CN116523662A

CN116523662A - Prediction method and device based on artificial intelligence, computer equipment and storage medium

Info

Publication number: CN116523662A
Application number: CN202310505396.4A
Authority: CN
Inventors: 江黄波; 龚亭如
Original assignee: Ping An Technology Shenzhen Co Ltd
Current assignee: Ping An Technology Shenzhen Co Ltd
Priority date: 2023-05-06
Filing date: 2023-05-06
Publication date: 2023-08-01

Abstract

The embodiment of the application belongs to the field of artificial intelligence, and relates to an artificial intelligence-based prediction method, which comprises the following steps: acquiring historical policy data of a target client; acquiring policy data of a temporary policy from the historical policy data; extracting target feature data from the policy data based on a preset feature extraction algorithm; inputting the target feature data into a pre-trained renewal prediction model, predicting the target feature data through the renewal prediction model, and outputting a corresponding renewal prediction result; and determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result. The application also provides a prediction device, computer equipment and a storage medium based on artificial intelligence. In addition, the present application relates to blockchain techniques in which the renewal prediction probabilities may be stored. The method and the device realize quick and accurate prediction of the continuous guarantee will of the target client by utilizing the pre-trained continuous guarantee prediction model.

Description

Prediction method and device based on artificial intelligence, computer equipment and storage medium

Technical Field

The present disclosure relates to the field of artificial intelligence development, and in particular, to an artificial intelligence-based prediction method, apparatus, computer device, and storage medium.

Background

With the popularity of insurance services, more and more individuals or companies choose to purchase insurance products to provide more assurance of future uncertainty. For an insurance company, after the insurance product purchased by the insurance user expires, the insurance company may desire to conduct continuous insurance on the insurance user, and even if the insurance user has a wish to conduct continuous insurance, the insurance company may desire to conduct personalized customization on the user so as to enable continuous insurance.

At present, for a customer renewal prediction mode, customer renewal will is usually predicted according to personal experience of insurance marketers, and the personal experience of different insurance marketers has differences, so that the manual judgment mode has the problems of lower accuracy and instability.

Disclosure of Invention

The embodiment of the application aims to provide an artificial intelligence-based prediction method, an artificial intelligence-based prediction device, computer equipment and a storage medium, so as to solve the technical problems that the accuracy is low and the artificial judgment mode is unstable because the personal experiences of different insurance marketers are different in the conventional client renewal prediction mode, namely, the client renewal will is predicted according to the personal experiences of the insurance marketers.

In order to solve the above technical problems, the embodiments of the present application provide an artificial intelligence based prediction method, which adopts the following technical scheme:

acquiring historical policy data of a target client;

acquiring policy data of a temporary policy from the historical policy data;

extracting target feature data from the policy data based on a preset feature extraction algorithm;

inputting the target feature data into a pre-trained renewal prediction model, predicting the target feature data through the renewal prediction model, and outputting a corresponding renewal prediction result;

and determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result.

Further, the step of extracting the target feature data from the policy data based on the preset feature extraction algorithm specifically includes:

performing feature extraction on the policy data based on a preset first feature extraction algorithm to obtain corresponding first feature data;

performing feature extraction on the policy data based on a preset second feature extraction algorithm to obtain corresponding second feature data;

performing feature extraction on the policy data based on a preset third feature extraction algorithm to obtain corresponding third feature data;

The target feature data is constructed based on the first feature data, the second feature data, and the third feature data.

Further, the step of constructing the target feature data based on the first feature data, the second feature data, and the third feature data specifically includes:

carrying out data combination on the first characteristic data, the second characteristic data and the third characteristic data to obtain a corresponding characteristic data set;

performing data matching on all the characteristic data in the characteristic data set, and screening fourth characteristic data repeatedly appearing in the characteristic data set;

and taking the fourth characteristic data as the target characteristic data.

Further, before the step of inputting the target feature data into a pre-trained renewal prediction model, performing prediction processing on the target feature data by using the renewal prediction model and outputting a renewal prediction result corresponding to the target feature data, the method further comprises:

acquiring pre-collected policy sample data;

preprocessing the policy sample data to obtain processed policy sample data;

dividing the processed policy sample data into a training set and a testing set according to a preset proportion;

Acquiring a pre-constructed initial prediction model;

training the initial prediction model by using the training set to obtain a trained initial prediction model;

verifying the trained initial prediction model by using the test set;

and if the trained initial renewal prediction model passes the verification, taking the trained initial prediction model as the renewal prediction model.

