CN113723795A

CN113723795A - Information delivery strategy testing method and device, storage medium and electronic equipment

Info

Publication number: CN113723795A
Application number: CN202110987250.9A
Authority: CN
Inventors: 周国平; 史振光
Original assignee: Taikang Insurance Group Co Ltd; Taikang Online Property Insurance Co Ltd
Current assignee: Taikang Insurance Group Co Ltd; Taikang Online Property Insurance Co Ltd
Priority date: 2021-08-26
Filing date: 2021-08-26
Publication date: 2021-11-30
Anticipated expiration: 2041-08-26
Also published as: CN113723795B

Abstract

The application provides an information delivery strategy testing method, an information delivery strategy testing device, a storage medium and electronic equipment, and relates to the technical field of data processing. Obtaining policy granularity characteristics corresponding to each policy type based on historical policy granularity information; for any policy type in the policy types, the following operations are respectively executed: determining a customer quality score corresponding to each policy granularity characteristic based on each policy granularity characteristic corresponding to the policy type and the policy label of each historical policy corresponding to each policy granularity characteristic; and respectively determining the test result of the information delivery strategy corresponding to each policy type according to each customer quality score corresponding to each policy type, so as to adjust each information delivery strategy according to the test result. According to the method and the device, the test result of the information delivery strategy is determined based on the customer quality score, so that the user can adjust the information delivery strategy in time according to the test result, and therefore the loss caused by the information delivery strategy is reduced.

Description

Information delivery strategy testing method and device, storage medium and electronic equipment

Technical Field

The application relates to the technical field of data processing, in particular to an information delivery strategy testing method, an information delivery strategy testing device, a storage medium and electronic equipment.

Background

In recent years, with the rise of internet traffic, benefits brought to companies for acquiring customers through a high traffic platform have become mainstream, and users can determine the cost and the benefits brought by putting multimedia information on the high traffic platform.

However, currently, only the investment and resulting conversion, cost to Customers (CAP), and resulting revenue of a company on a traffic platform can be seen by the customer. However, future input and output cannot be predicted, and the fund cannot be well planned, so that active and effective information delivery strategy adjustment cannot be made at the delivery end.

Therefore, a method for testing and adjusting the information delivery policy is needed.

Disclosure of Invention

In order to solve the existing technical problem, embodiments of the present application provide an information delivery policy testing method, apparatus, storage medium, and electronic device, so as to adjust an information delivery policy and reduce future loss of a user.

In order to achieve the above purpose, the technical solution of the embodiment of the present application is implemented as follows:

in a first aspect, an embodiment of the present application provides an information delivery policy testing method, where the method includes:

obtaining policy granularity characteristics corresponding to each policy type based on the obtained historical policy granularity information corresponding to each policy type;

for any policy type in the policy types, respectively executing the following operations: determining a customer quality score corresponding to each policy granularity characteristic based on each policy granularity characteristic corresponding to the policy type and a policy label of each historical policy corresponding to each policy granularity characteristic; the policy label comprises a removed label and an unremoved label; the customer quality score is used for representing the probability of the customer insurance application corresponding to the policy granularity characteristic;

respectively determining test results corresponding to the information delivery strategies corresponding to the policy types according to the quality scores of the clients corresponding to the policy types so as to adjust the information delivery strategies according to the test results; and the test result is used for representing whether the information releasing strategy achieves a preset effect in the releasing process.

The method and the device for determining the client quality score representing the client insurance probability are based on the policy granularity characteristics corresponding to each policy type and the policy labels of the historical policies corresponding to the policy granularity characteristics; and determining the test result of the information release strategy corresponding to the corresponding policy type based on the determined quality scores of the clients, and adjusting the corresponding information release strategy according to the determined test result, so that the information release strategy can be adjusted timely and effectively, and the loss caused by the information release strategy is reduced.

An optional implementation manner is that, the obtaining, based on the obtained historical policy granularity information corresponding to each policy type, policy granularity characteristics corresponding to each policy type includes:

for any policy type in the policy types, respectively executing the following operations:

acquiring historical policy information of each historical policy corresponding to the policy type, and performing data cleaning on the historical policy information to obtain historical policy granularity information corresponding to each historical policy;

performing feature extraction on the historical policy granularity information to obtain initial policy granularity features;

respectively performing characteristic derivation on each initial policy granularity characteristic to obtain a derived policy granularity characteristic;

respectively determining derivative characteristic values corresponding to the derivative policy granularity characteristics according to the derivative policy granularity characteristics and policy labels of the historical policies corresponding to the derivative policy granularity characteristics; the derived feature value is used to characterize the predictive power of the derived policy-granularity feature on the customer quality score;

and taking the derived policy granularity characteristic with the derived characteristic value larger than a first preset threshold value as the policy granularity characteristic.

According to the method and the device, after the historical policy information is obtained, data cleaning is carried out, useless information in the historical policy information is removed, and the historical policy granularity information is obtained; and then, after obtaining initial policy-keeping granularity characteristics based on the obtained policy-keeping granularity information, performing characteristic derivation to obtain derived policy-keeping granularity characteristics, and screening the derived policy-keeping granularity characteristics based on derived characteristic values corresponding to the derived policy-keeping granularity characteristics to remove useless characteristics, thereby improving accurate data basis for the subsequently obtained customer quality scores.

An optional implementation manner is that, the determining, based on the policy label of each policy-granularity feature corresponding to the policy type and each historical policy corresponding to each policy-granularity feature, a customer quality score corresponding to each policy-granularity feature includes:

respectively determining the quantity of non-quit historical policy and the quantity of quit historical policy corresponding to each policy granularity characteristic according to the policy label of each historical policy corresponding to each policy granularity characteristic;

respectively determining an evidence weight corresponding to each policy granularity characteristic according to the number of the non-quit history policies and the number of the quit history policies, and taking the determined evidence weight as a policy characteristic value corresponding to each policy granularity characteristic;

and inputting each policy characteristic value into a prediction model to obtain a customer quality score corresponding to each policy granularity characteristic.

In the embodiment of the application, evidence weights obtained based on non-quit historical policy data and quit historical policy data are used as policy characteristic values corresponding to various policy granularity characteristics; and then, obtaining a customer quality score based on the obtained policy keeping characteristic value, thereby improving the accuracy of the customer quality score and providing an accurate basis for subsequently adjusting an information delivery strategy.

An optional implementation manner is that, the obtaining, based on the prediction model, a customer quality score corresponding to each policy granularity feature includes:

for one policy characteristic value in each policy characteristic value, the following operations are respectively executed:

inputting the policy characteristic value into a logistic regression layer in the prediction model, and determining a scoring coefficient corresponding to the policy characteristic value based on the logistic regression layer;

inputting the policy characteristic value into a classification layer in the prediction model, and classifying each historical policy corresponding to the policy characteristic value based on the classification layer to obtain characteristic sub-values corresponding to each type corresponding to the policy characteristic value;

respectively determining sub-quality scores corresponding to the types according to the characteristic sub-values and the scoring coefficients;

and taking the sum of the sub-quality scores as the customer quality score corresponding to the policy granularity characteristic.

