1. A premium collection method, said method comprising:

determining credit data, payment related data and a loss risk of a target insurance policy of an applicant based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance fee is to be paid;

inputting the credit data into a credit assessment model, the credit assessment model outputting the applicant credit assessment results;

inputting the payment related data into a payment prediction model, wherein the payment prediction model outputs a payment probability prediction result of the applicant on the target insurance policy;

determining a user quality of the applicant based on the credit assessment result, the payment probability prediction result and a risk of reimbursement for the target policy of the applicant;

and determining the collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme.

2. The premium collection method according to claim 1, wherein the payment prediction model comprises at least two pre-trained payment prediction submodels;

the inputting the payment related data into a payment prediction model, the payment prediction model outputting a payment probability prediction result of the applicant to the target insurance policy, includes:

the payment related data are respectively input into the payment forecasting submodels, and each payment forecasting submodule respectively outputs a payment probability forecasting result of the policyholder to the target insurance policy;

and searching an optimal weight combination by using a grid search method, and summing products of all weights in the weight combination and the payment probability prediction result judged by the corresponding payment prediction sub-model to obtain a payment probability prediction result.

3. The premium collection method according to claim 2, wherein the payment predictor model comprises at least two of a lightgbm model, an xgboost model, an lr model, a gbdt model, an dnn model, a cnn model.

4. The premium collection method of claim 1, wherein said entering the credit data into a credit assessment model, said credit assessment model outputting the applicant credit assessment results, comprises:

and calculating the credit score of the applicant through a weighting algorithm based on preset weights according to the acquired credit data, and comparing the credit score with a threshold value in a preset credit requirement to determine the credit evaluation result of the applicant.

5. The premium collection method according to claim 1, wherein said determining the risk of reimbursement for the target policy comprises:

acquiring a plurality of historical policy which are consistent with the application category of the target policy from a historical claims database according to policy information of the target policy;

querying whether a claim-settled historical policy meeting a preset condition exists in the plurality of historical policies, wherein the preset condition is that the applicant of the claim-settled historical policy is similar to the applicant of the target policy;

when the claims-settled historical insurance policy meeting the preset condition exists, determining the claim amount of each claim-settled historical insurance policy and the proportion of the claim-settled historical insurance policy in the historical insurance policy;

and inputting the claim amount and the proportion of each settled historical policy into a pre-trained claim risk model to obtain the claim risk of the target policy.

6. The premium collection method according to claim 5, wherein inquiring among the plurality of historical policies as to whether there is a claims-settled historical policy satisfying a preset condition that an applicant of the claims-settled historical policy is similar to an applicant of the target policy includes:

determining similar policemen of the policemen according to the similarity between the basic data of the policemen and the basic data of each historical policemen;

and judging whether the similar applicant is an applicant of the claim-settled historical insurance policy, if so, acquiring the historical insurance policy of the claim-settled applicant.

7. The premium collection method of claim 1, wherein the user quality of the applicant comprises: low, medium and high;

the step of determining the collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme comprises the following steps:

outputting the premium charge call information to the applicant through a target premium charge call-receiving system for a first target premium with high user quality;

sending a second target policy with medium user quality and the premium call collection information to a first user terminal so that a first user of the first user terminal can check the second target policy;

and sending a third target policy with low user quality and the premium charge collection call information to a second user terminal so that a second user of the second user terminal can check the third target policy, wherein the authority of the second user terminal is higher than that of the first user terminal.

8. A premium collection apparatus, comprising:

the data collection unit is used for determining credit data, payment related data and a paying risk of a target insurance policy based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance policy is to be paid;

a credit evaluation unit for inputting the credit data into a credit evaluation model, the credit evaluation model outputting the applicant credit evaluation result;

the payment probability evaluation unit is used for inputting the payment related data into a payment prediction model, and the payment prediction model outputs a payment probability prediction result of the applicant on the target insurance policy;

the user quality evaluation unit is used for determining the user quality of the applicant based on the credit evaluation result of the applicant, the payment probability prediction result and the paying risk of the target insurance policy;

and the collection urging scheme determining unit is used for determining a collection urging scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection urging scheme, wherein the collection urging scheme comprises the following steps: premium collection call information and premium collection process information.

9. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions which, when executed by the processor, cause the processor to carry out the steps of the premium inducement method according to any one of claims 1 to 7.

10. A storage medium having stored thereon computer readable instructions which, when executed by one or more processors, cause the one or more processors to carry out the steps of the premium inducement method according to any one of claims 1 to 7.

