CN115357687A

CN115357687A - Data processing method and device of insurance product and electronic equipment

Info

Publication number: CN115357687A
Application number: CN202211117399.2A
Authority: CN
Inventors: 韩权杰; 杜新凯; 吕超; 谷姗姗; 孙雅琳
Original assignee: Sunshine Insurance Group Co Ltd
Current assignee: Sunshine Insurance Group Co Ltd
Priority date: 2022-09-14
Filing date: 2022-09-14
Publication date: 2022-11-18

Abstract

The invention provides a data processing method and device of an insurance product and electronic equipment, relates to the technical field of insurance product question-answering service, and solves the technical problem that the cost for realizing the insurance product question-answering service is high in the prior art. The method comprises the following steps: acquiring clause data of insurance products; creating a knowledge graph of the insurance product based on the clause data; acquiring a query text from a client, and determining corresponding query content from the query text; determining a query result corresponding to the query content from the knowledge graph based on the query content; and sending the query result to the client.

Description

Data processing method and device for insurance product and electronic equipment

Technical Field

The present application relates to the field of insurance product question-answering service technologies, and in particular, to a data processing method and apparatus for an insurance product, and an electronic device.

Background

Aiming at the question-answering service of insurance products, a plurality of manufacturers develop customer service robots with corresponding scenes based on self-service at present. The robot question-answer architecture of the current insurance product is mainly designed based on frequent-ask questions and answers (faq), similarity is calculated by performing semantic analysis on input intentions of a user, then standard questions similar to the intentions of the user are retrieved from a maintained faq library, and finally answers corresponding to the standard questions are returned to the user.

However, the technical problem of high cost for realizing the question-answering service of the insurance products exists in the prior art.

Disclosure of Invention

The application aims to provide a data processing method and device of an insurance product and electronic equipment, so as to alleviate the technical problem of high cost of implementing the question-answering service of the insurance product in the prior art.

In a first aspect, an embodiment of the present application provides a data processing method for an insurance product, where the method includes:

acquiring clause data of the insurance product;

creating a knowledge-graph of the insurance product based on the clause data;

acquiring a query text from a client, and determining corresponding query content from the query text;

determining a query result corresponding to the query content from the knowledge graph based on the query content;

and sending the query result to the client.

In one possible implementation, the obtaining query text from the client and determining corresponding query content from the query text includes:

acquiring a query text from a client, and extracting keywords of the query text by using a natural language processing technology to obtain the keywords in the query text;

and determining the query content corresponding to the query text based on the keywords.

In one possible implementation, the creating a knowledge-graph of the insurance product based on the clause data includes:

performing data extraction on the clause data to generate a two-dimensional table of the insurance product;

and saving the two-dimensional table as a Comma-Separated Values (CSV) file;

and importing the CSV file into a graph database to generate a knowledge graph of the insurance product.

In one possible implementation, the Graph database is a Nebula Graph database.

In one possible implementation, the determining, from the knowledge-graph, a query result corresponding to the query content based on the query content includes:

determining an initial query result corresponding to the query content from the knowledge graph based on the query content;

and combining the initial query result with a preset conversation template to generate a query result corresponding to the query content.

In one possible implementation, the query content includes any one or more of:

insurance product attribute, insurance product attribute comparison and insurance product classification.

In one possible implementation, the insurance product attributes include any one or more of:

waiting period, hesitation period, insurance guarantee age, payment period and guarantee period.

In a second aspect, an embodiment of the present application provides a data processing apparatus for an insurance product, where the apparatus includes:

the acquisition module is used for acquiring clause data of the insurance product;

a creation module for creating a knowledge graph of the insurance product based on the clause data;

the first determining module is used for acquiring a query text from a client and determining corresponding query content from the query text;

the second determining module is used for determining a query result corresponding to the query content from the knowledge graph based on the query content;

and the sending module is used for sending the query result to the client.

In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory and a processor, where the memory stores a computer program that is executable on the processor, and the processor implements the steps of the method according to the first aspect when executing the computer program.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the steps of the method of the first aspect.

The embodiment of the application brings the following beneficial effects:

the embodiment of the application provides a data processing method and device for an insurance product and electronic equipment. In the scheme, by constructing the insurance product knowledge graph and developing the knowledge graph question-answer service based on various scenes, the question-answer service capable of supporting three scenes, namely insurance product attribute query, insurance product attribute comparison and insurance product classification, can effectively supplement the insurance product question-answer service, greatly relieves the maintenance pressure of customer service staff, and relieves the technical problem of high cost of realizing the insurance product question-answer service in the prior art.