An optional implementation manner is that, according to each customer quality score corresponding to each policy type, the determining, respectively, a test result corresponding to the information delivery policy corresponding to each policy type, so as to adjust each information delivery policy according to the test result, includes:

for each policy type, the following operations are respectively executed:

respectively determining the quality scores of the information delivery channels corresponding to the information delivery channels and the quality scores of the information delivery platforms corresponding to the information delivery platforms according to the quality scores of the clients;

and determining a test result corresponding to the adjustment of the information delivery strategy according to the quality scores of the clients, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, so as to adjust the information delivery strategy according to the test result.

According to the method and the device, after the customer quality score is determined, the information delivery channel quality score and the information delivery platform quality score are determined according to the determined customer quality score, then the test result of the information delivery strategy is determined based on the determined customer quality score, the determined information delivery channel quality score and the determined information delivery platform quality score, the information delivery strategy is adjusted according to the determined test result, and therefore effectiveness and accuracy of the information delivery strategy are improved.

An optional implementation manner is that, according to the quality scores of the customers, respectively determining the quality scores of the information delivery channels corresponding to the information delivery channels and the quality scores of the information delivery platforms corresponding to the information delivery platforms includes:

determining policy quality scores corresponding to the historical policies according to the client quality scores;

determining the quality scores of the information delivery channels corresponding to the information delivery channels according to the information delivery channels corresponding to the historical policy and the policy quality scores;

and determining the quality scores of the information delivery platforms corresponding to the information delivery platforms according to the information delivery platforms corresponding to the historical policy and the policy quality scores.

According to the quality score determining method and device, after the policy quality score corresponding to each historical policy is determined based on each customer quality score, the quality score of the information delivery channel and the quality score of the information delivery platform are accurately determined according to the determined policy quality score, and therefore accurate data basis is provided for subsequent adjustment of information delivery strategies.

In an optional embodiment, the quality score includes the customer quality score, the information delivery channel quality score, and the information delivery platform quality score; determining a test result corresponding to the information delivery strategy according to the quality scores of the clients, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, so as to adjust the information delivery strategy according to the test result, wherein the method comprises the following steps:

if at least one quality score is smaller than a second preset threshold value, determining that the test result is that the information delivery strategy does not reach the preset effect in the delivery process, generating early warning information, and determining a withdrawal rate corresponding to one policy granularity characteristic in the policy granularity characteristics; if the receding rate is larger than a third preset threshold, determining abnormal change in the information delivery strategy delivery process, and adjusting the information delivery strategy according to the abnormal change; if the withdrawal rate is less than or equal to the third preset threshold, continuing to determine withdrawal rates corresponding to other policy granularity characteristics until the withdrawal rates are greater than the third preset threshold; the other policy granularity characteristics are any policy granularity characteristic except the policy granularity characteristic with the determined rejection rate in each policy granularity characteristic;

and if the quality scores are all larger than the second preset threshold value, determining that the test result is that the information delivery strategy achieves the preset effect in the delivery process, and delivering the multimedia information based on the information delivery strategy.

According to the embodiment of the application, after at least one quality score is determined to be smaller than a second preset threshold, the test result is determined to be that the information releasing strategy does not reach the preset effect in the releasing process, the early warning information is generated, and after the release rate corresponding to the policy granularity characteristic is determined to be larger than a third preset threshold, the abnormal change of the information releasing strategy in the releasing process is determined, so that a user can adjust the information releasing strategy according to the abnormal change, the information releasing strategy can be accurately and effectively adjusted, and the loss caused by the information releasing strategy is reduced.

In a second aspect, an embodiment of the present application further provides an information delivery policy testing apparatus, including:

the acquiring unit is used for acquiring policy granularity characteristics corresponding to each policy type based on the acquired historical policy granularity information corresponding to each policy type;

a determining unit, configured to perform the following operations for any one of the policy types, respectively: determining a customer quality score corresponding to each policy granularity characteristic based on each policy granularity characteristic corresponding to the policy type and a policy label of each historical policy corresponding to each policy granularity characteristic; the policy label comprises a removed label and an unremoved label; the customer quality score is used for representing the probability of the customer insurance application corresponding to the policy granularity characteristic;

the adjusting unit is used for respectively determining the test results corresponding to the information release strategies corresponding to the policy types according to the quality scores of the clients corresponding to the policy types so as to adjust the information release strategies according to the test results; and the test result is used for representing whether the information releasing strategy achieves a preset effect in the releasing process.

In a third aspect, an embodiment of the present application further provides an electronic device, including a memory and a processor, where the memory stores a computer program that is executable on the processor, and when the computer program is executed by the processor, the processor is enabled to implement the information delivery policy testing method according to any one of the first aspects.

In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the information delivery policy testing method according to any one of the first aspects is implemented.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is an application scenario diagram of an information delivery policy testing method according to an embodiment of the present application;

fig. 2 is a schematic flowchart of an information delivery policy testing method according to an embodiment of the present application;

fig. 3 is a schematic view of an input interface of an information delivery policy testing method according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a prediction model used in an information delivery policy testing method according to an embodiment of the present application;

fig. 5 is a schematic diagram of a bar graph corresponding to a quality score of an information delivery platform according to an embodiment of the present application;

fig. 6 is a schematic diagram of a bar graph corresponding to the quality score of an information delivery channel according to an embodiment of the present application;

fig. 7 is a schematic view of a result display interface of the information delivery policy testing method according to the embodiment of the present application;

fig. 8 is a schematic view of a result display interface of another information delivery policy testing method according to an embodiment of the present application;

fig. 9 is a schematic view of a problem root cause display interface of the information delivery policy testing method according to the embodiment of the present application;

fig. 10 is a schematic flowchart of a method for adjusting an information delivery policy according to an embodiment of the present application;

fig. 11 is a schematic view of a complete flow of an information delivery policy testing method according to an embodiment of the present application;

fig. 12 is a block diagram of a structure of an information delivery policy testing apparatus according to an embodiment of the present application;

fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

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

It should be noted that references in the specification of the present application to the terms "comprises" and "comprising," and variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Some terms in the embodiments of the present application are explained below to facilitate understanding by those skilled in the art.

(1) Evidence Weight (Weight of Evidence, WOE): WOE is a form of encoding on original arguments, a method of measuring the difference between normal sample (Good) and default sample (Bad) distributions.

(2) Information Value (innovation Value, IV): the term "prediction ability" is used to indicate the degree of contribution of a feature to a target prediction, i.e., the prediction ability of a feature, and generally speaking, the higher the IV value, the stronger the prediction ability of the feature, and the higher the information contribution degree.

The word "exemplary" is used hereinafter to mean "serving as an example, embodiment, or illustration. Any embodiment described as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The terms "first" and "second" are used herein for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of embodiments of the application, unless stated otherwise, "plurality" means two or more.

In order to better understand the technical solution provided by the embodiment of the present application, some brief descriptions are provided below for application scenarios to which the technical solution provided by the embodiment of the present application is applicable, and it should be noted that the application scenarios described below are only used for illustrating the embodiment of the present application and are not limited. In specific implementation, the technical scheme provided by the embodiment of the application can be flexibly applied according to actual needs.

In the prior art, when a user puts multimedia information according to an information putting strategy, the cost and the income brought by the information putting strategy can be determined only according to historical policy data obtained after the multimedia information is put, the input and output brought by the information putting strategy in the future time cannot be predicted, and the information putting strategy cannot be adjusted timely.