Background

The insurance payment forecasting is one of important works of management departments of the insurance industry, and the accurate payment forecasting is favorable for helping insurance companies to master client appeal and behavior characteristics more clearly, cross-sell old clients, accurately recommend products, improve secondary conversion efficiency of the clients and solve the problem of increase of enterprise clients. For insurance customers, accurate payment prediction represents more accurate demand matching, so that the customers can be helped to select insurance products more suitable for the customers, unnecessary marketing disturbance is reduced, and the insurance company and the insurance customers realize win-win.

The existing neural network is widely applied to the insurance prediction industry, but in the complex prediction process of insurance premium, as the characteristics are not invariable but dynamically changed along with all influence factors and all the characteristics have serious collinearity problems, the payment difficulty of predicting insurance according to the neural network learning training sample is relatively increased, the training result can not achieve the practical application effect all the time, and the prediction precision is low.

Disclosure of Invention

The application provides a premium collection method and device, computer equipment and a storage medium.

In a first aspect, there is provided a premium collection method comprising:

determining credit data, payment related data and a loss risk of a target insurance policy of an applicant based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance fee is to be paid;

inputting the credit data into a credit assessment model, the credit assessment model outputting the applicant credit assessment results;

inputting the payment related data into a payment prediction model, wherein the payment prediction model outputs a payment probability prediction result of the applicant on the target insurance policy;

determining a user quality of the applicant based on the credit assessment result, the payment probability prediction result and a risk of reimbursement for the target policy of the applicant;

and determining the collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme.

In some embodiments, the payment prediction model comprises at least two pre-trained payment prediction submodels;

the inputting the payment related data into a payment prediction model, the payment prediction model outputting a payment probability prediction result of the applicant to the target insurance policy, includes:

the payment related data are respectively input into the payment forecasting submodels, and each payment forecasting submodule respectively outputs a payment probability forecasting result of the policyholder to the target insurance policy;

and searching an optimal weight combination by using a grid search method, and summing products of all weights in the weight combination and the payment probability prediction result judged by the corresponding payment prediction sub-model to obtain a payment probability prediction result.

In some embodiments, the contribution predictor model includes at least two of a lightgbm model, an xgboost model, an lr model, a gbdt model, an dnn model, a cnn model.

In some embodiments, said entering said credit data into a credit assessment model, said credit assessment model outputting said applicant credit assessment results, comprises:

and calculating the credit score of the applicant through a weighting algorithm based on preset weights according to the acquired credit data, and comparing the credit score with a threshold value in a preset credit requirement to determine the credit evaluation result of the applicant.

In some embodiments, said determining a risk of reimbursement for said target policy comprises:

acquiring a plurality of historical policy which are consistent with the application category of the target policy from a historical claims database according to policy information of the target policy;

querying whether a claim-settled historical policy meeting a preset condition exists in the plurality of historical policies, wherein the preset condition is that the applicant of the claim-settled historical policy is similar to the applicant of the target policy;

when the claims-settled historical insurance policy meeting the preset condition exists, determining the claim amount of each claim-settled historical insurance policy and the proportion of the claim-settled historical insurance policy in the historical insurance policy;

and inputting the claim amount and the proportion of each settled historical policy into a pre-trained claim risk model to obtain the claim risk of the target policy.

In some embodiments, querying the plurality of historical policies for the presence of a claims-settled historical policy that satisfies a preset condition that an applicant of the claims-settled historical policy is similar to an applicant of the target policy comprises:

determining similar policemen of the policemen according to the similarity between the basic data of the policemen and the basic data of each historical policemen;

and judging whether the similar applicant is an applicant of the claim-settled historical insurance policy, if so, acquiring the claim amount of the claim-settled applicant's historical insurance policy.

In some embodiments, the applicant's user quality includes: low, medium and high;

the step of determining the collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme comprises the following steps:

outputting the premium charge call information to the applicant through a target premium charge call-receiving system for a first target premium with high user quality;

sending a second target policy with medium user quality and the premium call collection information to a first user terminal so that a first user of the first user terminal can check the second target policy;

and sending a third target policy with low user quality and the premium charge collection call information to a second user terminal so that a second user of the second user terminal can check the third target policy, wherein the authority of the second user terminal is higher than that of the first user terminal.