Drawings

In order to more clearly illustrate the detailed description of the present application or the technical solutions in the prior art, the drawings used in the detailed description or the prior art description will be briefly described below, and it is obvious that 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 efforts.

FIG. 1 is a schematic flow chart illustrating data processing of an insurance product according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram of a knowledge-graph-based insurance product question-answering process provided by an embodiment of the present application;

FIG. 3 is a schematic diagram of a data processing structure of an insurance product according to an embodiment of the present disclosure;

fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "comprising" and "having," and any variations thereof, as referred to in the embodiments of the present application, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

For the question-answer service of the insurance products, faq has the defects that a large amount of manpower is needed for maintenance, and the faq is not suitable for the question-answer scene of the insurance products needing logic judgment. The current insurance product question-answering service is mainly realized by constructing faq form. And (3) inputting a product consultation sentence by a client, calculating the similarity between the consultation sentence and the standard questions in the faq knowledge base by the insurance product customer service robot through a semantic similarity algorithm, then recalling the result, finely arranging the recalled standard questions, and finally returning the answers corresponding to the most similar standard questions to the client. Although faq is well applied at present, it is not suitable for solving the problem that logic judgment needs to be performed, such as "do i buy a guaranteed-term medical risk (Huimin version) this year xx", faq cannot be designed for all ages aiming at the problem, and if each faq is designed, the design cost is high. In addition, the insurance products have various attributes, and the space formed by the products and the attributes of the products is huge, so that the maintained faq knowledge base also needs higher labor cost.

From the defects, the technical problem that the cost for realizing the question-answering service of the insurance products is high exists in the prior art.

Based on the above, the embodiment of the application provides a data processing method and device for insurance products and electronic equipment, wherein the established insurance product knowledge graph is used, the open-source distributed graph database supporting clustered deployment is used for storing the insurance product knowledge graph, and finally the application of the insurance product question answering is enabled by developing the knowledge graph question answering service. The method provided by the embodiment of the application can solve the technical problem of high cost of realizing the insurance product question-answering service in the prior art.

Embodiments of the present application are further described below with reference to the accompanying drawings.

Fig. 1 is a schematic flowchart of a data processing method of an insurance product according to an embodiment of the present disclosure. As shown in fig. 1, the method includes:

step S110, acquiring the clause data of the insurance product.

For example, the system may obtain the clause data of the insurance product through external input or online acquisition.

Step S120, based on the clause data, a knowledge graph of the insurance product is created.

For example, the system may perform data extraction on the provision data of the insurance product, creating a knowledge-graph of the insurance product based on the extracted data.

Step S130, obtaining the query text from the client, and determining the corresponding query content from the query text.

For example, as shown in fig. 2, the system can determine the query intention of the user according to the query text input by the user, i.e. the query sentence of the user, such as "how long the guarantee period of the a product i wants to ask" is ", the" shorter waiting period of the two products, i.e. the a product and the B product ", the" which products can be bought by the age 52 ", and so on. The system can analyze the query text and further determine corresponding query content from the query text.

Step S140, based on the query content, determining a query result corresponding to the query content from the knowledge graph.

Illustratively, as shown in fig. 2, in order to return the answer required by the user, the user's consultation intention is required to be converted into a graph query statement to be retrieved in a graph database (a knowledge graph). Because the natural question sometimes contains short names or alias names of insurance products, the natural question can be mapped into the standard names of the insurance products in a knowledge alignment mode, and then graph data query is carried out by using the standard product names to obtain corresponding query results.

Step S150, the inquiry result is sent to the client.

In the embodiment of the application, by constructing the insurance product knowledge graph and developing the knowledge graph question-answer service based on various scenes, the question-answer service of three scenes of insurance product attribute query, insurance product attribute comparison and insurance product classification can be supported to effectively supplement the insurance product question-answer service, so that the maintenance pressure of customer service staff is greatly reduced, and the technical problem of high cost in the prior art of realizing the insurance product question-answer service is solved.

The above steps are described in detail below.

In some embodiments, the step S130 may specifically include the following steps:

step a), acquiring a query text from a client, and extracting keywords from the query text by using a natural language processing technology to obtain the keywords in the query text.

And b), determining the query content corresponding to the query text based on the keywords.

