CN112927091B - Complaint early warning method and device for annual gold insurance, computer equipment and medium - Google Patents
Complaint early warning method and device for annual gold insurance, computer equipment and medium Download PDFInfo
- Publication number
- CN112927091B CN112927091B CN202110377125.6A CN202110377125A CN112927091B CN 112927091 B CN112927091 B CN 112927091B CN 202110377125 A CN202110377125 A CN 202110377125A CN 112927091 B CN112927091 B CN 112927091B
- Authority
- CN
- China
- Prior art keywords
- complaint
- insurance
- life
- policy
- annual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 title abstract description 32
- 229910052737 gold Inorganic materials 0.000 title abstract description 32
- 239000010931 gold Substances 0.000 title abstract description 32
- 230000008451 emotion Effects 0.000 claims abstract description 62
- 238000004590 computer program Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 238000007621 cluster analysis Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000009897 systematic effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000010485 coping Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 238000013019 agitation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the application provides a complaint early warning method, a device, computer equipment and a medium for annual gold risk, wherein the method comprises the following steps: acquiring client access condition information and emotion information during client access according to policy data and client information of annual insurance to be pre-warned; dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be pre-warned, and judging whether the current time of the annuity insurance policy to be pre-warned corresponds to the life nodes of the annuity insurance policy; and inputting the client access condition information, emotion information during client access and condition information of a life node of which whether the current time of the annual insurance to be early-warned corresponds to the annual insurance or not into an early-warning model, and outputting policy information predicted as complaints by the early-warning model. The scheme can more objectively predict complaints, is not influenced by subjective factors, improves the efficiency of complaint early warning, can provide systematic complaint prediction for business scenes with huge quantity of insurance policies, and is also beneficial to improving the accuracy of complaint early warning.
Description
Technical Field
The application relates to the technical field of insurance, in particular to a complaint early warning method, a complaint early warning device, computer equipment and a medium for annual gold insurance.
Background
With the development of society, the right-to-right awareness of consumers is gradually enhanced. Handling customer complaints and bad reviews has gained attention in different industries. Insurance is different from other products, and as a heterogeneous product, complaints are various in expression form, and the treatment of complaints is quite complex. Customer complaint demands affect customer service experience first, cognition of products, and social reputation of companies second. More seriously, complaints of which an effective negotiation scheme cannot be obtained can cause customers to complain to the silver insurance prison, and excessive complaint amount can promote the supervision risk of the company.
In a general business scenario, complaints of clients are more direct, and clients directly present written or oral objections, claims and other behaviors. The insurance person has limited coping time, the complaint coping ability is related to personal working experience, if the personal experience is insufficient, the customer emotion cannot be calmed in a short time, and a reasonable solution is provided, the negative influence of the complaint may be enlarged. Thus, in a short period of time, the complaint treatment effect is a big test of the internal workplace industry experience.
The early warning work of complaints is perfected, and the preparation work of possibly occurring complaint cases is well done, so that important influence is brought to the positive response of the complaint cases. The prior complaint early warning work mainly depends on artificial subjective judgment, lacks objective judgment standards, has higher prediction cost, and is difficult to ensure the prediction efficiency and accuracy. In addition, depending on the complaint of the artificially predicted insurance policy, the method has certain feasibility under the condition of limited insurance policy quantity, but in the actual business scene, the insurance policy quantity is huge and the insurance policy states are different (such as the states of insurance application, continuous insurance, refund insurance, insurance and the like), and the manually prediction cannot provide systematic prediction. In conclusion, the insurance early warning work needs to be perfected.
Disclosure of Invention
The embodiment of the application provides a complaint early warning method for annual gold risk, which aims to solve the technical problems of low efficiency, poor accuracy and high cost of complaint early warning in the prior art. The method comprises the following steps:
acquiring client access condition information and emotion information during client access according to policy data and client information of annual insurance to be pre-warned;
dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be pre-warned, and judging whether the current time of the annuity insurance policy to be pre-warned corresponds to the life nodes of the annuity insurance policy;
inputting client access condition information, emotion information during client access and condition information of life nodes of annual insurance to be pre-warned about whether the current time of the annual insurance corresponds to the annual insurance or not into a pre-warning model, and outputting insurance policy information predicted as complaints by the pre-warning model, wherein the pre-warning model is obtained by training samples according to the client access condition information of the annual insurance history insurance policy, the emotion information during client access, the complaint condition and the condition information of the life nodes of the annual insurance about whether the complaint time corresponds to the annual insurance or not.