Further, after the step of determining the renewal prediction probability of the target client for the temporary policy based on the renewal prediction result, the method further includes:

acquiring a preset probability threshold;

numerical comparison is carried out on the renewal prediction probability and the probability threshold value to obtain a corresponding comparison result;

and generating a client type corresponding to the target client based on the comparison result.

acquiring specified data corresponding to a specified data type from the historical policy data based on a preset specified data type;

based on the renewal prediction probability and the specified data, a preset first calculation formula is called to calculate a future value score corresponding to the target client;

Generating a composite value score for the target customer based on the future value score.

Further, the step of generating the composite value score of the target client based on the future value score specifically includes:

acquiring a target client state of the target client;

determining a target value calculation formula corresponding to the target client state;

based on target parameters contained in the target value calculation formula, calling a preset second calculation formula to calculate a target data value corresponding to the target parameters;

substituting the future value score and the target data value into the target value calculation formula to perform numerical calculation to obtain a corresponding calculation result;

and taking the calculation result as the comprehensive value score of the target client.

In order to solve the above technical problems, the embodiments of the present application further provide an artificial intelligence based prediction apparatus, which adopts the following technical scheme:

the first acquisition module is used for acquiring historical policy data of the target client;

the second acquisition module is used for acquiring the policy data of the temporary policy from the history policy data;

the extraction module is used for extracting target feature data from the policy data based on a preset feature extraction algorithm;

The prediction module is used for inputting the target characteristic data into a pre-trained renewal prediction model, predicting the target characteristic data through the renewal prediction model and outputting a corresponding renewal prediction result;

and the first generation module is used for determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result.

In order to solve the above technical problems, the embodiments of the present application further provide a computer device, which adopts the following technical schemes:

acquiring historical policy data of a target client;

acquiring policy data of a temporary policy from the historical policy data;

In order to solve the above technical problems, embodiments of the present application further provide a computer readable storage medium, which adopts the following technical solutions:

Acquiring historical policy data of a target client;

acquiring policy data of a temporary policy from the historical policy data;

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

firstly, acquiring historical policy data of a target client; then acquiring the policy data of the temporary policy from the history policy data; then extracting target feature data from the policy data based on a preset feature extraction algorithm; inputting the target characteristic data into a pre-trained renewal prediction model, predicting the target characteristic data through the renewal prediction model, and outputting a corresponding renewal prediction result; and finally, determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result. According to the method and the device for predicting the renewal of the target client, the target characteristic data are extracted from the policy data of the temporary policy of the target client, and then the target characteristic data are processed by utilizing the renewal prediction model trained in advance so as to rapidly and accurately predict the renewal intention of the target client, so that the renewal prediction probability of the corresponding target client is generated, the insurance marketer is facilitated to carry out matched insurance business follow-up processing on the target client according to the obtained renewal prediction probability, the intelligence of the insurance business follow-up processing of the target client is improved, and the customer renewal rate is facilitated to be improved.

Drawings

For a clearer description of the solution in the present application, a brief description will be given below of the drawings that are needed in the description of the embodiments of the present application, it being obvious that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of one embodiment of an artificial intelligence based prediction method according to the present application;

FIG. 3 is a schematic diagram of one embodiment of an artificial intelligence based predictive device according to the application;

FIG. 4 is a schematic structural diagram of one embodiment of a computer device according to the present application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description and claims of the present application and in the description of the figures above are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to better understand the technical solutions of the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture Experts Group Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.

It should be noted that, the prediction method based on artificial intelligence provided in the embodiments of the present application is generally executed by a server/terminal device, and accordingly, the prediction apparatus based on artificial intelligence is generally disposed in the server/terminal device.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow chart of one embodiment of an artificial intelligence based prediction method according to the present application is shown. The prediction method based on artificial intelligence comprises the following steps:

Step S201, obtain the historical policy data of the target client.