For example, as shown in fig. 2, the system first parses the query text of the user, and parses the query text by using a natural language processing technique, to determine which of three major scenes, namely product attribute query, product attribute comparison and product classification, the intention of the user for consultation belongs to. The question can be analyzed in a rule-based mode by using the attribute keywords, and meanwhile, the question can also be analyzed in a text word vector mode. After parsing, question classification may be performed, and if a rule-based form is employed, a mapping of attribute keywords to attributes is constructed. If the guarantee period is a key word of the insurance product attribute, the classification of the question "how long the guarantee period of the product A is asked for by me" is the attribute query of the guarantee period in the insurance product attribute query scene. The "waiting period" is a keyword of the "waiting period" of the insurance product attribute, and the classification of the question "shorter waiting period of the two products, product a and product B" is the insurance product attribute comparison, namely the waiting period in the insurance product attribute comparison scene. The "which products can be bought by the age of 52" corresponds to the scenario that the guaranteed age is 52 years under the product classification scenario.

The query text is obtained from the client, and the keywords in the query text are obtained by extracting the keywords from the query text by using a natural language processing technology, so that the query content corresponding to the query text is determined based on the keywords, and the content which the user wants to consult can be determined more accurately.

In some embodiments, the step S120 may specifically include the following steps:

and c), performing data extraction on the clause data to generate a two-dimensional table of the insurance product.

And step b), saving the two-dimensional table as a CSV file.

And d), importing the CSV file into a database to generate a knowledge graph of the insurance product.

Illustratively, the insurance product knowledge graph construction designs a corresponding knowledge graph schema for product attributes by analyzing insurance product clause data, and the insurance product attributes may include more than 50 attributes such as waiting period, hesitation period, insurance age, payment period, guarantee period, and the like. And then extracting corresponding product attributes from insurance clauses by using a text analysis tool to generate a two-dimensional table of the insurance product attributes, storing the two-dimensional table of the insurance product as a CSV file, and importing the two-dimensional table of the insurance product into a database in batches to form a knowledge map.

Based on step d) above, the graph database may include multiple types, thereby making the functionality richer. As one example, the Graph database is a Nebula Graph database.

Exemplarily, in order to support the subsequent large-scale development and application of the insurance product knowledge Graph, an open-source, distributed and easily-expanded native Graph database Nebula Graph database can be used as a knowledge Graph storage engine, can bear a super-large-scale data set with hundreds of millions of nodes and hundreds of millions of edges, and provides millisecond-level query. In order to support the attribute batch storage of the insurance products, the attributes of the insurance products can be extracted by using a text analysis tool and stored in the CSV file. And then, a Nebula importer tool is used for importing the CSV files into a Nebula Graph database in batches.

In some embodiments, the step S140 may specifically include the following steps:

and e), determining an initial query result corresponding to the query content from the knowledge graph based on the query content.

And f), combining the initial query result with a preset conversation template to generate a query result corresponding to the query content.

For example, the system may retrieve attribute values, attribute comparison results, or results of product categorization from a graph database (knowledge graph) based on previously determined graph database query statements, i.e., query content. If the corresponding content is inquired, namely the returned result is not empty, constructing a corresponding dialect template and splicing the returned result, and returning the processed result to the user; if the corresponding content is not inquired, namely the returned result is empty, the question cannot be answered by using the current knowledge graph, and therefore the question can be returned to the user through a preset linguistics.

The system determines an initial query result corresponding to the query content from the knowledge graph based on the query content, and combines the initial query result with a preset dialect template to generate a query result corresponding to the query content. Corresponding information can be returned to the user based on different query results, so that the user can clearly know the answers of the content consulted by the user, and the use experience of the user is improved.

In some embodiments, the query content may include multiple types, so that the function is richer, and the query of more content can be realized through the knowledge graph. As one example, the query content includes any one or more of:

Illustratively, "if the user asks a question: "how long the guarantee period of the product A is asked for by me", the classification of the question is the attribute query of the guarantee period under the insurance product attribute query scene. If the user asks a question: "the waiting time of the products A and B is shorter", the classification of the question is the insurance product attribute comparison of waiting time in the insurance product attribute comparison scene. If the user asks a question: "what products can be bought by age 52" the category corresponding to the question is the scene of saving age 52 in the product classification scene.

The types of the query contents include multiple types, so that the functions are richer, the query contents can be classified more accurately according to the query texts of the user, more accurate answers are provided, and the user experience is improved.