The embodiment of the application also provides a complaint early warning device for annual gold insurance, which aims to solve the technical problems of low efficiency, poor accuracy and high cost of the complaint early warning in the prior art. The device comprises:
the information acquisition module is used for acquiring client access condition information and emotion information during client access according to the policy data and the client information of the annual insurance to be early-warned;
the time determining module is used for dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be early-warned, and judging whether the current time of the annuity insurance policy to be early-warned corresponds to the life node of the annuity insurance policy;
the early warning module is used for inputting customer access condition information, emotion information during customer access and condition information of whether the current time of the annual insurance to be early warned corresponds to the annual insurance life node or not into the early warning model, and outputting insurance policy information predicted as complaints by the early warning model, wherein the early warning model is obtained by training samples according to the customer access condition information of the annual insurance historical insurance policy, emotion information during customer access, complaint condition and condition information of whether the complaint time corresponds to the annual insurance life node or not.
The embodiment of the application also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program is executed by the processor to realize any complaint early warning method for annual insurance, so as to solve the technical problems of low efficiency, poor accuracy and high cost in the prior art of complaint early warning.
The embodiment of the application also provides a computer readable storage medium which stores a computer program for executing any complaint early warning method for annual gold risk, so as to solve the technical problems of low efficiency, poor accuracy and high cost in the prior art of complaint early warning.
In the embodiment of the application, the early warning model is obtained by training a sample based on the client access condition information of the annuity insurance history policy, the emotion information during client access, the complaint condition and the condition information of whether the complaint time corresponds to the life node of the annuity insurance, further, the client access condition information and the emotion information during client access are obtained according to the policy data and the client information of the annuity insurance to be early warned, the life cycle of the annuity insurance policy is divided into different life nodes according to the staged characteristics of the policy data of the annuity insurance to be early warned, whether the current time of the annuity insurance to be early warned corresponds to the life node of the annuity insurance is judged, and finally, inputting the client access condition information of the annual insurance to be pre-warned, affective information during client access and condition information of a life node of the annual insurance to be pre-warned or not, outputting insurance policy information predicted as complaints by the pre-warning model, namely predicting the insurance policy where the complaints will occur by the pre-warning model, and outputting the insurance policy information, wherein compared with the technical scheme of manually and subjectively predicting the complaints in the prior art, the insurance policy where the complaints will occur is predicted by the pre-warning model driven by data, so that the complaints can be predicted more objectively, the influence of the subjective factors is avoided, the efficiency of complaint pre-warning can be improved, the complaint prediction of a system can be provided for a business scene with huge quantity of the insurance policy, the accuracy of complaint pre-warning is also facilitated to be improved, and reference data is provided for the advance preparation work corresponding to the complaints; meanwhile, because whether the states of the corresponding life nodes are different or not can influence the probability of complaint generation, the customer access condition information can reflect the attention degree of customers to the insurance policy, and the emotion information during customer access can reflect the customer dissatisfaction degree and the complaint possibility, the early warning model is provided for carrying out complaint early warning by taking the information of whether the current time corresponds to the life nodes of annual insurance, the customer access condition information and the emotion information during customer access as indexes, so that the accuracy of complaint early warning is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate and together with the description serve to explain the application. In the drawings:
FIG. 1 is a flow chart of a complaint early warning method for annuity risk provided by an embodiment of the application;
FIG. 2 is a flowchart of a method for early warning complaints for annuity according to an embodiment of the present application;
FIG. 3 is a block diagram of a computer device according to an embodiment of the present application;
fig. 4 is a block diagram of a complaint early warning device for annuity risk according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following embodiments and the accompanying drawings, in order to make the objects, technical solutions and advantages of the present application more apparent. The exemplary embodiments of the present application and the descriptions thereof are used herein to explain the present application, but are not intended to limit the application.
In an embodiment of the present application, a complaint early warning method for annuity risk is provided, as shown in fig. 1, the method includes:
step 102: acquiring client access condition information and emotion information during client access according to policy data and client information of annual insurance to be pre-warned;
step 104: dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be pre-warned, and judging whether the current time of the annuity insurance policy to be pre-warned corresponds to the life nodes of the annuity insurance policy;
step 106: inputting client access condition information, emotion information during client access and condition information of life nodes of annual insurance to be pre-warned about whether the current time of the annual insurance corresponds to the annual insurance or not into a pre-warning model, and outputting insurance policy information predicted as complaints by the pre-warning model, wherein the pre-warning model is obtained by training samples according to the client access condition information of the annual insurance history insurance policy, the emotion information during client access, the complaint condition and the condition information of the life nodes of the annual insurance about whether the complaint time corresponds to the annual insurance or not.
