to the first input data is 1 and the classification result of the second input data is 1, or if the classification result corresponding to the first input data is 0 and the classification result of the second input data is 1, then there is no need to update the first general question in the general underwriting questionnaire; if the classification result corresponding to the first input data is 1 and the classification result of the second input data is 0, then the first general question in the general underwriting questionnaire needs to be updated.

[0070] Further, the underwriting questionnaire question processing device 400 further includes: a prompt module, configured to send a prompt message if it is determined that the first general question in the general underwriting questionnaire needs to be updated, and the prompt message is used to prompt the business personnel to update the first general question in the general underwriting questionnaire.

[0071] Further, the underwriting questionnaire question processing device 400 further includes: a model construction module, configured to: obtain a sample underwriting questionnaire and determine the sample diseases in the sample underwriting questionnaire; match and label the second underwriting questions corresponding to the sample diseases in the sample underwriting questionnaire with the second general questions corresponding to the sample diseases in the general underwriting questionnaire to generate labeled data; perform format conversion on the labeled data to generate training sample data; input the training sample data into the BERT sentence pair classification model for model training to construct a classification model.

[0072] Further, the format of the training sample data is "[CLS]sentence1[SEP]sentence2 [SEP], and the label is 1 or 0", where if sentence1 represents the second underwriting question, then sentence2 represents the second general question, and if sentence1 represents the second general question, then sentence2 represents the second underwriting question.

[0073] The underwriting questionnaire question processing device 400 in the embodiments of the present application may be a computer device or a component in a computer device, such as an integrated circuit or a chip. The computer device may be a terminal or other devices other than a terminal. Exemplarily, the computer device may be a mobile phone, a tablet computer, a laptop computer, a handheld computer, a Mobile Internet Device (MID), a robot, an Ultra-Mobile Personal Computer (UMPC), a netbook, or a Personal Digital Assistant (PDA), etc., and may also be a server, a Network Attached Storage (NAS), a Personal Computer (PC), a teller machine, or a self-service machine, etc. The embodiments of the present application do not make specific limitations.

[0074] The underwriting questionnaire question processing device 400 in the embodiments of the present application may be a device with an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems. The embodiments of the present application do not make specific limitations.

[0075] The underwriting questionnaire question processing device 400 provided by the embodiments of the present application can implement Figure 1 each process implemented by the underwriting questionnaire question processing method embodiments. To avoid repetition, it will not be elaborated here.

[0076] The embodiments of the present application further provide a computer device, as Figure 5 shown. The computer device 500 includes a processor 501 and a memory 502. A program or instruction that can run on the processor 501 is stored on the memory 502. When the program or instruction is executed by the processor 501, it implements each step of the above-mentioned underwriting questionnaire question processing method embodiments and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0077] It should be noted that the computer devices in the embodiments of the present application include the above-mentioned mobile computer devices and non-mobile computer devices.

[0078] The memory 502 can be used to store software programs and various data. The memory 502 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. Among them, the first storage area may store an operating system, application programs or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 502 may include a volatile memory or a non-volatile memory, or the memory 502 may include both a volatile memory and a non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synch link dynamic random access memory (SLDRAM), and a direct rambus random access memory (DRRAM). The memory 502 in the embodiments of the present application includes, but is not limited to, these and any other suitable types of memories.

[0079] The processor 501 may include one or more processing units; optionally, the processor 501 integrates an application processor and a modem processor. Among them, the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above modem processor may not be integrated into the processor 501 either.

[0080] The embodiments of the present application further provide a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, it implements each process of the above-mentioned embodiment of the underwriting questionnaire question processing method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0081] An embodiment of the present application also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is configured to run programs or instructions to implement each process of the above-mentioned underwriting questionnaire question processing method embodiment, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0082] It should be understood that the chip mentioned in the embodiment of the present application may also be referred to as a system-on-chip, system chip, chip system, or system-on-chip, etc.