In some embodiments, the insurance product attributes include any one or more of:

Illustratively, by including insurance product attributes in multiple categories, functionality can be enriched. For example, the attribute of the insurance product is set as the keyword, so that the system can accurately judge the content to be queried by the user according to the keyword after analyzing the query text of the user, further provide a corresponding answer, and improve the use experience of the user.

Fig. 3 is a schematic structural diagram of a data processing apparatus of an insurance product according to an embodiment of the present application. As shown in fig. 3, the data processing device 300 of the insurance product includes:

an obtaining module 301, configured to obtain clause data of an insurance product;

a creation module 302 for creating a knowledge graph of the insurance product based on the clause data;

a first determining module 303, configured to obtain a query text from a client, and determine corresponding query content from the query text;

a second determining module 304, configured to determine, based on the query content, a query result corresponding to the query content from the knowledge graph;

a sending module 305, configured to send the query result to the client.

In some embodiments, the first determining module 303 is specifically configured to:

acquiring a query text from a client, and extracting keywords from the query text by using a natural language processing technology to obtain keywords in the query text;

In some embodiments, the creation module 302 is specifically configured to:

performing data extraction on the receipt data to generate a two-dimensional table of insurance products;

saving the two-dimensional table as a CSV file;

In some embodiments, the Graph database is a Nebula Graph database.

In some embodiments, the second determining module 304 is specifically configured to:

and combining the initial query result with a preset dialect template to generate a query result corresponding to the query content.

In some embodiments, the query content includes any one or more of:

The device provided by the embodiment of the present invention has the same implementation principle and technical effect as the method embodiments, and for the sake of brief description, no mention is made in the system embodiments, and reference may be made to the corresponding contents in the method embodiments.

The embodiment of the invention provides electronic equipment, which particularly comprises a processor and a storage device, wherein the processor is used for processing a plurality of data files; the storage means has stored thereon a computer program which, when executed by the processor, performs the method of any of the above embodiments.

Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, where the electronic device includes: a processor 401, a memory 402, a bus 403 and a communication interface 404, the processor 401, the communication interface 404 and the memory 402 being connected by the bus 403; the processor 401 is adapted to execute executable modules, such as computer programs, stored in the memory 402.

The Memory 402 may include a Random Access Memory (RAM) and a Non-volatile Memory (Non-volatile Memory), such as at least one disk Memory. The communication connection between the network element of the system and at least one other network element is realized through at least one communication interface 404 (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, and the like can be used.

Bus 403 can be an ISA bus, PCI bus, EISA bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one double-headed arrow is shown in FIG. 4, but that does not indicate only one bus or one type of bus.

The memory 402 is used for storing a program, the processor 401 executes the program after receiving an execution instruction, and the method performed by the apparatus defined by the flow program disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 401, or implemented by the processor 401.

The processor 401 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 401. The Processor 401 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, etc. as is well known in the art. The storage medium is located in the memory 402, and the processor 401 reads the information in the memory 402, and combines the hardware thereof to complete the steps of the method.

The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer readable storage medium storing a program code, where instructions included in the program code may be used to execute the method in the foregoing method embodiment, and for specific implementation, reference may be made to the foregoing method embodiment, which is not described herein again.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.

Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method of data processing for an insurance product, the method including:

acquiring clause data of the insurance product;

creating a knowledge-graph of the insurance product based on the clause data;

and sending the query result to the client.

2. The method of claim 1, wherein obtaining query text from a client and determining corresponding query content from the query text comprises:

acquiring a query text from a client, and extracting keywords of the query text by using a natural language processing technology to obtain keywords in the query text;

3. The method of claim 1, wherein creating a knowledge-graph of the insurance product based on the clause data comprises:

storing the two-dimensional table as a CSV file;

4. The method according to claim 3, wherein the Graph database is a Nebula Graph database.

5. The method of claim 1, wherein determining the query result corresponding to the query content from the knowledge-graph based on the query content comprises:

6. The method of claim 1, wherein the query content comprises any one or more of:

7. The method of claim 6, wherein the insurance product attributes comprise any one or more of:

8. A data processing apparatus for an insurance product, the apparatus comprising:

the second determination module is used for determining a query result corresponding to the query content from the knowledge graph based on the query content;

and the sending module is used for sending the query result to the client.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program operable on the processor, and wherein the processor implements the steps of the method of any of claims 1 to 7 when executing the computer program.

10. A computer readable storage medium having stored thereon computer executable instructions which, when invoked and executed by a processor, cause the processor to execute the method of any of claims 1 to 7.