As can be seen from the flow shown in fig. 1, in the embodiment of the application, the client access condition information of the annual gold insurance history policy, affective information during client access, complaint conditions and condition information of the life node of the annual gold insurance are provided as sample training to obtain an early warning model, further, the client access condition information and affective information during client access are obtained according to the policy data and client information of the annual gold insurance to be early warned, the life cycle of the annual gold insurance policy is divided into different life nodes according to the staged characteristics of the policy data of the annual gold insurance to be early warned, whether the current time of the annual gold insurance to be early warned corresponds to the life node of the annual gold insurance is judged, finally, the condition information of the client access condition information of the annual gold insurance to be early warned, the affective information during client access and the condition information of the life node of the annual gold insurance to be early warned are input into an early warning model, the early warning model outputs the policy information predicted to be complaint, namely, the policy to be complaint is predicted to be complaint by the early warning model, and the policy information is predicted to be complaint, compared with the prior art, the policy is predicted by the early warning model, the prediction system is more accurate than the policy, and the service is predicted to have a great influence on the life node of the service by the prediction policy, and the service is more predicted by the aspect of the prediction policy, compared with the prior art, and the service is more predicted by the prediction policy; meanwhile, because whether the states of the corresponding life nodes are different or not can influence the probability of complaint generation, the customer access condition information can reflect the attention degree of customers to the insurance policy, and the emotion information during customer access can reflect the customer dissatisfaction degree and the complaint possibility, the early warning model is provided for carrying out complaint early warning by taking the information of whether the current time corresponds to the life nodes of annual insurance, the customer access condition information and the emotion information during customer access as indexes, so that the accuracy of complaint early warning is improved.
In specific implementation, the early warning model is obtained by training a sample with the client access condition information of the annual gold insurance history policy, the emotion information of the client access, the complaint condition (divided into complaint and non-complaint) and the condition information of whether the complaint time corresponds to the annual gold insurance life node, and specifically, the early warning model can be obtained by training a neural network, a two-class classifier and other machine learning components.
In practice, the present inventors have found that, unlike other dangerous products, the annual insurance product clauses define the responsibility of each policy year, and cover the time of the responsibility of the claim in addition to payment and security. For example, for products such as accident insurance, medical insurance, life insurance, etc., the specific time of accident occurrence cannot be known, so the exact time of occurrence of claim settlement cannot be given, and annual insurance is taken as a special insurance product, the periodic characteristics of the insurance policy data are strong, for example, the time of occurrence of responsibility such as payment, claim settlement, etc. has corresponding relation with the insurance policy, and in the life cycle of the insurance policy, when the insurance policy is just in important nodes such as payment or claim settlement, the probability of complaint is high, therefore, in order to further combine the periodic characteristics of the annual insurance, the accuracy of early warning is improved by referring to time-specific information, in this embodiment, the life cycle of the insurance policy of the annual insurance is divided into different life nodes according to the time-period and the responsibility corresponding to different time-periods of the insurance policy data of the annual insurance policy to be early warned.
Specifically, the policy can be divided into a plurality of different life nodes according to time sequence according to different responsibilities (namely service states and service contents) corresponding to different time periods in the policy, so as to monitor whether the policy is in different life nodes to predict complaints.
In particular implementations, life nodes of the policy lifecycle partition for annuity insurance may include validation nodes, payment nodes, and claim settlement nodes. Specifically, the number of the life nodes of each class may be one or more, for example, there may be one effective node, multiple payment nodes and multiple claim settlement nodes in the life cycle of a policy.
In particular, in order to accurately predict complaints, in this embodiment, it is required to accurately determine whether the current time corresponds to a life node, and in order to facilitate subsequent determination of a cause of complaints, it may also be determined which life node the current time corresponds to, for example, whether the current time corresponds to a life node, and which life node the current time corresponds to may be determined by the following formula:
the closer the current time is to the nearest policy life node, the larger the above formula value is, so that when the maximum value of the above formula is larger than a preset threshold value for each life node, the current time of the instruction corresponds to the life node, and specifically, the life node corresponding to the maximum value of the above formula is the life node of the annual gold risk corresponding to the current time. Wherein t is the current time; l (L) k,i The time may be in months for the ith life node in the kth class of life nodes. Through the method, the node state of the current policy corresponding to the three different life nodes of the life node, the payment node and the claim settlement node can be obtained. If let t=t+Δt, it is possible to obtain whether the current policy corresponds to the state of the policy node in a future period of time (Δt).