A second aspect provides a premium collection apparatus comprising:

the data collection unit is used for determining credit data, payment related data and a paying risk of a target insurance policy based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance policy is to be paid;

a credit evaluation unit for inputting the credit data into a credit evaluation model, the credit evaluation model outputting the applicant credit evaluation result;

the payment probability evaluation unit is used for inputting the payment related data into a payment prediction model, and the payment prediction model outputs a payment probability prediction result of the applicant on the target insurance policy;

the user quality evaluation unit is used for determining the user quality of the applicant based on the credit evaluation result of the applicant, the payment probability prediction result and the paying risk of the target insurance policy;

and the collection urging scheme determining unit is used for determining a collection urging scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection urging scheme, wherein the collection urging scheme comprises the following steps: premium collection call information and premium collection process information.

A third aspect provides a computer device comprising a memory and a processor, the memory having stored therein computer readable instructions, which, when executed by the processor, cause the processor to perform the steps of the premium collection method described above.

A fourth aspect provides a storage medium having stored thereon computer-readable instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of the premium inducement method described above.

The premium collection method, the premium collection device, the computer equipment and the storage medium firstly determine credit data, payment related data and claim risk of the target policy of the applicant based on policy information of the target policy, wherein the target policy is a policy of which the premium is to be paid; secondly, inputting the credit data into a credit evaluation model, and outputting a credit evaluation result of the applicant by the credit evaluation model; inputting the payment related data into a payment prediction model, and outputting a payment probability prediction result of the applicant on the target insurance policy by the payment prediction model; determining the user quality of the applicant again based on the credit evaluation result, the payment probability prediction result and the paying risk of the target insurance policy of the applicant; and finally, determining a collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme. Therefore, based on the relevant data of the applicant, the possibility of the re-purchasing of the dangerous species and the corresponding probability of the re-purchasing of the dangerous species of the client are accurately predicted, the actual requirements of the client are met, and meanwhile, the communication cost is effectively reduced; through multi-channel data fusion, the dynamic perception of market change of data is realized, data push decision and auxiliary realization of product marketing targets are realized, and the purpose of recommending appropriate insurance products to appropriate users in an intelligent manner is finally realized.

Drawings

FIG. 1 is a diagram of an exemplary environment for implementing a premium collection method in one embodiment;

FIG. 2 is a block diagram showing an internal configuration of a computer device according to an embodiment;

FIG. 3 is a flow chart of a premium collection method in one embodiment;

FIG. 4 is a block diagram of a premium hastening device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

It will be understood that, as used herein, the terms "first," "second," and the like may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another.

Fig. 1 is a diagram of an implementation environment of the premium collection method provided in an embodiment, as shown in fig. 1, in which the implementation environment may include a computer device 110 and a terminal 120.

The computer device 110 is a data provider device, and the computer device 110 has an Interface, which may be, for example, an API (Application Programming Interface). The terminal 120 is a target policy input party and has an interface configuration interface, and when the premium is charged, the user can input the target policy through the terminal 120 to make the computer device 110 charge the next premium.

It should be noted that the terminal 120 and the computer device 110 may be, but are not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, and the like. The computer device 110 and the terminal 110 may be connected through bluetooth, USB (Universal Serial Bus), or other communication connection methods, which is not limited herein.

FIG. 2 is a diagram showing an internal configuration of a computer device according to an embodiment. As shown in fig. 2, the computer device may include a processor, a storage medium, a memory, and a network API interface connected by a system bus. The storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store control information sequences, and the computer readable instructions when executed by the processor can enable the processor to realize a premium collection method. The processor of the computer device is used for providing calculation and control capability and supporting the operation of the whole computer device. The memory of the computer device may have stored thereon computer readable instructions that, when executed by the processor, cause the processor to perform a premium collection method. The network API interface of the computer device is used for connecting and communicating with the terminal. Those skilled in the art will appreciate that the architecture shown in fig. 2 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

As shown in fig. 3, in an embodiment, a premium charging method is provided, which can be applied to the computer device 110, and specifically includes the following steps:

step 301, determining credit data and payment related data of an applicant of a target insurance policy and a claim risk of the target insurance policy based on policy information of the target insurance policy; wherein the target policy is a policy of the premium to be paid;

the present invention can be applied to most existing risks, such as health risks for medical reimbursement, and scene risks such as order risks and freight risks, when a user may have a target insurance policy for insurance, the user may initiate the target insurance policy through a corresponding insurance platform, the target insurance policy referred to in step 301 may refer to a target insurance policy directly initiated by the user, or may refer to a target insurance policy initiated by the user or other operations, and the target insurance policy is generated after corresponding process processing (e.g., vehicle insurance needs field damage checking), which is not limited in the embodiments of the present specification. In an application scenario, a regular scanning detection system detects whether a policy call exists, and mainly comprises external policy input and an unsuccessful policy collection. If the policy calling is detected, data acquisition is carried out; otherwise, the policy is called again after a certain time delay. And judging whether the hastening is successful or not. If not, the policy is re-entered into policy calling after a certain time delay; if yes, the next round of detection is performed.