In this embodiment, the electronic device (e.g., the server/terminal device shown in fig. 1) on which the prediction method based on artificial intelligence operates may acquire the historical policy data of the target client through a wired connection manner or a wireless connection manner. It should be noted that the wireless connection may include, but is not limited to, 3G/4G/5G connection, wiFi connection, bluetooth connection, wiMAX connection, zigbee connection, UWB (ultra wideband) connection, and other now known or later developed wireless connection. Historical policy data for the target customer may be looked up from a pre-built policy database based on the identification information (e.g., name) of the target customer.

Step S202, acquiring the policy data of the clinical policy from the history policy data.

In this embodiment, the term policy refers to a policy that the duration of the insurance contract is about to expire. The policy data at least comprises data such as insurance category of purchased insurance, name of purchased insurance, premium paid, insurance marketer, premium total, customer personal asset information, customer age information, number of claims of purchased insurance.

Step S203, extracting target feature data from the policy data based on a preset feature extraction algorithm.

In this embodiment, the specific implementation process of extracting the target feature data from the policy data based on the preset feature extraction algorithm is described in further detail in the following specific embodiments, which will not be described herein.

Step S204, inputting the target feature data into a pre-trained renewal prediction model, performing prediction processing on the target feature data through the renewal prediction model, and outputting a corresponding renewal prediction result.

In this embodiment, the renewal prediction result at least includes renewal prediction probability. The foregoing process of generating the continuous predictive model will be described in further detail in the following embodiments, which are not described herein.

And step S205, determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result.

In this embodiment, data corresponding to the probability value may be extracted from the renewal prediction result output by the renewal prediction model, and the data corresponding to the probability value may be used as the renewal prediction probability of the target client for the temporary policy.

Firstly, acquiring historical policy data of a target client; then acquiring the policy data of the temporary policy from the history policy data; then extracting target feature data from the policy data based on a preset feature extraction algorithm; inputting the target characteristic data into a pre-trained renewal prediction model, predicting the target characteristic data through the renewal prediction model, and outputting a corresponding renewal prediction result; and finally, determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result. According to the method and the system, the target characteristic data are extracted from the policy data of the temporary policy of the target client, and then the target characteristic data are processed by utilizing the pre-trained renewal prediction model to rapidly and accurately predict the renewal intention of the target client, so that the renewal prediction probability of the corresponding target client is generated, the insurance marketer is facilitated to carry out matched insurance business follow-up processing on the target client according to the obtained renewal prediction probability, the processing intelligence of the insurance business follow-up of the target client is improved, and the client renewal rate is facilitated to be improved.

In some alternative implementations, step S203 includes the steps of:

and carrying out feature extraction on the policy data based on a preset first feature extraction algorithm to obtain corresponding first feature data.

In this embodiment, the first feature extraction algorithm may specifically be a feature extraction algorithm corresponding to a correlation coefficient, and may include, for example, a pearson correlation coefficient, a spearman correlation coefficient, and the like.

And carrying out feature extraction on the policy data based on a preset second feature extraction algorithm to obtain corresponding second feature data.

In this embodiment, the second feature extraction algorithm may specifically be an mRMR (Max-Redundancy and Min-Redundancy) algorithm, and feature extraction is performed on policy data by using the mRMR algorithm, so that correlation between model input and model output is maximized, redundancy between model inputs is minimized, and further, a follow-up prediction model may have better prediction accuracy when performing prediction processing on target feature data.

And carrying out feature extraction on the policy data based on a preset third feature extraction algorithm to obtain corresponding third feature data.

In this embodiment, the third feature extraction algorithm may specifically be a principal component analysis algorithm. The principal component analysis is a multivariate statistical method which converts a plurality of indexes into a plurality of comprehensive indexes on the premise of losing little information by using the thought of dimension reduction. The resulting composite of transformations is often referred to as principal components, where each principal component is a linear combination of the original variables and the individual principal components are uncorrelated with each other, which results in some superior performance of the principal component over the original variables. Therefore, when complex problems are studied, only a few main components can be considered without losing too much information, so that main contradictions are easier to grasp, regularity among internal variables of things is revealed, meanwhile, the problems are simplified, and analysis efficiency is improved.

In this embodiment, the specific implementation process of constructing the target feature data based on the first feature data, the second feature data and the third feature data will be described in further detail in the following specific embodiments, which will not be described herein.