Based on the above problems, the embodiments of the present application provide an information delivery policy testing method and apparatus. The parking space state detection method can be applied to a terminal, such as a computer; but also to the server.

As shown in fig. 1, an application scenario of the information delivery policy testing method is provided in the embodiment of the present application. Referring to fig. 1, the electronic device 100 is communicatively connected to a server 200 through a network, and the server 200 is communicatively connected to a terminal device 300 through the network, where the network may be, but is not limited to, a local area network, a metropolitan area network, a wide area network, or the like, the number of servers 200 connected to the electronic device 100 may be plural, and the number of terminal devices 300 connected to the servers 200 may be plural. The electronic device 100 may transmit communication data and messages to each other through the network and the server 200, and the terminal device 300 may transmit communication data and messages to each other through the network and the server 200.

The terminal 300 may be a portable device (e.g., a mobile phone, a tablet Computer, a notebook Computer, etc.), or may be a Computer, a smart screen, a Personal Computer (PC), etc. The terminal device 300 may be installed with various clients including a video playing client, a social application client, and the like, and is an electronic device capable of displaying each operation interface provided in the installed client and each object in the operation interface.

The client is software installed on the terminal device 300, for example, APP installed on a mobile terminal such as a mobile phone, and belongs to software resources in the terminal device 300. For example, the terminal device 300 may download an installation package of the client via the network, install the client using the installation package, and after the installation is completed, the client may run on the terminal device 300.

The electronic device 100 may be a terminal, e.g., a computer; the electronic device 100 may also be a server, for example, a server or a server cluster or distributed system composed of a plurality of servers, or a virtualization platform, or a personal computer, a large and medium-sized computer, or a computer cluster, or a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and large data and artificial intelligence platform. According to implementation needs, the application scenario in the embodiment of the present application may have any number of terminal devices and servers. The embodiment of the present application is not particularly limited to this.

Illustratively, the terminal device 300 has a third-party application client installed thereon, and the third-party application client may be a browser, a video player, a music player, or the like. In the process of performing the information delivery policy test, the electronic device 100 sends an information delivery sub-policy to the server 200 corresponding to the agent according to the information delivery policy; the information delivery sub-strategy is a multimedia information delivery strategy corresponding to each agent in the information delivery strategy. The server 200 will deliver multimedia information, such as advertisement video or advertisement pictures, etc., to each target account of the third party application. When the user logs in the third-party application client by using the target account, the third-party application client displays the multimedia information to the user so that the user can receive the related information, know the related content of the insurance promotion and apply the insurance.

Specifically, when the information delivery policy is tested, historical policy within a set time period may be obtained, and the historical policy may be classified according to policy types corresponding to the historical policies. The electronic device 100 obtains policy granularity characteristics of each policy type pair based on the obtained historical policy granularity information corresponding to each policy type, and determines a customer quality score corresponding to each policy granularity characteristic according to the policy granularity characteristics and the policy label of each historical policy corresponding to the policy granularity characteristics; and then, respectively determining a test result corresponding to the information release strategy corresponding to each policy type based on each customer quality score corresponding to each policy type, so as to adjust each information release strategy according to the test result.

As shown in fig. 2, an embodiment of the present application provides an information delivery policy testing method, including the following steps:

step S201, obtaining policy granularity characteristics corresponding to each policy type based on the obtained historical policy granularity information corresponding to each policy type.

It should be noted that the policy granularity information is used to represent the basic information of the minimum dimension of the policy; policy granularity information includes, but is not limited to: annual premium, age of the applicant, presence or absence of social security of the applicant, province of the applicant, sex, and time of application.

According to the method and the device, historical policy granularity information corresponding to each policy type in a set time period is obtained, and policy granularity characteristics corresponding to each policy type are obtained according to the obtained historical policy granularity information.

In specific implementation, after acquiring historical policy granularity information corresponding to each historical policy in a historical time period, the embodiment of the application classifies the acquired historical policy granularity information according to corresponding policy types; and obtaining policy granularity characteristics corresponding to the policy types based on the historical policy granularity information corresponding to the policy types.

For example, the embodiment of the application can acquire historical policy granularity information in the past year, and determine policy types corresponding to the acquired historical policy granularity information as vehicle insurance and personal accident insurance; and obtaining policy granularity characteristics corresponding to the vehicle insurance and the personal accident insurance respectively according to the obtained historical policy granularity information.

Step S202, for any policy type in the policy types, the following operations are respectively performed: and determining the customer quality scores corresponding to the policy granularity characteristics based on the policy granularity characteristics corresponding to the policy types and the policy labels of the historical policies corresponding to the policy granularity characteristics.

Note that the policy label includes a removed label and an unremoved label; the customer quality score is used to characterize the probability of a customer's application corresponding to the policy granularity characteristic.

After the policy granularity features corresponding to the policy types are obtained, the policy labels corresponding to the policy granularity features are determined to be historical policies with labels removed and the policy labels are historical policies without labels removed based on the policy labels corresponding to the historical policies corresponding to the policy granularity features.

In some embodiments, the customer quality score corresponding to each policy granularity feature is determined according to each determined policy granularity feature, and each removed historical policy and each unremoved historical policy corresponding to each policy granularity feature.

Step S203, determining test results corresponding to the information release strategies corresponding to the policy types respectively according to the quality scores of the clients corresponding to the policy types, and adjusting the information release strategies according to the test results.

It should be noted that the test result is used to represent whether the information delivery policy achieves a preset effect in the delivery process.

In some embodiments, the information delivery policy achieving the preset effect in the delivery process may be expressed as: in the information release strategy release process, the quality of the client released by the information release strategy, the quality of the insurance policy corresponding to the information release channel corresponding to the information release strategy and the quality of the insurance policy corresponding to the information release platform reach preset quality scores.

After determining the customer quality scores corresponding to the policy granularity characteristics, determining the customer quality scores and the corresponding policy types.

Specifically, in the embodiment of the application, based on the determined quality scores of the clients, the test results corresponding to the information delivery strategies corresponding to the policy types are determined, so that the information delivery strategies are adjusted according to the test results.

In some embodiments, the application embodiment determines, according to the determined customer quality score, an insurance probability of the customer corresponding to the policy granularity feature corresponding to the customer quality score.

For example, the higher the customer quality score, the higher the probability of insuring a customer for the policy granularity feature corresponding to the customer quality score.

According to the insurance policy testing method and device, the testing results corresponding to the information releasing strategies corresponding to the insurance policy types are respectively determined according to the determined insurance policy application probabilities of the customers corresponding to the insurance policy granularity characteristics corresponding to the insurance policy types, and the corresponding information releasing strategies are adjusted according to the determined testing results.

The method and the device for determining the client quality score representing the client insurance probability are based on the policy granularity characteristics corresponding to each policy type and the policy labels of the historical policies corresponding to the policy granularity characteristics; and determining a test result corresponding to the information release strategy corresponding to the corresponding policy type based on the determined quality scores of the clients, and adjusting each information release strategy according to the test result, so that the information release strategy can be effectively adjusted in time, and the loss caused by the information release strategy is reduced.

In some embodiments, the electronic device determines, according to a policy type input by a user, a policy granularity characteristic corresponding to the policy type.

For example, as shown in fig. 3, the user may input the policy type to be predicted in the input interface to obtain the policy granularity characteristic corresponding to the policy type.