In the embodiment, in order to further accurately predict complaints, to fully reflect the customer access condition information and reflect the attention degree of customers to the policy, the customer access condition information may be customer access frequency, customer access duration, change condition of customer access frequency, etc., where the customer access condition information in the embodiment is the change condition of customer access frequency in a preset duration unit, for example, the change condition of customer access frequency in a week unit, whether the customer access frequency is increased or decreased, if the customer access frequency is increased, it is indicated that the customer attention degree to the policy is higher, and there may be problems of related business objections, etc., and these factors are likely to generate complaints; if the customer access frequency becomes smaller, the customer is less concerned about the policy, and there may be no objection or appeal to the relevant service, and the probability of complaints is small in these cases.
In the implementation, the number of times of client access can be obtained through inserting codes of products such as a client access insurance system or an application program, and further the change condition of the client access frequency is obtained by taking a preset duration as a unit.
In the specific implementation, in order to further accurately predict complaints, the emotion state of the client at the time of access is fully reflected through emotion information, and in the embodiment, according to the interaction information at the time of client access, the emotion information at the time of client access is acquired.
Specifically, the emotion information of the client during access can be analyzed and obtained through interactive information such as text, voice and the like of online and offline customer service conversations during the client access. For example, when a customer accesses negative emotion information such as agitation, anger, gas generation and the like, the customer is not satisfied, and the probability of complaints is high; when the clients have positive emotion information such as like, good, very excellent and the like during access, the clients are satisfied, and the probability of complaints is very low.
In specific implementation, the inventor discovers that each of three information, namely the access condition, the emotion information and the relation between the current time of the policy and the life node, can be fed back in real time to show whether a customer has a possibility of complaint or not. Specifically, the early warning model can predict the probability of complaints of each policy in a period of time in the future, and a clear early warning effect is realized in the future, and can output the policies with the probability of complaints larger than a preset value so as to early warn the policies and further implement related intervention work, and can consider that early warning or intervention is not needed at present for policies with the probability of complaints smaller than the preset value, and the early warning model can not output the policies.
In the practical business scenario, even if a customer complaint cause is not known, a customer service worker can not find an entry point easily when visiting the customer, the intervention effect of possibly generating a complaint case is not obvious, and the problems of manpower waste and low communication efficiency with the customer are also caused to a certain extent, so that the intervention work before the complaint can be pertinently and directionally developed so as to effectively improve the intervention effect, further reference information is provided for the policy management work and provide convenience, in the embodiment, support is provided for the complaint coping work, for example, the complaint cause of each historical policy is obtained by carrying out cluster analysis on customer access condition information corresponding to the complaint in the sample, emotion information corresponding to the complaint in the customer access, and life node type corresponding to the complaint time, and the complaint condition is customer access condition information corresponding to the complaint, emotion information corresponding to the complaint time and node type corresponding to the complaint time; and aiming at the insurance policy predicted as complaint by the early warning model, comparing the customer access condition information of the insurance policy, the emotion information of the customer when accessing, the life node type corresponding to the current time and the data in the complaint data set, obtaining and outputting the complaint reason of the insurance policy. The method not only can predict the possible insurance policy of complaints in a period of time (such as a week, a month, a quarter or a half year, etc.), but also can analyze and output possible complaint reasons so as to provide specific guidance for complaint intervention work, can be used for solving complaint early warning work before important complaint nodes, further achieves the effects of having complaints with a specific case and having no complaint key return visit, is beneficial to improving the service quality to a certain extent, and has a certain guiding significance for the fine operation of insurance company clients and maintenance of company reputation.
In specific implementation, KMeans cluster analysis can be performed on customer access condition information corresponding to a policy of complaints in the samples, emotion information during customer access and life node types corresponding to complaint time, so as to obtain complaint reasons.
In this embodiment, a process of implementing the complaint early warning method for annuity risk is described in detail with reference to fig. 2, and as shown in fig. 2, the method includes the following steps:
the first step, based on the history policy data and the customer information, a history data set is provided, the data with complaints records corresponds to complaints Y being 1, and the data without complaints records corresponds to complaints Y being 0.
And secondly, sorting the historical data set and extracting relevant information.
Policy period:
(1) according to different responsibilities corresponding to different time periods in the historical policy year, dividing the life cycle of each year of the policy into a plurality of different life nodes according to the time sequence, and determining whether the complaint time corresponds to the life node and the life node type corresponding to the complaint time. The life node may include an effective node, a payment node, and a claim settlement node. Specifically, whether the complaint time corresponds to the life node and the life node type corresponding to the complaint time can be determined according to the correspondence and intersection conditions of the complaint time and the time of the life node, and whether the complaint time corresponds to the life node and the life node type corresponding to the complaint time can be determined by the following formula:
the closer the complaint time is to the nearest policy node, the larger the above equation value. Wherein, t is complaint time at this time; l (L) k,i For the ith life node in the kth class of life nodes,and when the maximum value of the formula is larger than a preset threshold value, the complaint time corresponds to the life node, and the life node corresponding to the maximum value is the life node of the annual gold risk corresponding to the current time.