According to the method, feature extraction is carried out on the policy data based on a preset first feature extraction algorithm, so that corresponding first feature data are obtained; and carrying out feature extraction on the policy data based on a preset second feature extraction algorithm to obtain corresponding second feature data; the target feature data is subsequently constructed based on the first feature data and the second feature data. According to the method and the device, the first feature data, the second feature data and the third feature data obtained by the feature extraction algorithm are further combined and matched, so that the target feature data related to continuous prediction is finally determined from the warranty data of the temporary warranty, and the accuracy and the effectiveness of the generated target feature data are effectively guaranteed. In addition, the obtained target characteristic data keeps the characteristic data with stronger correlation with the continuous prediction processing, so that noise data in the characteristic space can be effectively deleted, the saturation recognition modeling error is reduced, and the interference effect of noise on the model is reduced. And the dimension of input data of the model is reduced, so that the processing time required by the prediction processing of the follow-up predictive model is reduced, and the prediction processing efficiency of the follow-up predictive model is effectively improved.

In some optional implementations of the present embodiment, the constructing the target feature data based on the first feature data, the second feature data, and the third feature data includes:

and carrying out data combination on the first feature data, the second feature data and the third feature to obtain a corresponding feature data set.

In this embodiment, the data merging means that the first feature data, the second feature data, and the third feature data are merged and summarized, so as to obtain a feature data set including the first feature data, the second feature data, and the third feature data.

And carrying out data matching on all the characteristic data in the characteristic data set, and screening out fourth characteristic data repeatedly appearing in the characteristic data set.

In this embodiment, the data matching specifically may refer to similarity calculation processing between feature data, by calculating similarity between any plurality of feature data, if the calculated similarity is greater than a preset similarity threshold, it is determined that the plurality of feature data are feature data that match each other, that is, the plurality of feature data belong to the same feature data that repeatedly appears in the feature data set. In addition, the value of the similarity threshold is not particularly limited, and may be set according to actual requirements.

And taking the fourth characteristic data as the target characteristic data.

The method comprises the steps of carrying out data combination on the first characteristic data, the second characteristic data and the third characteristic data to obtain a corresponding characteristic data set; and then carrying out data matching on all the characteristic data in the characteristic data set, screening out fourth characteristic data repeatedly appearing in the characteristic data set, and taking the fourth characteristic data as the target characteristic data. According to the method and the device, the first feature data, the second feature data and the third feature data obtained by the feature extraction algorithm are further combined and matched, so that the target feature data related to continuous prediction is finally determined from the warranty data of the temporary warranty, and the accuracy and the effectiveness of the generated target feature data are effectively guaranteed. In addition, the obtained target characteristic data keeps the characteristic data with stronger correlation with the continuous prediction processing, so that noise data in the characteristic space can be effectively deleted, the saturation recognition modeling error is reduced, and the interference effect of noise on the model is reduced. And the dimension of input data of the model is reduced, so that the processing time required by the prediction processing of the follow-up predictive model is reduced, and the prediction processing efficiency of the follow-up predictive model is effectively improved.

In some alternative implementations, before step S204, the electronic device may further perform the following steps:

and acquiring pre-collected policy sample data.

In this embodiment, the policy sample data refers to specific policy data collected in advance for model training. When specific policy data is acquired, the policy can be extracted from dangerous seeds of various categories so as to improve the prediction accuracy of a machine learning model which is trained later. Wherein, at least, the policy sample data comprises: the type of insurance purchased, the name of insurance purchased, the premium paid, the insurance marketer, the premium total, the customer personal asset information, the customer age information, the number of claims for insurance purchased by the customer, etc.

And preprocessing the policy sample data to obtain processed policy sample data.

In this embodiment, the preprocessing of the policy sample data refers to the feature extraction of the policy sample data by using the feature extraction algorithm, and a specific processing procedure may refer to the extraction procedure of the target feature data, which is not described herein.

Dividing the processed policy sample data into a training set and a testing set according to a preset proportion.

In this embodiment, a preset percentage of data may be randomly extracted from the processed policy sample data to be used as a training set, and then the remaining other data in the processed policy sample data may be used as a test set, or (1-preset percentage) of data may also be randomly extracted from the processed policy sample data to be used as a test set. In addition, the above-mentioned preset ratio is not particularly limited, and may be set according to actual requirements, for example, the preset ratio may be set to 7: and 3, randomly extracting 70% of data from the processed policy sample data to serve as a training set, and randomly extracting 30% of data from the processed policy sample data to serve as a test set.