In other embodiments, the information delivery policies corresponding to all policy types are directly tested in the embodiments of the present application.

Specifically, in the embodiment of the present application, policy granularity features corresponding to each policy type are obtained, and an information delivery policy is tested according to the obtained policy granularity features.

In some embodiments, the policy granularity characteristic corresponding to the policy type may be obtained by the following method.

In implementation, according to the obtained policy type, the embodiment of the application obtains the historical policy information of each historical policy corresponding to the policy type, and performs data cleaning on the obtained historical policy information to obtain the historical policy granularity information corresponding to each historical policy.

In some embodiments, before Data cleaning is performed on historical policy information, Data Exploration (EDA) is performed on the acquired historical policy information to determine the integrity and authenticity of the historical policy information.

According to the embodiment of the application, after data exploration is carried out on the historical policy information, the complete historical policy information corresponding to each historical policy is determined, the incomplete historical policy information corresponding to each historical policy is eliminated, and the complete historical policy information corresponding to each historical policy is obtained.

In some embodiments, the method includes the steps that historical policy information with complete data corresponding to each historical policy corresponding to the policy type is collected, and the historical policy information is subjected to data cleaning, so that vacancy values and abnormal values in the historical policy information are removed; and through the service rule and the statistical method corresponding to the policy type, the historical policy information is cleaned into the intuitive historical policy granularity information.

For example, according to the business rule corresponding to the policy type, the embodiment of the present application performs data cleaning on the obtained historical policy information with complete data, removes irrelevant information, and sorts the processed historical policy information by a statistical method to obtain a policy data table corresponding to the policy type, and uses the obtained policy data table as the historical policy granularity information.

After the historical policy granularity information is obtained, the obtained historical policy granularity information is packaged to a big data platform or other data bins.

In some embodiments, after obtaining the historical policy granularity information, the embodiments of the present application perform feature extraction on the historical policy granularity information to obtain initial policy granularity features.

For example, initial policy granularity characteristics include, but are not limited to: age, time of insuring, sex, province of insuring person.

After the initial policy granularity characteristics are obtained, the embodiment of the application performs characteristic derivation on each initial policy granularity characteristic to obtain derived policy granularity characteristics.

In some embodiments, in the embodiments of the present application, feature derivation is performed on the initial policy granularity features according to the derivation rules corresponding to the initial policy granularity features, so as to obtain derived policy granularity features corresponding to the initial policy granularity features.

In specific implementation, the embodiment of the present application performs feature derivation on the initial policy granularity characteristic, and uses a more detailed characteristic obtained by resolving the initial policy granularity characteristic as a derived policy granularity characteristic.

For example, the initial policy granularity is characterized by age, and the age-corresponding derivation rules are divided into a plurality of age ranges; specifically, the age characteristics are subjected to characteristic derivation based on a derivation rule corresponding to the age, and the derived policy granularity characteristics corresponding to the initial policy granularity characteristic age are 18-25, 26-35, 36-50 and 51-65. When the initial policy granularity characteristic is the insurance time, and the insurance time is subjected to characteristic derivation, time interval derivation and week derivation can be performed; specifically, if time interval derivation is carried out on the guarantee time, the guarantee time can be disassembled into guarantee times of 1:00pm to 6:00pm, 7:00am to 12:00am and the like; when week derivation is performed on the application time, the application time can be disassembled into week 1, week 2, week 3 and the like.

After the derived policy granularity characteristics are obtained, the derived characteristic values corresponding to the derived policy granularity characteristics are respectively determined according to the obtained derived policy granularity characteristics and the policy labels of the historical policies corresponding to the derived policy granularity characteristics.

It is noted that the derived feature values are used to characterize the predictive power of the derived policy-granularity feature on the customer quality score.

In some embodiments, the policy labels of the respective historical policies corresponding to the derived policy granularity characteristics may be determined in the following manner.

In implementation, in the embodiment of the present application, each historical policy corresponding to each derived policy granularity feature is determined according to policy granularity information corresponding to each historical policy.

In some embodiments, the policy removing period of each historical policy is determined according to policy granularity information corresponding to each historical policy, the determined policy removing period is matched with the set policy removing period, and the policy label of each historical policy is determined according to the matching result.

In some embodiments, the set withdrawal period in the embodiments of the present application may be the 2 nd period, or may also be the 3 rd period, or may also be the 2 nd or 3 rd period.

It should be noted that the 2 nd and 3 rd stages are payment period numbers corresponding to the policy.

In some embodiments, upon determining that the matching result is a successful match, embodiments of the present application determine that the policy label of the historical policy is a disengaged label.

In other embodiments, upon determining that the match result is a match identification, embodiments of the present application determine that the policy label of the historical policy is an unreleased label.

After the policy labels of the historical policies corresponding to the derived policy granularity characteristics are determined, the removed historical policy and the unremoved historical policy corresponding to the derived policy granularity characteristics are determined.

In the embodiment of the application, before determining the derived feature value according to each derived policy granularity feature and the policy label of each historical policy corresponding to each derived policy granularity feature, the obtained derived policy granularity feature needs to be preliminarily screened.

In particular, the embodiments of the present application may perform a preliminary screening on derived policy size characteristics in the following manner.

In some embodiments, the XGBOOST algorithm may be used to perform a preliminary screening on derived policy granularity characteristics.

In specific implementation, the XGBOOST algorithm is used to classify the derived policy granularity features, determine the historical policy corresponding to each derived policy granularity feature, determine the importance corresponding to each derived policy granularity feature according to the number of the determined historical policies corresponding to each derived policy granularity, and remove the derived policy granularity features with the importance smaller than the importance threshold value to complete the primary screening of the derived policy granularity features.

After the derived policy size features are preliminarily screened, the derived feature values corresponding to the derived policy size features are respectively determined according to the screened derived policy size features and policy labels of historical policies corresponding to the derived policy size features, and the derived policy size features after preliminary screening are secondarily screened according to the obtained derived feature values.

In specific implementation, in the embodiment of the present application, the amount of the retired historical policy and the amount of the unreleased historical policy corresponding to each derived policy granularity characteristic are determined according to the policy label of each historical policy corresponding to each derived policy granularity characteristic.

In some embodiments, the derivative feature values corresponding to the derived policy granularity features are respectively determined according to the determined removed historical policy quantity and the determined unremoved historical policy quantity corresponding to the derived policy granularity features.

In specific implementation, the embodiment of the present application may determine the derived feature value corresponding to each derived policy granularity feature in the following manner.

According to the method and the device, the derived evidence weight corresponding to each derived policy granularity characteristic can be determined according to the determined removed historical policy quantity and unremoved historical policy quantity corresponding to each derived policy granularity characteristic, and then the derived characteristic value corresponding to each derived policy granularity characteristic is determined according to the determined evidence weight.

In specific implementation, the embodiment of the present application may determine the derived evidence weight corresponding to each policy granularity feature in the following manner.

The method comprises the steps of determining the total number of unremoved historical insurance policies and the total number of removed historical insurance policies in the obtained historical insurance policies; and determining the derived evidence weight corresponding to the granularity characteristics of each derived policy according to the determined number of the retired historical policies, the determined number of the unreleased historical policies, the determined total number of the unreleased historical policies and the determined total number of the retired historical policies.