(2) Access condition information:
the number of client accesses can be obtained through products such as a terminal, an application program and the like which are logged in by the user. For example, the change in the access frequency of the client is acquired in units of weeks.
(3) Customer emotion information:
and acquiring a voice file and a text file of a conversation with online and offline customer service when a customer accesses, and performing emotion analysis, for example, determining emotion information when the customer accesses through semantic analysis and other technologies.
Thirdly, inputting life nodes, customer access condition information and emotion information when the customers access with the complaint time corresponding to the annual insurance or not as main index information into a classifier, and performing classification training to obtain the early warning model.
Fourth, a sample with complaint condition of 1 is extracted, KMeans cluster analysis is carried out on the customer access condition information, emotion information during customer access, and information such as life node types corresponding to the determined complaint time, which are extracted from the sample, so that complaint reasons of each historical policy are obtained, and a complaint data set is formed by the customer access condition information, emotion information during customer access, life node types corresponding to the complaint time and corresponding complaint reasons of the sample with complaint condition of 1.
Fifthly, in practical application, inputting customer access condition information of annual insurance to be pre-warned, emotion information during customer access and condition information of life node of annual insurance to be pre-warned, wherein the pre-warning model outputs insurance policy information predicted as complaints (for example, the insurance policy information can comprise insurance policy numbers, insurance policy names and the like), predicts insurance policies predicted as complaints for the pre-warning model, and further inputs the customer access condition information of the insurance policy, emotion information during customer access and the condition information corresponding to the current time and the number of life node types and complaint data sets in the current timeAnd (5) according to the comparison, obtaining possible complaint reasons of the policy and outputting the possible complaint reasons. For example, the current time of the annuity risk to be early-warned is in a renewal state (L Duration of time ) (namely, the current time corresponds to the payment node), the client access condition information is high access frequency (V), and the emotion information at the client access is negative emotion (S) N ) The information reflects that the annuity risk customer group to be pre-warned has a strong negative emotion to the problem related to the renewal, and at the moment, the pre-warning model can accurately predict a possible complaint policy and output a possible complaint reason based on information such as customer access condition information, emotion information when the customer accesses and condition information of a life node corresponding to the annuity risk at the current time of the annuity risk to be pre-warned, and the like, so that a list of the policy and the complaint reason can be issued, specific complaint intervention work can be done in advance, and complaint is reduced or avoided. If the early warning model does not output the policy information and the complaint reasons, namely, the policy possibly complaining in a future period is not predicted currently, the related data of the policy is continuously monitored.
In this embodiment, a computer device is provided, as shown in fig. 3, including a memory 302, a processor 304, and a computer program stored in the memory and capable of running on the processor, where the processor implements any of the foregoing complaint early warning methods for annual gold risk when executing the computer program.
In particular, the computer device may be a computer terminal, a server or similar computing means.
In the present embodiment, a computer-readable storage medium storing a computer program that executes any of the above-described complaint early-warning methods for annual gold risk is provided.
In particular, computer-readable storage media, including both permanent and non-permanent, removable and non-removable media, may be used to implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable storage media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
Based on the same inventive concept, the embodiment of the application also provides a complaint early warning device for annuity risk, as described in the following embodiment. The principle of solving the problem by the complaint early-warning device for the annuity risk is similar to that of the complaint early-warning method for the annuity risk, so that the implementation of the complaint early-warning device for the annuity risk can be referred to the implementation of the complaint early-warning method for the annuity risk, and repeated parts are omitted. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements the intended function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
FIG. 4 is a block diagram of a complaint warning device for annuity risk according to an embodiment of the present application, as shown in FIG. 4, the device includes:
the information acquisition module 402 is configured to acquire client access condition information and emotion information during client access according to policy data and client information of annual insurance to be pre-warned;
the time determining module 404 is configured to divide the life cycle of the insurance policy of the annuity into different life nodes according to the staged feature of the insurance policy data of the annuity to be early-warned, and determine whether the current time of the annuity to be early-warned corresponds to the life node of the annuity;
the complaint early warning module 406 is configured to input customer access condition information, emotion information during customer access, and condition information of whether a current time of a to-be-early-warned annuity insurance corresponds to an annuity insurance life node into an early warning model, and output policy information predicted as complaints by the early warning model, where the early warning model is obtained by training samples based on the customer access condition information of the annuity insurance history policy, emotion information during customer access, complaint condition, and condition information of whether the complaint time corresponds to the annuity insurance life node.