And obtaining a pre-constructed initial prediction model.

In this embodiment, the initial prediction model may be a machine learning model. The algorithm used in the machine learning model is not limited, and for example, a random forest algorithm or an Xgboost algorithm may be used, and preferably an Xgboost algorithm may be used. The Xgboost algorithm is an integrated machine learning algorithm based on a decision tree, and adopts a gradient boosting (GradientBoosting) framework, and can be used for classifying or regressing through inputting parameter data.

And training the initial prediction model by using the training set to obtain a trained initial prediction model.

In this embodiment, the training set may be used to train the initial prediction model by a search algorithm (e.g., a random search algorithm or a grid search algorithm) so that the loss function reaches an expected value, so that a trained initial prediction model may be generated. The specific training process for the initial prediction model may refer to the existing model training process, which is not described herein.

And verifying the trained initial prediction model by using the test set.

In this embodiment, when the initial prediction model is input by using the test set and the corresponding continuous prediction result meets the preset prediction accuracy, it is indicated that the performance of the initial prediction model is better. The value of the prediction accuracy is not particularly limited.

The method comprises the steps of obtaining pre-collected policy sample data, and preprocessing the policy sample data to obtain processed policy sample data; dividing the processed policy sample data into a training set and a testing set according to a preset proportion; then, acquiring a pre-built initial prediction model, and training the initial prediction model by using the training set to obtain a trained initial prediction model; subsequently, verifying the trained initial prediction model by using the test set; and if the trained initial renewal prediction model passes the verification, taking the trained initial prediction model as the renewal prediction model. According to the method and the device, the required renewal prediction model is quickly and accurately trained and generated by using the pre-collected policy sample data, so that target feature data corresponding to the input policy data of the temporary policy can be intelligently predicted by using the renewal prediction model, a renewal prediction result of a target client for the temporary policy is obtained, and the renewal prediction processing efficiency and the processing intelligence of the target user are improved.

In some alternative implementations, after step S205, the electronic device may further perform the following steps:

and acquiring a preset probability threshold.

In this embodiment, the value of the probability threshold is not specifically limited, and may be set according to actual use requirements.

And carrying out numerical comparison on the renewal prediction probability and the probability threshold value to obtain a corresponding comparison result.

In this embodiment, the comparison result may include that the renewal prediction probability of the target client is greater than or equal to the probability threshold, or that the renewal prediction probability of the target client is less than the probability threshold.

In this embodiment, if the renewal prediction probability of the target client is greater than or equal to the probability threshold, the client type of the target client belonging to the renewing intention client is generated, and if the renewal prediction probability of the target client is less than the probability threshold, the client type of the target client belonging to the non-renewing intention client is generated. In addition, a plurality of probability thresholds can be set, and the renewal will of the clients is divided into a plurality of grades, so that the marketing efficiency of insurance marketers is further improved.

The method comprises the steps of obtaining a preset probability threshold value; then, carrying out numerical comparison on the renewal prediction probability and the probability threshold value to obtain a corresponding comparison result; and generating a client type corresponding to the target client based on the comparison result. According to the method and the device, the probability threshold is set, so that the client with higher renewal will can be selected by comparing the renewal prediction probability with the probability threshold, and the insurance marketer can be facilitated to make a corresponding insurance marketing strategy according to the screening result, so that the working efficiency of the insurance marketer is improved.

In some optional implementations of this embodiment, after step S205, the electronic device may further perform the following steps:

and acquiring the specified data corresponding to the specified data type from the historical policy data based on the preset specified data type.

In this embodiment, the specified data type specifically refers to the profit of the customer's annual insurance requirement. The profit brought by the one-year insurance requirement of the customer can be calculated according to the current profit summation of the insurance policy and the profit of the same insurance policy, which is not ensured by the customer, and the conversion rate.

And calling a preset first calculation formula to calculate a future value score corresponding to the target client based on the renewal prediction probability and the specified data.

In this embodiment, the first calculation formula is a calculation formula related to the renewal prediction probability and the specified data, which is preset according to an actual service usage requirement. The first calculation formula specifically includes: future value score = customer annual insurance demand brings profit x renewal prediction probability.

In this embodiment, the foregoing implementation process of generating the composite value score of the target client based on the future value score will be described in further detail in the following embodiments, which will not be described herein.