In some embodiments, the derived evidence weight corresponding to each derived policy granularity feature may be determined by the following formula:

wherein, woe_iRepresenting the derived evidence weight corresponding to the derived policy granularity characteristic;

representing the proportion of the number of retired historical policies corresponding to the derived policy granularity features to the total number of retired historical policies;

representing the proportion of the number of unretired historical policies corresponding to the derived policy granularity characteristics to the total number of unretired historical policies; y is_iRepresenting the number of retired historical policies corresponding to the derived policy granularity features; n is_iRepresenting the number of unretired historical policies corresponding to the derived policy granularity characteristics; y is_tRepresenting a total number of outstanding historical policies in the historical policy; n is_tRepresenting the total number of unredeemed historical policies in the historical policy.

After the derived evidence weight corresponding to each derived policy granularity characteristic is determined, the derived characteristic value corresponding to each derived policy granularity characteristic is determined according to the determined derived evidence weight.

In some embodiments, the derived feature value corresponding to each derived policy granularity feature may be determined by the following formula:

wherein IV_iRepresenting a derivative characteristic value corresponding to each derivative policy granularity characteristic; woe_iRepresenting the derived evidence weight corresponding to the derived policy granularity characteristic;

After the derivative characteristic values corresponding to the derivative policy granularity characteristics are determined, secondary screening is carried out on the derivative policy granularity characteristics according to the determined derivative characteristic values, and policy granularity characteristics are obtained.

In specific implementation, the derived policy granularity characteristic with the derived characteristic value greater than the first preset threshold is used as the policy granularity characteristic in the embodiment of the present application.

For example, derived policy size feature A, B, C corresponds to derived feature values of 0.05, 0.04, 0.015; the first preset threshold value is 0.02; wherein 0.05 and 0.04 are greater than the first preset threshold value 0.02, the derived policy granularity characteristic A, B corresponding to 0.05 and 0.04 is taken as the policy granularity characteristic.

In other embodiments, the present application eliminates derived policy-granularity features having derived feature values less than or equal to a first preset threshold.

In other embodiments, the derived policy granularity features with the derived feature value less than or equal to the first preset threshold are removed, and the derived policy granularity features with the derived feature value greater than the first preset threshold are used as the residual features; according to the embodiment of the application, the remaining features are used for carrying out Bernoulli binomial distribution linear fitting on the policy-keeping label, and the confidence level of the remaining features is determined; and deleting the residual features with the confidence level larger than the confidence level threshold, continuing to perform fitting until the confidence level is lower than the confidence level threshold, and taking the last residual features as policy-preserving granularity features.

For example, the first preset threshold is 0.02, the confidence level threshold is 0.05; the derived policy size features comprise A, B, C, D, E, F corresponding derived feature values of 0.02, 0.06, 0.03, 0.04, 0.05, 0.025, the derived policy size feature A with the derived feature value of 0.02 is removed, and the residual derived policy size feature B, C, D, E, F is used as residual feature; performing Bernoulli binomial distribution linear fitting on the residual characteristics B, C, D, E, F to the policy label to determine the confidence level of the residual characteristics; the confidence levels corresponding to the residual characteristics B, C, D, E, F are respectively 0.05, 0.04, 0.03, 0.05 and 0.06, the residual characteristics F with the confidence level of 0.06 are removed, and the residual characteristics B, C, D, E are used as policy granularity characteristics.

After the policy granularity characteristics are determined, the client quality scores corresponding to the policy granularity characteristics are determined according to the determined policy granularity characteristics and the policy labels of the historical policies corresponding to the policy granularity characteristics.

In some embodiments, the customer quality score corresponding to each policy granularity feature may be determined by the embodiments of the present application in the following manner.

In specific implementation, according to the policy label of each historical policy corresponding to each policy granularity characteristic, the embodiment of the present application determines the number of non-quit historical policies and the number of quit historical policies corresponding to each policy granularity characteristic, respectively.

According to the determined number of non-quit historical policy and the determined number of quit historical policy, the evidence weight corresponding to each policy granularity characteristic is respectively determined, and the determined evidence weight value is used as the policy characteristic value corresponding to each policy granularity characteristic.

In some embodiments, the method determines the total number of non-quit historical policy and the total number of quit historical policy in the obtained historical policy, and then determines policy feature values corresponding to each policy granularity feature according to the determined total number of the non-quit historical policy and the determined total number of the quit historical policy, and the number of the non-quit historical policy and the number of the quit historical policy corresponding to each policy granularity feature.

In specific implementation, the policy characteristic value corresponding to each policy granularity characteristic may be determined by the following formula:

wherein, WOE_iRepresenting evidence weight corresponding to the policy granularity characteristic;

representing the proportion of the number of the removed historical policy corresponding to the policy granularity characteristic to the total number of the removed historical policy;

representing the proportion of the number of unremoved historical policy corresponding to the policy granularity characteristic to the total number of unremoved historical policies; y is_iRepresenting the number of retired historical policy corresponding to policy granularity characteristics; n is_iRepresenting the number of unreleased historical policy corresponding to the policy granularity characteristic; y is_tRepresenting a total number of outstanding historical policies in the historical policy; n is_tRepresenting the total number of unredeemed historical policies in the historical policy.

In some embodiments, after the evidence weight corresponding to each policy granularity characteristic is determined, the determined evidence weight value is used as the policy characteristic value corresponding to each policy granularity characteristic.

After the policy characteristic value corresponding to each policy granularity characteristic is obtained, each policy characteristic value is input into the prediction model to obtain the customer quality score corresponding to each policy granularity characteristic.

It should be noted that, as shown in fig. 4, the prediction model in the embodiment of the present application includes a logistic regression layer and a classification layer.

In specific implementation, in the embodiment of the present application, each policy feature value is input to a logistic regression layer in the prediction model, and based on the logistic regression layer, a scoring coefficient corresponding to each policy feature value is respectively determined.

In some embodiments, the scoring coefficients corresponding to the policy feature values may be determined in the following manner.

After the policy characteristic values are input into the logistic regression layer, the corresponding ratios of the policy characteristic values are determined, and the scoring coefficients corresponding to the policy characteristic values are determined according to the determined ratios.

In some embodiments, the present embodiments may determine the ratio by the following formula:

wherein θ represents a ratio; p represents the proportion of the number of the released historical policy to the total number of the historical policies; alpha is alpha₀，α_iA regression coefficient representing a logistic regression layer; WOE_iRepresenting a policy feature value.

After the proportion is determined, the scoring coefficient corresponding to each policy feature value can be determined through the following formula:

wherein, A and B represent scoring coefficients; p_D0Represents 2 theta₀Corresponding score, P₀Denotes theta₀A corresponding score value; theta₀Indicating the ratio corresponding to each policy feature value.

The embodiment of the application inputs each policy characteristic value into a classification layer in a prediction model, and classifies each historical policy corresponding to each policy characteristic value based on the classification layer to obtain the characteristic sub-value corresponding to each type corresponding to the policy characteristic value.

It should be noted that the classification layer may be a decision tree binning layer.

In specific implementation, after the policy characteristic values are input into the classification layer, the historical policies corresponding to the policy characteristic values are classified based on the classification layer, so that the proportion of the number of the withdrawn historical policies in the obtained types to the total number of the historical policies accords with the monotonicity of the badges.