The policy information may include policy identification (e.g., policy number), user identification (e.g., buyer ID), and the claim settlement rules corresponding to the policy information of the target policy, such as the maximum payment amount for the order, whether the target policy can be paid or not corresponding to different reasons for goods return, etc. In addition, credit data of the buyer, such as credit points, historical shopping records, historical return records, historical transaction evaluation records, and the like of the E-commerce platform can be obtained according to the buyer ID. Predetermined rule features can then be extracted from the claims and predetermined credit data can be extracted from the credit data.

Step 302, inputting the credit data into a credit evaluation model, and outputting a credit evaluation result of the applicant by the credit evaluation model;

wherein, the applicant credit assessment result determines whether the applicant meets the preset credit requirement and/or the fraud risk reaches the preset threshold.

In some embodiments, the above steps may include: and calculating the credit score of the applicant through a weighting algorithm based on preset weights according to the acquired credit data, and comparing the credit score with a threshold value in a preset credit requirement to determine the credit evaluation result of the applicant.

Step 303, inputting the payment related data into a payment prediction model, and outputting a payment probability prediction result of the applicant on the target insurance policy by the payment prediction model;

it will be appreciated that the policy information can indicate the applicant's basic data, which can include:

the target user may be any user, and the target user may be a user with insurance requirements. The insurance business information can refer to related business of paying insurance premium to the applicant according to the prearranged agreement, and the applicant undertakes insurance fund responsibility for paying for the insurance fund due to property loss caused by the occurrence of the prearranged accident, or undertakes the behavior of paying the insurance fund responsibility when the applicant dies, is disabled, has diseases or reaches the prearranged age, deadline and the like. Insurance services may include one or more insurance categories, such as personal safety-type insurance, health-type insurance, and property-type insurance, among others. The payment-related data may be information of a user's demand for a certain insurance category or multiple insurance categories, and the payment-related data may include one or more of user age, gender, work condition (which may include work nature, work place, etc.), owned vehicle condition (which may include vehicle purchase age, vehicle model, etc.), driving age, travel condition, health condition, etc.

In some embodiments, the payment prediction model includes at least two pre-trained payment prediction submodels;

the step 303 of inputting the payment-related data into the payment prediction model, where the payment prediction model outputs the prediction result of the payment probability of the applicant for the target insurance policy, may include:

3031, inputting the credit data into payment forecasting submodels respectively, and outputting a payment probability forecasting result of the policyholder to the target insurance policy by each payment forecasting submodule;

step 3032, using a grid search method to search an optimal weight combination, and summing products of each weight in the weight combination and the payment probability prediction result determined by the corresponding payment prediction submodel to obtain a payment probability prediction result.

Wherein, the payment forecasting sub-module may include: the method comprises the steps of utilizing a grid search method to automatically adjust the weight of combinations of lightgbm, xgboost, lr, gbdt, dnn, cnn and the like by utilizing a grid search method, recording errors corresponding to each weight combination, sequencing the weights according to the sequence of the errors from small to large, selecting an algorithm with the first K smallest errors as an algorithm required by final prediction of the feature, predicting future values by using corresponding K optimal parameters to obtain K prediction results, and solving the average value of the K prediction results as the final prediction result.

In some embodiments, the determining the risk of reimbursement for the target policy in step 301 above includes:

3011, obtaining a plurality of historical policy in accordance with the application category of the target policy from the historical claims database according to policy information of the target policy;

step 3012, querying whether there is a claim settled historical policy satisfying a preset condition among the plurality of historical policies, the preset condition being that the applicant of the claim settled historical policy is similar to the applicant of the target policy;

in one embodiment, querying a plurality of historical policies for the presence of a claim-settled historical policy that satisfies a predetermined condition that an applicant of the claim-settled historical policy is similar to an applicant of the target policy comprises:

3012a, determining similar policyholder of the policyholder according to similarity between the basic data of the policyholder and the basic data of each historical policyholder;

and 3012b, determining whether the similar applicant is an applicant of the claim-settled historical policy, and if so, obtaining the claim amount of the claim-settled applicant's historical policy.

Step 3013, when there are claims-settled historical policies that satisfy the preset conditions, determining the claim amount of each claim-settled historical policy and the proportion of the claims-settled historical policies in the historical policies;

step 3014, inputting the payment amount and the percentage of each settled historical policy into a pre-trained risk model for paying, so as to obtain the risk for paying of the target policy.