[0083] An embodiment of the present application also provides a computer program product. The program product is stored in a storage medium and is executed by at least one processor to implement each process of the above-mentioned method embodiment, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0084] It should be noted that in this article, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed. It may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0085] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

Claims

1. A method for processing underwriting questionnaire questions, characterized in that Including: Obtain a new underwriting questionnaire and determine the name of the target disease in the new underwriting questionnaire; Screen out the first underwriting questions corresponding to the name of the target disease in the new underwriting questionnaire, and search for the first general questions corresponding to the name of the target disease in the general underwriting questionnaire; Perform format conversion on the first underwriting questions and the first general questions to generate first input data and second input data respectively; Perform binary classification on the first input data and the second input data based on the trained classification model to obtain a classification result; Judge whether to update the first general question in the general underwriting questionnaire according to the classification result; The judging whether to update the first general question in the general underwriting questionnaire according to the classification result includes: If the classification result corresponding to the first input data is 1 and the classification result of the second input data is 1, or if the classification result corresponding to the first input data is 0 and the classification result of the second input data is 1, then there is no need to update the first general question in the general underwriting questionnaire; If the classification result corresponding to the first input data is 1 and the classification result of the second input data is 0, then update the first general question in the general underwriting questionnaire; Wherein, the classification result corresponding to the first input data being 1 means that the first general question can be deduced from the first underwriting question, the classification result corresponding to the first input data being 0 means that the first general question cannot be deduced from the first underwriting question, the classification result corresponding to the second input data being 1 means that the first underwriting question can be deduced from the first general question, and the classification result corresponding to the second input data being 0 means that the first underwriting question cannot be deduced from the first general question.

2. The method according to claim 1, wherein The first input data is [CLS]sentenceA[SEP]sentenceB[SEP], and the second input data is [CLS]sentenceB[SEP]sentenceA[SEP], where sentenceA represents the first underwriting question and sentenceB represents the first general question.

3. The method according to claim 1, wherein After judging whether to update the first general question in the general underwriting questionnaire according to the classification result, it further includes: If it is determined that the first general question in the general underwriting questionnaire needs to be updated, a prompt message is sent, and the prompt message is used to prompt the business personnel to update the first general question in the general underwriting questionnaire.

4. The method according to any one of claims 1 to 3, characterized in that Before obtaining the new underwriting questionnaire, it further includes: Obtain a sample underwriting questionnaire and determine the sample diseases in the sample underwriting questionnaire; Match and label the second underwriting questions corresponding to the sample diseases in the sample underwriting questionnaire with the second general questions corresponding to the sample diseases in the general underwriting questionnaire to generate labeled data; Perform format conversion on the labeled data to generate training sample data; Input the training sample data into the BERT sentence pair classification model for model training to construct the classification model.

5. The method according to claim 4, wherein the format of the training sample data is "[CLS]sentence1[SEP]sentence2[SEP], and the label is 1 or 0", where if sentence1 represents the second underwriting question, then sentence2 represents the second general question, and if sentence1 represents the second general question, then sentence2 represents the second underwriting question.

6. A device for processing underwriting questionnaire questions, characterized in that, It includes: A questionnaire acquisition module, configured to acquire a new underwriting questionnaire and determine a target disease name in the new underwriting questionnaire; A question acquisition module, configured to screen out the first underwriting questions corresponding to the target disease name in the new underwriting questionnaire, and search for the first general questions corresponding to the target disease name in the general underwriting questionnaire; A classification module, configured to perform format conversion on the first underwriting question and the first general question to obtain a first input data and a second input data respectively; And perform binary classification on the first input data and the second input data based on the trained classification model to obtain a classification result; A judgment module, configured to judge whether to update the first general question in the general underwriting questionnaire according to the classification result; The judgment module is specifically configured to: If the classification result corresponding to the first input data is 1 and the classification result of the second input data is 1, or if the classification result corresponding to the first input data is 0 and the classification result of the second input data is 1, then there is no need to update the first general question in the general underwriting questionnaire; If the classification result corresponding to the first input data is 1 and the classification result of the second input data is 0, then update the first general question in the general underwriting questionnaire; Wherein, the classification result corresponding to the first input data being 1 means that the first general question can be deduced from the first underwriting question, the classification result corresponding to the first input data being 0 means that the first general question cannot be deduced from the first underwriting question, the classification result corresponding to the second input data being 1 means that the first underwriting question can be deduced from the first general question, and the classification result corresponding to the second input data being 0 means that the first underwriting question cannot be deduced from the first general question.

7. A computer device, characterized in that, It includes a processor and a memory, and the memory stores a program or instruction that runs on the processor. When the program or instruction is executed by the processor, the steps of the underwriting questionnaire question processing method according to any one of claims 1 to 5 are implemented.

8. A readable storage medium, on which a program or instructions are stored, characterized in that, When the program or instruction is executed by the processor, the steps of the underwriting questionnaire question processing method according to any one of claims 1 to 5 are implemented.

Images