According to the embodiment of the application, after the historical policy corresponding to each policy eigenvalue is classified to obtain a plurality of types, the eigenvalue corresponding to each type corresponding to each policy eigenvalue is respectively determined.

In specific implementation, the characteristic sub-values corresponding to each type corresponding to the policy characteristic value are determined in the following manner in the embodiment of the present application.

In implementation, in the embodiment of the present application, the number of removed historical policy and the number of unremoved historical policy corresponding to each type, as well as the number of removed historical policy and the number of unremoved historical policy corresponding to the policy feature value are respectively determined, the evidence weight corresponding to each type is determined, and the obtained evidence weight is used as the feature sub-value corresponding to each type.

After determining each characteristic sub-value and scoring coefficient corresponding to each policy characteristic value, determining sub-quality scores corresponding to each type corresponding to each policy characteristic value respectively.

In specific implementation, in the embodiment of the present application, the sum of each sub-quality score corresponding to each policy feature value is used as the customer quality score corresponding to the policy granularity feature.

In some embodiments, the customer quality score corresponding to each policy granularity feature may be determined according to the following formula:

wherein, score_{General assembly}Representing a customer quality score corresponding to the policy granularity characteristic; a + B alpha₀Representing a sub-quality score corresponding to type 0; a and B represent scoring coefficients; alpha is alpha₀，α_iRepresenting a regression coefficient; WOE_iAnd representing the characteristic sub-value corresponding to the ith type.

After determining the quality scores of the customers corresponding to the policy types, determining the test results corresponding to the information delivery strategies corresponding to the policy types according to the quality scores of the customers corresponding to the policy types, so as to adjust the information delivery strategies according to the test results.

The information delivery strategy corresponding to the policy type can be adjusted in the following manner.

In specific implementation, after determining each customer quality score corresponding to each policy type, according to the determined customer quality scores, respectively determining an information delivery channel quality score corresponding to each information delivery channel and an information delivery platform quality score corresponding to each information delivery platform;

according to the quality scores of the clients, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, the testing result corresponding to the information delivery strategy is determined, and the information delivery strategy is adjusted according to the testing result.

In some embodiments, the information delivery channel quality score corresponding to each information delivery channel and the information delivery platform quality score corresponding to each information delivery platform may be determined by a classifier in the prediction model shown in fig. 4.

Specifically, the policy quality scores corresponding to the historical policies are determined according to the determined customer quality scores corresponding to the policy granularity features.

Specifically, according to each customer quality score and the historical policy corresponding to each customer quality score, the average value of each customer quality score corresponding to the historical policy is determined, and the sum of the average values of each customer quality score is used as the policy quality score corresponding to each historical policy.

In some embodiments, the present application embodiments take the quotient of the number of historical policies corresponding to the customer quality score and the customer quality score as the mean of the customer quality scores corresponding to the historical policies.

After the policy quality score corresponding to each historical policy is determined, determining the information release channel quality score corresponding to each information release channel according to the information release channel and the policy quality score corresponding to each historical policy; and determining the quality scores of the information delivery platforms corresponding to the information delivery platforms according to the information delivery platforms corresponding to the historical policy and the policy quality scores.

For example, the customer quality scores corresponding to the policy granularity characteristic A, B, C are 30, 40 and 50, respectively, wherein the number of the historical policies corresponding to the policy granularity characteristic A, B, C is 10, 20 and 10, respectively, and the average of the customer quality scores corresponding to the historical policies is 3, 2 and 5, respectively; determining the corresponding policy quality score of each historical policy as 10; the information release platforms corresponding to the historical warranties are two platforms a and b, wherein the number of the historical warranties corresponding to the information release platform a is 6, the quality score of the information release platform corresponding to the information release platform a is 60, the number of the historical warranties corresponding to the information release platform b is 4, and the quality score of the information release platform corresponding to the information release platform b is 40; the corresponding information release channels in the historical policy are two channels e and f, wherein if the number of the historical policies corresponding to the information release channel e is 6, the quality score of the information release channel corresponding to the information release channel e is 60, and if the number of the historical policies corresponding to the information release channel f is 4, the quality score of the information release channel corresponding to the information release channel f is 40.

In some embodiments, after determining the policy quality score corresponding to each policy type, the embodiment of the present application predicts the renewal rate and withdrawal rate in a period of time in the future according to the preset correspondence between the policy quality score and the renewal rate and the number of policies.

For example, the corresponding relationship between the policy quality score and the renewal rate and the policy number in the embodiment of the present application is shown in table 1:

table 1: corresponding relation between quality score of policy and renewal rate and policy number

In some embodiments, the corresponding relationship between the policy quality score and the renewal rate and the policy number is updated according to a preset period.

For example, the policy quality score is updated once a month according to the correspondence between the renewal rate and the number of policies.

In a specific implementation, the embodiment of the present application represents the renewal rate corresponding to the quality score of the historical policy according to the aggregate value of the renewal conditions accumulated by the historical policy.

In some embodiments, after determining the quality scores of the information delivery channels, the quality scores of the information delivery platforms and the quality scores of the customers, the determined quality scores may be displayed in a display interface.

For example, as shown in fig. 5, after the quality score of the information delivery platform is determined, the information delivery platform is displayed in a display interface in a form of a bar graph; as shown in fig. 6, after the quality score of the information delivery channel is determined, the information delivery channel is displayed in a display interface in a form of a bar graph.

In some embodiments, after determining the renewal rate, the number of insurance policies, and the quality score corresponding to the policy type through the prediction model, the embodiment of the present application displays the prediction result in the display interface.

For example, as shown in fig. 7, the embodiment of the present application further provides a result display interface for displaying the policy quality score, the renewal rate and the number of policies corresponding to the policy type. The user can determine whether to adjust the information delivery strategy according to the display content of the result display interface.

According to the quality scores of all clients, the quality scores of all information delivery channels and the quality scores of all information delivery platforms, the testing result corresponding to the information delivery strategy is determined, and the information delivery strategy is adjusted according to the testing result.

In specific implementation, the information delivery strategy can be adjusted in the following manner in the embodiment of the present application.

The method and the device for determining the quality score determine whether the quality score is smaller than a second preset threshold value or not, determine a test result corresponding to the information delivery strategy, and adjust the information delivery strategy according to the determined test result.

It should be noted that the quality score includes a customer quality score, an information delivery channel quality score, and an information delivery platform quality score. The second preset threshold may be an average of the historical quality scores.

In specific implementation, if it is determined that at least one quality score is smaller than a second preset threshold, the test result is determined that the information delivery strategy does not reach a preset effect in the delivery process, the early warning information is generated, and the withdrawal rate corresponding to one policy granularity feature of the policy granularity features is determined.

For example, if the customer quality score corresponding to the policy granularity characteristic is 20 points and the customer quality score corresponding to the historical policy granularity characteristic is 70 points, it is determined that the test result is that the information delivery strategy does not reach the preset effect in the delivery process, and the early warning information is generated.

In some embodiments, if it is determined that the fallback rate corresponding to the policy granularity feature is greater than a third preset threshold, determining an abnormal change in the information delivery policy during delivery, so as to adjust the information delivery policy according to the abnormal change.

It should be noted that the abnormal change includes, but is not limited to: the information delivery strategy is changed in the delivery process, when the information delivery platform delivers the multimedia information according to the information delivery strategy, the information delivery platform is abnormal, and when the information delivery platform delivers the multimedia information according to the information delivery strategy, whether the information delivery channel is changed or not is judged.