In one embodiment, the time determining module is further configured to divide the life cycle of the policy of the annuity insurance into different life nodes according to the time stage of the policy data of the annuity insurance to be pre-warned and responsibilities corresponding to different time stages.
In one embodiment, the life nodes include an validation node, a payment node, and a claims node.
In one embodiment, further comprising:
the life node determining module is used for calculating the life node of the annuity risk corresponding to the current time through the following formula:
wherein t is the current time; l (L) k,i And when the maximum value of the formula is larger than a preset threshold value, the life node corresponding to the maximum value is the life node of the annual risk corresponding to the current time.
In one embodiment, the client access condition information is a change condition of a client access frequency in units of a preset duration.
In one embodiment, further comprising:
the cluster analysis module is used for carrying out cluster analysis on customer access condition information corresponding to a complaint policy, emotion information during customer access and life node types corresponding to complaint time in the sample to obtain complaint reasons of each historical policy, wherein the complaint conditions comprise customer access condition information corresponding to the complaint policy, emotion information during customer access, life node types corresponding to complaint time and corresponding complaint reasons to form a complaint data set;
the complaint cause determining module is used for comparing the customer access condition information of the complaint, the emotion information of the customer when accessing and the life node type corresponding to the current time with the data in the complaint data set to obtain and output the complaint cause of the complaint aiming at the warrant predicted by the early warning model.
In one embodiment, the cluster analysis module is specifically configured to perform KMeans cluster analysis on customer access condition information corresponding to a policy with complaints in the sample, emotion information during customer access, and a life node type corresponding to complaint time.
The embodiment of the application realizes the following technical effects: the method comprises the steps of providing client access condition information based on an annual gold insurance policy, affective information during client access, complaint conditions and condition information of life nodes of annual gold insurance or not, training samples to obtain an early warning model, further obtaining client access condition information and affective information during client access according to the annual gold insurance policy data to be early-warned and the client information, dividing the annual gold insurance policy life cycle into different life nodes according to the staged characteristics of the annual gold insurance policy data to be early-warned, judging whether the current time of the annual gold insurance to be early-warned corresponds to the life nodes of the annual gold insurance or not, finally, inputting the annual gold insurance policy information to be early-warned, the affective information during client access and the condition information of the life nodes of the annual gold insurance to be early-warned or not according to the early-warning model, outputting the insurance policy information predicted to be complaint, namely predicting the insurance policy to be complained through the model, and outputting the insurance policy information compared with the technical scheme of the prior art that the insurance policy to be early-warned is subjective and the insurance policy, predicting the current time of the annual gold insurance policy is provided by driving the early-warning data to be early-warned, and the service is more accurate than the early-warning system is provided, and the early-warning system is more convenient to predict the service and has a large number of the insurance policy to be early-complaint, and can be more important than the service and has the service and is more predicted to be early for the service and has the service and to be early-complaint; meanwhile, because whether the states of the corresponding life nodes are different or not can influence the probability of complaint generation, the customer access condition information can reflect the attention degree of customers to the insurance policy, and the emotion information during customer access can reflect the customer dissatisfaction degree and the complaint possibility, the early warning model is provided for carrying out complaint early warning by taking the information of whether the current time corresponds to the life nodes of annual insurance, the customer access condition information and the emotion information during customer access as indexes, so that the accuracy of complaint early warning is improved.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.
Claims (7)
1. The complaint early warning method for annuity risk is characterized by comprising the following steps:
acquiring client access condition information and emotion information during client access according to policy data and client information of annual insurance to be pre-warned;
dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be pre-warned, and judging whether the current time of the annuity insurance policy to be pre-warned corresponds to the life nodes of the annuity insurance policy; the life node comprises an effective node, a payment node and a claim settlement node;
inputting client access condition information, emotion information during client access and condition information of life nodes of annual insurance to be early-warned whether the current time of the annual insurance corresponds to the annual insurance or not into an early warning model, and outputting insurance policy information predicted as complaints by the early warning model, wherein the early warning model is obtained by taking the client access condition information of the annual insurance history insurance policy, the emotion information during client access, the complaint condition and the condition information of the life nodes of the annual insurance whether the complaint time corresponds to the annual insurance or not as sample training;
dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be pre-warned, comprising:
dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the time stage of the insurance policy data of the annuity insurance policy to be pre-warned and the responsibilities corresponding to different time stages;
determining whether the current time corresponds to a life node of the annual risk by the following formula:
wherein t is the current time; l (L) k,i And when the maximum value of the formula is larger than a preset threshold value, the life node corresponding to the maximum value is the life node of the annual risk corresponding to the current time.