In this step, the risk model of claim payment may be a decision tree classifier, which inputs the amount of claim payment of the settled historical policy, the similarity between the historical applicant and applicant of the historical policy and the proportion of the historical policy in the historical policy into the root node of the decision tree model; and in each non-root node, performing condition calculation on candidate time slices classified into the current non-root node after condition calculation of the parent node of the current non-root node to obtain the claim risk of the target policy.

Step 304, determining the user quality of the applicant based on the credit evaluation result, the payment probability prediction result and the claim risk of the target insurance policy of the applicant;

step 305, determining a corresponding collection scheme of the applicant based on the preset corresponding relation between the user quality and the collection scheme, wherein the collection scheme comprises: premium collection call information and premium collection process information.

In some embodiments, the user quality of the applicant includes: low, medium and high;

in some application scenarios, the credit evaluation result is a credit score of the applicant, the payment probability prediction result is a probability that the applicant pays the target insurance policy, the claim risk of the target insurance policy refers to a probability that the applicant may pay, and the applicant user quality determination method may be to set weights of the three, and then perform weighted calculation on the three to obtain the user quality of the applicant.

The step 305 may include:

step 305a, outputting premium call-receiving information to the applicant through a target premium payment call-receiving system for a first target premium with high user quality;

step 305b, sending the second target policy with the user quality being middle and the premium call collection information to the first user terminal so that the first user of the first user terminal can check the second target policy;

and 305c, sending the third target policy with low user quality and the premium call collection information to a second user terminal so that a second user of the second user terminal can check the third target policy, wherein the authority of the second user terminal is higher than that of the first user terminal.

It can be understood that both the first user terminal and the second user terminal are provided with corresponding audit authorities, and the audit authority of the second user terminal is higher than the audit authority of the first user terminal. For example: the first user is a common operator, the second user is an expert, the common operator checks the written data of the case with the medium user quality, the examination of the case with the medium user quality can be finished when the checking is finished, the expert needs to carry out deep investigation on the case with the low user quality, and the examination of the case with the low user quality is finished after the investigation is finished.

As shown in fig. 4, in an embodiment, there is provided a premium charging apparatus, which may be integrated in the computer device 110, and specifically includes:

the data collection unit 411 is configured to determine credit data, payment related data, and a risk of paying a target policy of an applicant of the target policy based on policy information of the target policy, where the target policy is a policy to be paid a policy;

a credit evaluation unit 412 for inputting credit data into a credit evaluation model, the credit evaluation model outputting a result of the applicant's credit evaluation;

the payment probability evaluation unit 413 is used for inputting payment related data into a payment prediction model, and the payment prediction model outputs a payment probability prediction result of the applicant on the target insurance policy;

a user quality evaluation unit 414, configured to determine the user quality of the applicant based on the credit evaluation result, the payment probability prediction result, and the risk of reimbursement for the target insurance policy of the applicant;

a collection urging scheme determining unit 415, configured to determine a collection urging scheme corresponding to the applicant based on a preset correspondence between the user quality and the collection urging scheme, where the collection urging scheme includes: premium collection call information and premium collection process information.

In one embodiment, a computer device is provided, which may include a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program: determining credit data, payment related data and a paying risk of a target insurance policy of an applicant of the target insurance policy based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance policy is to be paid; inputting the credit data into a credit evaluation model, and outputting a credit evaluation result of the applicant by the credit evaluation model; inputting the payment related data into a payment prediction model, and outputting a payment probability prediction result of the applicant on the target insurance policy by the payment prediction model; determining the user quality of the applicant based on the credit evaluation result, the payment probability prediction result and the paying risk of the target insurance policy of the applicant; and determining a collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme.

In one embodiment, a storage medium is provided that stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of: receiving a target policy and determining target policy data based on the target policy, the target policy data may include: determining credit data, payment related data and a paying risk of a target insurance policy of an applicant of the target insurance policy based on insurance policy information of the target insurance policy, wherein the target insurance policy is an insurance policy of which the insurance policy is to be paid; inputting the credit data into a credit evaluation model, and outputting a credit evaluation result of the applicant by the credit evaluation model; inputting the payment related data into a payment prediction model, and outputting a payment probability prediction result of the applicant on the target insurance policy by the payment prediction model; determining the user quality of the applicant based on the credit evaluation result, the payment probability prediction result and the paying risk of the target insurance policy of the applicant; and determining a collection prompting scheme corresponding to the applicant based on the preset corresponding relation between the user quality and the collection prompting scheme.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the computer program is executed. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.