Specifically, after determining the abnormal change in the information delivery process of the information delivery strategy, the embodiment of the application displays the abnormal change of the information delivery strategy in the problem root cause display interface, so that the user can adjust the information delivery strategy according to the abnormal change displayed in the problem root cause display interface.

For example, as shown in fig. 8, in the embodiment of the present application, after it is determined that at least one quality score is smaller than a second preset threshold, a prompt message of "problem occurs in the information delivery policy" is displayed in the display interface, and the user may click a key for searching the problem root to obtain the problem root. As shown in fig. 9, the user can adapt to adjust the information delivery policy according to the abnormal change displayed on the problem root cause display interface.

In other embodiments, if it is determined that the withdrawal rate corresponding to the policy granularity feature is less than or equal to the third preset threshold, the withdrawal rates corresponding to other policy granularity features are continuously determined until the determined withdrawal rate is greater than the third preset threshold.

The other policy size characteristic is any one of the policy size characteristics excluding the determined rejection rate among the respective policy size characteristics.

In other embodiments, if it is determined that the quality scores are all greater than the second preset threshold value, it is determined that the test result is that the information delivery strategy achieves a preset effect in the delivery process, and the multimedia information is delivered based on the information delivery strategy.

As shown in fig. 10, an embodiment of the present application provides a method for determining an abnormal change in a delivery process of an information delivery policy, including the following steps:

step S1001, obtaining each quality score corresponding to the policy type.

It should be noted that the quality score includes a customer quality score, an information delivery channel quality score, and an information delivery platform quality score.

Step S1002, determining whether the average masses are all larger than a second preset threshold value; if yes, go to step S1012; if not, step S1003 is executed.

Step S1003, determining that the test result is that the information release strategy does not reach the preset effect in the release process, sending out early warning information, and determining the release rate corresponding to one policy granularity characteristic in each policy granularity characteristic.

Step S1004, determining whether the receding rate is greater than a third preset threshold value; if yes, go to step S1005; if not, step S1011 is executed.

Step S1005, determining whether the information releasing strategy changes in the releasing process; if yes, go to step S1010; if not, go to step S1006.

Step S1006, an information delivery platform corresponding to the information delivery strategy is determined.

Step 1007, determining whether the information delivery platform is abnormal; if yes, go to step S1010; if not, executing step S1008;

specifically, the embodiment of the application determines whether the information delivery platform is abnormal when delivering the multimedia information according to the information delivery strategy.

And step S1008, determining an information delivery channel corresponding to the information delivery strategy.

Step S1009, determine whether the information delivery channel changes the corresponding information delivery strategy; if yes, go to step S1010; if not, step S1011 is executed.

Step S1010, determining abnormal change in the information delivery strategy delivery process.

In step S1011, other policy granularity characteristics are determined, and step S1003 is executed.

Step S1012, determining that the test result is that the information delivery policy achieves a preset effect in the delivery process, and delivering the multimedia information based on the information delivery policy.

As shown in fig. 11, an embodiment of the present application provides a complete flowchart of a method for testing an information delivery policy, where an example is to test an information delivery policy corresponding to a policy type; the method comprises the following steps:

step 1101, obtaining historical policy information of each historical policy corresponding to the policy type;

step S1102, data cleaning is carried out on the historical policy information to obtain historical policy granularity information corresponding to each historical policy;

step S1103, extracting the characteristics of the historical policy granularity information to obtain initial policy granularity characteristics;

step S1104, respectively performing characteristic derivation on each initial policy granularity characteristic to obtain a derived policy granularity characteristic;

step S1105, respectively determining the derived feature values corresponding to each derived policy granularity characteristic according to each derived policy granularity characteristic and the policy label of each historical policy corresponding to each derived policy granularity characteristic;

it should be noted that the derived feature values are used to characterize the predictive ability of the derived policy granularity feature on the customer quality score;

step S1106, using the derived policy granularity characteristic with the derived characteristic value larger than the first preset threshold value as the policy granularity characteristic;

step S1107, respectively determining the number of non-quit historical policy and the number of quit historical policy corresponding to each policy granularity characteristic according to the policy label of each historical policy corresponding to each policy granularity characteristic;

step S1108, respectively determining evidence weights corresponding to the granularity characteristics of the policy according to the number of non-release historical policies and the number of release historical policies;

step S1109, the determined evidence weight value is used as a policy characteristic value corresponding to each policy granularity characteristic;

step S1110, inputting each policy feature value into a logistic regression layer in the prediction model, and respectively determining a scoring coefficient corresponding to each policy feature value based on the logistic regression layer;

step S1111, respectively inputting each policy feature value into a classification layer in the prediction model, and respectively classifying each historical policy corresponding to each policy feature value based on the classification layer to obtain feature sub-values corresponding to each type corresponding to each policy feature value;

step S1112, respectively determining sub-quality scores corresponding to each type corresponding to each policy eigenvalue according to each eigenvalue and score coefficient corresponding to each policy eigenvalue;

step S1113, using the sum of the sub-quality scores corresponding to the policy characteristic values as the customer quality score corresponding to the policy granularity characteristic;

step S1114, determining policy quality scores corresponding to the historical policies according to the quality scores of the clients;

step S1115, determining the quality scores of the information delivery channels corresponding to the information delivery channels according to the information delivery channels corresponding to the historical policy and the policy quality scores;

step S1116, determining the quality scores of the information delivery platforms corresponding to the information delivery platforms according to the information delivery platforms corresponding to the historical policy and the policy quality scores;

and S1117, determining a test result corresponding to the information delivery strategy according to the quality scores of the customers, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, and adjusting the information delivery strategy according to the test result.

The information delivery strategy testing method shown in fig. 2 is based on the same inventive concept, and an information delivery strategy testing device is also provided in the embodiment of the present application, and the information delivery strategy testing device can be arranged in a terminal device. Because the device is a device corresponding to the information delivery policy test method in the embodiment of the present application, and the principle of the device for solving the problem is similar to that of the method, the implementation of the device may refer to the implementation of the above method, and repeated details are not described again.

Fig. 12 shows a schematic structural diagram of an information delivery policy testing apparatus provided in an embodiment of the present application, and as shown in fig. 12, the information delivery policy testing apparatus includes: an acquisition unit 1201, a determination unit 1202, and an adjustment unit 1203; wherein,

an obtaining unit 1201, configured to obtain policy granularity characteristics corresponding to each policy type based on the obtained historical policy granularity information corresponding to each policy type;

a determining unit 1202, configured to perform the following operations for any one of the policy types, respectively: determining a customer quality score corresponding to each policy granularity characteristic based on each policy granularity characteristic corresponding to the policy type and the policy label of each historical policy corresponding to each policy granularity characteristic; the policy label comprises a removed label and an unremoved label; the client quality score is used for representing the probability of client application corresponding to the policy granularity characteristic;

an adjusting unit 1203, configured to determine, according to each client quality score corresponding to each policy type, a test result corresponding to the information delivery policy corresponding to each policy type, respectively, so as to adjust each information delivery policy according to the test result; the test result is used for representing whether the information releasing strategy achieves the preset effect in the releasing process.