2. The complaint warning method for annuity of claim 1, further comprising:
the client access condition information is the change condition of the client access frequency with the preset time length as a unit.
3. The complaint early warning method for annuity risk of any one of claims 1 to 2, further comprising:
clustering analysis is carried out on customer access condition information corresponding to a complaint policy, emotion information during customer access and life node types corresponding to complaint time in a sample to obtain complaint reasons of each historical policy, wherein the complaint conditions comprise customer access condition information corresponding to the complaint policy, emotion information during customer access, life node types corresponding to complaint time and corresponding complaint reasons to form a complaint data set;
and aiming at the insurance policy predicted as complaint by the early warning model, comparing the customer access condition information of the insurance policy, the emotion information of the customer when accessing, the life node type corresponding to the current time and the data in the complaint data set, obtaining and outputting the complaint reason of the insurance policy.
4. The complaint early warning method for annuity risk according to claim 3, wherein the clustering analysis of the customer access condition information corresponding to the policy in which the complaint condition is complaint, the emotion information at the time of customer access, and the life node type corresponding to the complaint time in the sample includes:
and carrying out KMeans cluster analysis on customer access condition information corresponding to a complaint policy, emotion information during customer access and life node types corresponding to complaint time in the sample.
5. A complaint early warning device for annuity risk, comprising:
the information acquisition module is used for acquiring client access condition information and emotion information during client access according to the policy data and the client information of the annual insurance to be early-warned;
the time determining module is used for dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the staged characteristics of the insurance policy data of the annuity insurance policy to be early-warned, and judging whether the current time of the annuity insurance policy to be early-warned corresponds to the life node of the annuity insurance policy; the life node comprises an effective node, a payment node and a claim settlement node;
the early warning module is used for inputting customer access condition information, emotion information during customer access and condition information of whether the current time of the annual insurance to be early warned corresponds to the annual insurance life node or not into an early warning model, and outputting insurance policy information predicted as complaints by the early warning model, wherein the early warning model is obtained by training samples according to the customer access condition information of the annual insurance historical insurance policy, emotion information during customer access, complaint condition and condition information of whether the complaint time corresponds to the annual insurance life node or not;
the time determining module is specifically configured to: dividing the life cycle of the insurance policy of the annuity insurance into different life nodes according to the time stage of the insurance policy data of the annuity insurance policy to be pre-warned and the responsibilities corresponding to different time stages;
determining whether the current time corresponds to a life node of the annual risk by the following formula:
wherein t is the current time; l (L) k,i And when the maximum value of the formula is larger than a preset threshold value, the life node corresponding to the maximum value is the life node of the annual risk corresponding to the current time.
6. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the complaint early warning method for annual risk according to any one of claims 1 to 4 when the computer program is executed by the processor.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program that executes the complaint early-warning method for annuity risk of any one of claims 1 to 4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110377125.6A CN112927091B (en) | 2021-04-08 | 2021-04-08 | Complaint early warning method and device for annual gold insurance, computer equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110377125.6A CN112927091B (en) | 2021-04-08 | 2021-04-08 | Complaint early warning method and device for annual gold insurance, computer equipment and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112927091A CN112927091A (en) | 2021-06-08 |
CN112927091B true CN112927091B (en) | 2023-11-10 |
Family
ID=76173725
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110377125.6A Active CN112927091B (en) | 2021-04-08 | 2021-04-08 | Complaint early warning method and device for annual gold insurance, computer equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112927091B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116760652B (en) * | 2023-08-23 | 2023-11-17 | 保大坊科技有限公司 | Method, apparatus and storage medium for simultaneously accessing multiple systems |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106529804A (en) * | 2016-11-09 | 2017-03-22 | 国网江苏省电力公司南京供电公司 | Client complaint early-warning monitoring analyzing method based on text mining technology |