Acquiring specified data corresponding to a specified data type from the historical policy data based on a preset specified data type; then, based on the renewal prediction probability and the specified data, a preset first calculation formula is called to calculate a future value score corresponding to the target client; a composite value score for the target customer is then generated based on the future value score. According to the method and the system, the specified data and the renewal prediction probability are calculated and processed based on the first calculation formula, the future value score of the target customer can be rapidly generated, the comprehensive value score of the target customer can be rapidly and accurately generated based on the future value score, the follow-up profit of the customer, which can be objectively evaluated according to the obtained comprehensive value score of the customer, can be further quantitatively guided, the resource is put in, the corresponding output strategy is generated, and the follow-up processing intelligence of the insurance business of the customer is improved.

In some optional implementations of the present embodiment, the generating the composite value score for the target client based on the future value score includes the steps of:

and acquiring the target client state of the target client.

In this embodiment, the states of clients may be categorized because the corresponding value structures are different for different client states. Customer status may include guaranteed, unsecured, potential customers. And for the clients in different client states, the calculation formulas which are preset according to the actual service use requirements and respectively correspond to the comprehensive value scores of the clients in different client states are calculated. Specifically, the composite value score of the insured customer = a1 x historical profit score + b1 x current value score + c1 x potential value score + d1 x future value score; composite value score for the deshielded customer = a2 x historical profit score + c2 x potential value score + d2 x future value score; composite value score for potential customer = c3 potential value score + d3 future value score. The values of a, b, c and d can be weight parameters set according to actual use requirements, are not particularly limited, and can be respectively seen according to different client states, and reference values are respectively given according to distribution quantiles to obtain the client value scores of the client in each stage. In addition, the same kind of clients: customer groups with similar factors such as customer industry, enterprise properties, scale and the like; customer potential demand products: the similar clients of the enterprise have high permeability; customer potential demand product profit: reasonable profit of high permeability products of similar clients of enterprises; some dangerous line conversion rate: the conversion rate of various dangerous varieties of the clients can be predicted based on the similar client permeability and modeling based on the client basic information, premium scale and the like; future M years renewal rate: the renewal rate can be predicted based on average future M years renewal rate statistics of similar clients, and also based on client basic information and historical underwriting behavior modeling.

And determining a target value calculation formula corresponding to the target client state.

In this embodiment, the target value calculation formula specifically refers to a calculation formula of a composite value score corresponding to the guaranteeing client. The target value calculation formula specifically comprises: composite value score at the insured customer = a1 x historical profit score + b1 x current value score + c1 x potential value score + d1 x future value score.

And calling a second preset calculation formula to calculate a target data value corresponding to the target parameter based on the target parameter contained in the target value calculation formula.

In this embodiment, the objective parameters specifically refer to the historical profit score, the current value score, and the potential value score. Wherein the client policy profit value can be calculated by the group in the last N years. Based on premium, pay, fee, share ratio, etc., the actual profit value of each policy of the customer is calculated, and the calculation logic is different for different stages of customer value. Specifically, historical value = sum of late N years' expiration policy profit; current value = current warranty pre-estimated expiration warranty profit sum; potential value = customer potential demand product profit value × conversion; future value = comprehensively considering the one-year insurance requirement of the customer to bring profit × the M-year renewal rate in the future.

Substituting the future value score and the target data value into the target value calculation formula to perform numerical calculation, and obtaining a corresponding calculation result.

In this embodiment, according to the similar calculation logic, the value of each dangerous line of the client, the average client value of the similar client group, and the average client value of each dangerous line of the similar client group may be calculated by dividing the lines. In addition, the obtained comprehensive value score of the client can be multiplied by factors such as a cooperation period coefficient, a product width coefficient, an industry coefficient and the like according to the operation and management requirements so as to obtain more accurate result data.

The method comprises the steps of obtaining a target client state of the target client; then determining a target value calculation formula corresponding to the target client state; then, based on target parameters contained in the target value calculation formula, a preset second calculation formula is called to calculate a target data value corresponding to the target parameters; and substituting the future value score and the target data value into the target value calculation formula to carry out numerical calculation, obtaining a corresponding calculation result, and taking the calculation result as the comprehensive value score of the target client. According to the method and the system, the target value calculation formula corresponding to the target client state of the target client is used, the comprehensive value score of the target client is quickly and accurately generated based on the future value score, the subsequent objective evaluation of long-term profits of the client according to the obtained comprehensive value score of the client is facilitated, the resource release is further quantitatively guided, the corresponding output strategy generation is performed, and the processing intelligence of follow-up of insurance business of the client is improved.

It should be emphasized that, to further ensure the privacy and security of the renewal prediction probabilities, the renewal prediction probabilities may also be stored in nodes of a blockchain.

The blockchain referred to in the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by computer readable instructions stored in a computer readable storage medium that, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

With further reference to fig. 3, as an implementation of the method shown in fig. 2, the present application provides an embodiment of an artificial intelligence-based prediction apparatus, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus is particularly applicable to various electronic devices.

As shown in fig. 3, the artificial intelligence based prediction apparatus 300 according to the present embodiment includes: a first acquisition module 301, a second acquisition module 302, an extraction module 303, a prediction module 304, and a first generation module 305. Wherein:

a first obtaining module 301, configured to obtain historical policy data of a target client;

a second obtaining module 302, configured to obtain policy data of a temporary policy from the historical policy data;

an extracting module 303, configured to extract target feature data from the policy data based on a preset feature extraction algorithm;

the prediction module 304 is configured to input the target feature data to a pre-trained renewal prediction model, perform prediction processing on the target feature data through the renewal prediction model, and output a corresponding renewal prediction result;

the first generation module 305 is configured to determine a renewal prediction probability of the target client for the temporary policy based on the renewal prediction result.

In this embodiment, the operations performed by the above modules or units are respectively corresponding to the steps of the artificial intelligence-based prediction method in the foregoing embodiment, and are not described herein again.

In some alternative implementations of the present embodiment, the extracting module 303 includes:

the first extraction sub-module is used for carrying out feature extraction on the policy data based on a preset first feature extraction algorithm to obtain corresponding first feature data;

the second extraction sub-module is used for carrying out feature extraction on the policy data based on a preset second feature extraction algorithm to obtain corresponding second feature data;

the third extraction sub-module is used for carrying out feature extraction on the policy data based on a preset third feature extraction algorithm to obtain corresponding third feature data;

and the construction submodule is used for constructing the target characteristic data based on the first characteristic data, the second characteristic data and the third characteristic data.

In some alternative implementations of the present embodiment, the building sub-module includes:

The merging unit is used for carrying out data merging on the first characteristic data, the second characteristic data and the third characteristic data to obtain a corresponding characteristic data set;

the matching unit is used for carrying out data matching on all the characteristic data in the characteristic data set, and screening fourth characteristic data repeatedly appearing in the characteristic data set;

and the determining unit is used for taking the fourth characteristic data as the target characteristic data.

In this embodiment, the operations performed by the modules or units respectively correspond to the steps of the artificial intelligence-based prediction method in the foregoing embodiment one by one, which is not described herein again.

In some optional implementations of this embodiment, the artificial intelligence based prediction apparatus further includes:

the third acquisition module is used for acquiring the pre-collected policy sample data;

the processing module is used for preprocessing the policy sample data to obtain processed policy sample data;

the dividing module is used for dividing the processed policy sample data into a training set and a testing set according to a preset proportion;

the fourth acquisition module is used for acquiring a pre-constructed initial prediction model;

The training module is used for training the initial prediction model by using the training set to obtain a trained initial prediction model;

the verification module is used for verifying the trained initial prediction model by using the test set;

and the determining module is used for taking the trained initial prediction model as the renewal prediction model if the trained initial renewal prediction model passes the verification.

a fifth acquisition module, configured to acquire a preset probability threshold;

the comparison module is used for carrying out numerical comparison on the renewal prediction probability and the probability threshold value to obtain a corresponding comparison result;

and the second generation module is used for generating a client type corresponding to the target client based on the comparison result.

a sixth acquisition module, configured to acquire specified data corresponding to a specified data type from the historical policy data based on a preset specified data type;

the calculation module is used for calling a preset first calculation formula to calculate a future value score corresponding to the target client based on the renewal prediction probability and the specified data;

and a third generation module, configured to generate a composite value score of the target client based on the future value score.

In some optional implementations of this embodiment, the third generating module includes:

the acquisition sub-module is used for acquiring the target client state of the target client;

the first determining submodule is used for determining a target value calculation formula corresponding to the target client state;

the first calculation sub-module is used for calling a preset second calculation formula to calculate a target data value corresponding to the target parameter based on the target parameter contained in the target value calculation formula;

The second calculation sub-module is used for substituting the future value score and the target data value into the target value calculation formula to carry out numerical calculation so as to obtain a corresponding calculation result;

and the second determination submodule is used for taking the calculation result as the comprehensive value score of the target client.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 4, fig. 4 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It should be noted that only computer device 4 having components 41-43 is shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 41 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 4. Of course, the memory 41 may also comprise both an internal memory unit of the computer device 4 and an external memory device. In this embodiment, the memory 41 is typically used to store an operating system and various application software installed on the computer device 4, such as computer readable instructions of an artificial intelligence based prediction method. Further, the memory 41 may be used to temporarily store various types of data that have been output or are to be output.

The processor 42 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions stored in the memory 41 or process data, such as computer readable instructions for executing the artificial intelligence based prediction method.

The network interface 43 may comprise a wireless network interface or a wired network interface, which network interface 43 is typically used for establishing a communication connection between the computer device 4 and other electronic devices.

in the embodiment of the application, firstly, historical policy data of a target client is obtained; then acquiring the policy data of the temporary policy from the history policy data; then extracting target feature data from the policy data based on a preset feature extraction algorithm; inputting the target characteristic data into a pre-trained renewal prediction model, predicting the target characteristic data through the renewal prediction model, and outputting a corresponding renewal prediction result; and finally, determining the renewal prediction probability of the target client for the temporary warranty based on the renewal prediction result. According to the method and the device for predicting the renewal of the target client, the target characteristic data are extracted from the policy data of the temporary policy of the target client, and then the target characteristic data are processed by utilizing the renewal prediction model trained in advance so as to rapidly and accurately predict the renewal intention of the target client, so that the renewal prediction probability of the corresponding target client is generated, the insurance marketer is facilitated to carry out matched insurance business follow-up processing on the target client according to the obtained renewal prediction probability, the intelligence of the insurance business follow-up processing of the target client is improved, and the customer renewal rate is facilitated to be improved.

The present application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of an artificial intelligence-based prediction method as described above.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present application.

It is apparent that the embodiments described above are only some embodiments of the present application, but not all embodiments, the preferred embodiments of the present application are given in the drawings, but not limiting the patent scope of the present application. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a more thorough understanding of the present disclosure. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing, or equivalents may be substituted for elements thereof. All equivalent structures made by the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the protection scope of the application.

Claims

1. An artificial intelligence based prediction method, comprising the steps of:

acquiring historical policy data of a target client;

acquiring policy data of a temporary policy from the historical policy data;

2. The method for predicting based on artificial intelligence according to claim 1, wherein the step of extracting target feature data from the policy data based on a preset feature extraction algorithm specifically comprises:

3. The artificial intelligence based prediction method according to claim 2, wherein the step of constructing the target feature data based on the first feature data, the second feature data, and the third feature data specifically comprises:

and taking the fourth characteristic data as the target characteristic data.

4. The artificial intelligence-based prediction method according to claim 1, further comprising, before the step of inputting the target feature data into a pre-trained renewal prediction model, performing prediction processing on the target feature data by the renewal prediction model and outputting a renewal prediction result corresponding to the target feature data:

acquiring pre-collected policy sample data;

Preprocessing the policy sample data to obtain processed policy sample data;

acquiring a pre-constructed initial prediction model;

verifying the trained initial prediction model by using the test set;

5. The artificial intelligence based prediction method according to claim 1, further comprising, after the step of determining a renewal prediction probability of the target client for the temporary policy based on the renewal prediction result:

acquiring a preset probability threshold;

6. The artificial intelligence based prediction method according to claim 1, further comprising, after the step of determining a renewal prediction probability of the target client for the temporary policy based on the renewal prediction result:

7. The artificial intelligence based prediction method of claim 6, wherein the step of generating a composite value score for the target client based on the future value score comprises:

acquiring a target client state of the target client;

8. An artificial intelligence based predictive device comprising:

9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the artificial intelligence based prediction method of any of claims 1 to 7.

10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the artificial intelligence based prediction method of any of claims 1 to 7.