In an optional implementation manner, the obtaining unit 1201 is specifically configured to:

for any policy type in the policy types, the following operations are respectively executed:

extracting the characteristics of the historical policy granularity information to obtain initial policy granularity characteristics;

respectively determining derivative characteristic values corresponding to the derivative policy granularity characteristics according to the derivative policy granularity characteristics and policy labels of the historical policies corresponding to the derivative policy granularity characteristics; the derived feature value is used for characterizing the prediction capability of the derived policy granularity feature on the customer quality score;

and taking the derived policy granularity characteristic with the derived characteristic value larger than the first preset threshold value as the policy granularity characteristic.

An optional implementation manner is that the determining unit 1202 is specifically configured to:

respectively determining an evidence weight corresponding to each policy granularity characteristic according to the number of non-quit historical policies and the number of quit historical policies, and taking the determined evidence weight as a policy characteristic value corresponding to each policy granularity characteristic;

and inputting the characteristic value of each policy into the prediction model to obtain a customer quality score corresponding to the granularity characteristic of each policy.

inputting the policy characteristic value into a classification layer in the prediction model, classifying each historical policy corresponding to the policy characteristic value based on the classification layer, and obtaining characteristic sub-values corresponding to each type corresponding to the policy characteristic value;

respectively determining sub-quality scores corresponding to the types according to the characteristic sub-values and the score coefficients;

and taking the sum of the sub-quality scores as a customer quality score corresponding to the policy granularity characteristic.

An optional implementation manner is that the adjusting unit 1203 is specifically configured to:

for each policy type, the following operations are respectively executed:

and determining a test result corresponding to the information delivery strategy according to the quality scores of the clients, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, so as to adjust the information delivery strategy according to the test result.

determining a policy quality score corresponding to each historical policy according to each customer quality score;

One optional implementation way is that the quality score comprises a customer quality score, an information delivery channel quality score and an information delivery platform quality score; the adjusting unit 1203 is specifically configured to:

if at least one quality score is smaller than a second preset threshold value, determining that the test result is that the information release strategy does not reach the preset effect in the release process, generating early warning information, and determining the release rate corresponding to one policy granularity characteristic in each policy granularity characteristic; if the receding rate is larger than a third preset threshold, determining abnormal change in the releasing process of the information releasing strategy so as to adjust the information releasing strategy according to the abnormal change; if the withdrawal rate is less than or equal to a third preset threshold, continuously determining the withdrawal rates corresponding to other policy granularity characteristics until the withdrawal rates are greater than the third preset threshold; the other policy granularity characteristics are any policy granularity characteristic except the policy granularity characteristic with the determined withdrawal rate in each policy granularity characteristic;

and if the quality scores are determined to be larger than the second preset threshold value, determining that the test result is that the information delivery strategy achieves the preset effect in the delivery process, and delivering the multimedia information based on the information delivery strategy.

The information delivery strategy testing method shown in fig. 2 is based on the same inventive concept, and the embodiment of the application further provides an electronic device. As shown in fig. 13, for convenience of explanation, only the parts related to the embodiments of the present application are shown, and specific technical details are not disclosed, and reference may be made to the parts of the embodiments of the method of the present application. The embodiment of the application also provides the electronic equipment. The electronic device may be a server, such as the electronic device 100 shown in fig. 1, or a terminal device such as a mobile terminal or a computer, such as the electronic device 100 shown in fig. 1.

The electronic device comprises at least a memory for storing data and a processor for data processing. The processor for data Processing may be implemented by a microprocessor, a CPU, a GPU (Graphics Processing Unit), a DSP, or an FPGA when executing Processing. For the memory, the memory stores an operation instruction, which may be a computer executable code, and the operation instruction implements each step in the flow of the information delivery policy testing method according to the embodiment of the present application.

Fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present application; as shown in fig. 13, the electronic device 100 in the embodiment of the present application includes: a processor 101, a display 102, a memory 103, an input device 106, a bus 105, and a communication module 104; the processor 101, memory 103, input device 106, display 102 and communication module 104 are all connected by a bus 105, the bus 105 being used to transfer data between the processor 101, memory 103, display 102, communication module 104 and input device 106.

The memory 103 may be configured to store software programs and modules, such as program instructions/modules corresponding to the information delivery policy testing method in the embodiment of the present application, and the processor 101 executes various functional applications and data processing of the electronic device 100 by running the software programs and modules stored in the memory 103, such as the information delivery policy testing method provided in the embodiment of the present application. The memory 103 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program of at least one application, and the like; the storage data area may store data created according to the use of the electronic device 100 (such as historical orders, warranty granularity characteristics, and the like), and the like. Further, the memory 103 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.

The processor 101 is a control center of the electronic apparatus 100, connects various parts of the entire electronic apparatus 100 using the bus 105 and various interfaces and lines, and performs various functions of the electronic apparatus 100 and processes data by running or executing software programs and/or modules stored in the memory 103 and calling data stored in the memory 103. Alternatively, processor 101 may include one or more processing units, such as a CPU, GPU, digital processing unit, etc.

The processor 101 may present the results of the information delivery policy test to the user via the display 102.

The processor 101 may also be connected to a network via the communication module 104 to obtain a historical policy, etc.

The input device 106 is mainly used for obtaining input operation of a user, and when the electronic devices are different, the input device 106 may be different. For example, when the electronic device is a computer, the input device 106 can be a mouse, a keyboard, or other input device; when the electronic device is a portable device such as a smart phone or a tablet computer, the input device 106 may be a touch screen.

According to an aspect of the application, a computer program product or computer program is provided, comprising computer instructions, the computer instructions being stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, so that the computer device executes the information delivery policy testing method in any one of the embodiments.

In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

As will be appreciated by one skilled in the art, 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 the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. An information delivery strategy testing method is characterized by comprising the following steps:

2. The method according to claim 1, wherein the obtaining policy granularity characteristics corresponding to each policy type based on the obtained historical policy granularity information corresponding to each policy type comprises:

3. The method according to claim 1, wherein the determining a customer quality score for each policy granularity feature based on each policy granularity feature for the policy type and a policy label for each historical policy for each policy granularity feature comprises:

4. The method according to claim 3, wherein said deriving a customer quality score corresponding to each policy granularity feature based on the predictive model comprises:

5. The method according to claim 1, wherein the determining, according to the quality scores of the customers corresponding to the policy types, the test results corresponding to the information delivery policies corresponding to the policy types, respectively, so as to adjust the information delivery policies according to the test results, comprises:

for each policy type, the following operations are respectively executed:

6. The method according to claim 5, wherein the determining the quality score of the information delivery channel corresponding to each information delivery channel and the quality score of the information delivery platform corresponding to each information delivery platform according to the quality scores of the customers comprises:

7. The method of claim 5, wherein a quality score comprises the customer quality score, the information delivery channel quality score, and the information delivery platform quality score; determining a test result corresponding to the information delivery strategy according to the quality scores of the clients, the quality scores of the information delivery channels and the quality scores of the information delivery platforms, so as to adjust the information delivery strategy according to the test result, wherein the method comprises the following steps:

8. An information delivery strategy testing device, characterized in that, the device includes:

9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program operable on the processor, the computer program, when executed by the processor, implementing the method of any of claims 1-7.

10. A computer-readable storage medium having a computer program stored therein, the computer program characterized by: the computer program, when executed by a processor, implements the method of any one of claims 1 to 7.