CN110349039A (en) * | 2019-06-13 | 2019-10-18 | 中国平安人寿保险股份有限公司 | Complaint risk appraisal procedure, system, computer equipment and readable storage medium storing program for executing |
CN110516836A (en) * | 2019-07-05 | 2019-11-29 | 中国平安财产保险股份有限公司 | Intelligent early-warning method, apparatus, equipment and storage medium based on big data |
CN110796554A (en) * | 2019-09-06 | 2020-02-14 | 中国平安财产保险股份有限公司 | User complaint early warning method and device, computer equipment and storage medium |
CN110889770A (en) * | 2019-10-12 | 2020-03-17 | 中国平安财产保险股份有限公司 | Data processing method, device, equipment and storage medium |
CN111612230A (en) * | 2020-05-13 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | Client appeal trend early warning analysis method |
CN111639102A (en) * | 2020-06-01 | 2020-09-08 | 阳光保险集团股份有限公司 | Client data resource sharing method and device and electronic equipment |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8725614B2 (en) * | 2011-01-19 | 2014-05-13 | Financial Engines, Inc. | Creating and maintaining a payout-ready portfolio within an investment plan to generate a sustainable income stream |
US11113704B2 (en) * | 2015-12-07 | 2021-09-07 | Daniel J. Towriss | Systems and methods for interactive annuity product services using machine learning modeling |
-
2021
- 2021-04-08 CN CN202110377125.6A patent/CN112927091B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106529804A (en) * | 2016-11-09 | 2017-03-22 | 国网江苏省电力公司南京供电公司 | Client complaint early-warning monitoring analyzing method based on text mining technology |
CN110349039A (en) * | 2019-06-13 | 2019-10-18 | 中国平安人寿保险股份有限公司 | Complaint risk appraisal procedure, system, computer equipment and readable storage medium storing program for executing |
CN110516836A (en) * | 2019-07-05 | 2019-11-29 | 中国平安财产保险股份有限公司 | Intelligent early-warning method, apparatus, equipment and storage medium based on big data |
CN110796554A (en) * | 2019-09-06 | 2020-02-14 | 中国平安财产保险股份有限公司 | User complaint early warning method and device, computer equipment and storage medium |
CN110889770A (en) * | 2019-10-12 | 2020-03-17 | 中国平安财产保险股份有限公司 | Data processing method, device, equipment and storage medium |
CN111612230A (en) * | 2020-05-13 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | Client appeal trend early warning analysis method |
CN111639102A (en) * | 2020-06-01 | 2020-09-08 | 阳光保险集团股份有限公司 | Client data resource sharing method and device and electronic equipment |
Non-Patent Citations (3)
Title |
---|
基于随机森林算法的投诉预警模型优化方法;朱龙珠 等;电力信息与通信技术(第08期);第60-65页 * |
质监大数据企业画像的研究及应用;吴行惠 等;数字通信世界(第02期);第204-208页 * |
转型时期寿险公司风险管控研究;唐庚荣;中国博士学位论文全文数据库 经济与管理科学辑(第10期);第1-147页 * |
Also Published As
Publication number | Publication date |
---|---|
CN112927091A (en) | 2021-06-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10719767B2 (en) | Computer implemented system for automating the generation of a business decision analytic model | |
US10671926B2 (en) | Method and system for generating predictive models for scoring and prioritizing opportunities | |
US10706359B2 (en) | Method and system for generating predictive models for scoring and prioritizing leads | |
CN111539585B (en) | Random forest-based power customer appeal sensitivity supervision and early warning method | |
CN110515811A (en) | Terminal artificial intelligence performance benchmark test method and device | |
CN113256969B (en) | Traffic accident early warning method, device and medium for expressway | |
ElBahrawy et al. | Wikipedia and cryptocurrencies: interplay between collective attention and market performance | |
CN112598443A (en) | Online channel business data processing method and system based on deep learning | |
CN111222994A (en) | Client risk assessment method, device, medium and electronic equipment | |
CN114676316A (en) | Method and system for constructing enterprise employee portrait based on big data | |
CN112927091B (en) | Complaint early warning method and device for annual gold insurance, computer equipment and medium | |
CN110163470B (en) | Event evaluation method and device | |
CN116302835A (en) | Fortune dimension abnormity detection device, method and storage medium | |
CN110175113A (en) | Business scenario determines method and apparatus | |
CN112712270B (en) | Information processing method, device, equipment and storage medium | |
CN113962656A (en) | Power grid data asset management method, system, equipment and storage medium | |
CN114547406A (en) | Data monitoring method, system, storage medium and electronic device | |
Gupta et al. | Forecasting Customer Churn in the Telecommunications Industry | |
López et al. | Application of a Data Mining Model to Predict Customer Defection. Case of a Telecommunications Company in Peru | |
CN117236715A (en) | Method, interaction system, equipment and storage medium for generating risk decision strategy | |
CN118212002A (en) | Customer life cycle early warning method, equipment and storage medium | |
CN118378138A (en) | Cross-platform scheduling method and device for shutdown operation flow and readable storage medium | |
Chen et al. | Semiconductor Price Index Predicting Based on a Novel Improved AdaBoost Feature-Weighted Combination Model | |
CN118041662A (en) | Network data security detection method and device | |
CN118429075A (en) | Prediction processing method, device and equipment for credit card default users |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |