KR20190004101A - Apparatus and method for analyzing investment tastes of customer using data mining technique based on a multitude add points of the plural index - Google Patents

Abstract

The present invention relates to an apparatus for analyzing a customer investment tendency using data mining based on a multi-point of a multi-index, which comprises: a query information database connected to any one main area of a customer risk management tendency, an investment period, an income state, an income stability, a health, a consumption tendency, an asset level, an asset type, and a debt type, and storing a plurality of analysis query information connected to a plurality of evaluation areas corresponding to each of the main areas; a query information providing unit for selecting at least one from the analysis query information in response to an investment tendency analysis request of a customer terminal, and providing the same for the customer terminal; a score calculation unit for calculating scores for the evaluation areas connected to the provided analysis query information when answer information on the analysis query information provided for the customer terminal is received; and an investment tendency information generating unit for calculating evaluation scores for each of the evaluation areas by using scores for each of the evaluation areas calculated by the score calculation unit, and generating investment tendency information on a customer on the basis of the calculated evaluation scores.

Description

TECHNICAL FIELD The present invention relates to an apparatus and method for analyzing customer investment propensity using a multi-valued point-based data mining technique,

The present invention relates to an apparatus and method for analyzing customer investment propensity using multiple point-based data mining of a plurality of indices.

In general, portfolio selection theory is widely used as a method for investing and allocating assets most advantageously.

According to the portfolio selection theory, which is one of the strategies for effective asset investment, if the expected return is the same for the asset investment, the combination of the asset type group with low risk is selected and if the risk burden is the same, It is possible to combine the products of the asset type group. While investors may manage their own assets by creating portfolios to manage their own assets, securities firms also provide services for investors visiting their branches to create a portfolio based on current investment products.

Since the composition of this portfolio is varied according to the investment propensity of each individual user, it has been difficult to construct a portfolio that matches his or her propensity. In order to conduct professional consultation, visit the branch office of the relevant securities company and wait for a long time , And it is difficult to receive various information in real time due to limited consulting time.

In addition, since the financial service branch office recommends the products directly to the customers and creates the portfolio, the quality of the portfolio may be different depending on the competence of the staff. Also, the inconsistent investment direction A portfolio can be constructed according to the reliability of the portfolio.

Even if a contract is made based on the creation of a primary portfolio, if it is maintained for a long time, the market will not be able to respond quickly to changing market conditions, and the likelihood of a decline in profitability will increase. To this end, it is extremely difficult for an investor to change a portfolio composed of various assets (adjustment of asset group allocation) in consideration of all market conditions.

Korean Patent Publication No. 10-2017-0034647 (published on Mar. 29, 2017)

The present invention provides analytical query information, which is linked to a main domain and a plurality of evaluation domains, to a customer, and calculates an evaluation score for each customer based on the answer information, thereby analyzing the customer's investment propensity information Provided is an apparatus and method for analyzing customer investment propensity using multi-valued point-based data mining of multiple indicators.

In order to solve the problems to be solved, the apparatus for analyzing customer investment propensity using multi-point based data mining according to an embodiment of the present invention includes a customer's risk management tendency, investment period, income status, A query information database storing a plurality of analysis query information linked to one of the main regions of the main region, the consumption tendency, the asset level, the asset type and the debt type, and connected to a plurality of evaluation regions corresponding to each of the main regions, A query information providing unit for selecting at least one of the plurality of query information for analysis according to a request for analysis of investment propensity of the client terminal and providing the selected query information to the client terminal; When the answer information is received, a plurality of evaluations A score calculating unit for calculating a score for each area based on the score obtained by the score calculating unit and a score for each evaluation area calculated by the score calculating unit, And an investment propensity information generating unit for generating information.

According to an embodiment of the present invention, the query information providing unit selects one query information for analysis from the query information database and provides the selected query information to the client terminal, and when the response information is received from the client terminal, A plurality of analytic query information can be provided by searching analytic query information related to information from the query information database.

According to an embodiment of the present invention, the investment propensity information generating unit counts the number of analytical query information for each main area as the analytical query information provided to the client terminal reaches a preset number, It is possible to calculate the evaluation score for each evaluation area by a method of weighting the scores calculated by the answer information according to the number of query information for each area.

According to an embodiment of the present invention, the investment propensity information generating unit assigns a weight greater than a reference value to a score of the same evaluation area as that of the main area linked to the analysis query information, It is possible to calculate the evaluation score for the evaluation area by assigning a weight smaller than the reference value and calculating the score to which the large weight is assigned or the small weight is assigned.

According to an embodiment of the present invention, the investment propensity information generating unit may calculate an average value for each evaluation area by summing the scores for the evaluation areas, averaging the scores, analyzing a correlation between the average value and the score, It is possible to calculate the evaluation score for the evaluation area by using the remaining score after excluding the score that exceeds the predetermined first value or less than the predetermined second value.

In order to solve the problems to be solved, a method for analyzing customer investment propensity using data mining based on multiple points of a plurality of indices according to an embodiment of the present invention includes a customer's risk management tendency, investment period, income status, A query information database which is connected to one of the main areas of the main area of the main area, the main area of the asset type, the asset type and the debt type, and which is connected to a plurality of evaluation areas corresponding to each of the main areas, A method of analyzing an investment propensity of a customer, comprising the steps of: selecting at least one of the plurality of analytical query information according to an investment tendency analysis request of a client terminal, and providing the at least one or more analytical query information to the client terminal; When the answer information for the usage query information is received, Calculating a score for a plurality of evaluation regions connected to the provided analysis query information by using the evaluation score for each evaluation region using the scores for the evaluation regions calculated; And generating the investment propensity information of the customer based on the score.

According to an embodiment of the present invention, the step of providing to the client terminal includes selecting one analysis query information from the query information database and providing the selected query information to the client terminal, It is possible to search for and provide analysis query information related to the received answer information in the query information database, and to provide a plurality of query information for analysis repeatedly.

According to an embodiment of the present invention, the step of calculating the evaluation score includes the steps of counting the number of analysis query information for each main area as the analysis query information provided to the client terminal reaches a predetermined number, And calculating the evaluation score for each evaluation area in such a manner that the score calculated as the answer information is weighted and calculated according to the counted number of analysis query information for each main area.

According to an embodiment of the present invention, the step of calculating the evaluation score includes the steps of: giving a first weight greater than a reference value to a score of the evaluation area that is the same as the main area connected to the analysis query information; And calculating a score for the evaluation area by calculating a score to which the first or second weight is assigned after assigning a second weight smaller than the reference value to the score of the evaluation area.

According to the embodiment of the present invention, the step of calculating the evaluation score includes a step of calculating an average value for the evaluation area by summing the scores for one of the evaluation areas among the plurality of evaluation areas and averaging the scores, Analyzing a correlation between the scores of any one of the evaluation regions and excluding a score exceeding a predetermined first value or a predetermined second value as a result of the analysis; And calculating an evaluation score for the evaluation area.

According to an aspect of the present invention, there is provided a method for providing analytical query information to a customer, the analytical query information including a main area and a plurality of evaluation regions linked to each other, By providing a device and method for analyzing customer investment propensity using multi-point based data mining capable of analyzing investment propensity information, it is possible to improve accuracy of analysis of customer's propensity to invest.

In addition, according to the present invention, it is possible to analyze the future propensity of the customer and the investment propensity of the customer by analyzing the customer propensity by including the information related to the customer's health.

In addition, according to the above-described task solution of the present invention, it is possible to prevent any one-sided analysis by calculating the evaluation score for each area for analyzing the customer investment propensity based on the correlation and the weight.

1 is a network configuration diagram showing an overall connection relationship of a robo advisor system according to an embodiment of the present invention.
2 is a block diagram showing a detailed configuration of a robot advisor system according to an embodiment of the present invention.
3 is a block diagram illustrating a detailed configuration of an investment propensity analysis server according to an embodiment of the present invention.
4 is a diagram for explaining the structure of query information for analysis according to an embodiment of the present invention.
5 is an exemplary diagram illustrating investment propensity information generated by an investment propensity analysis server of a robo advisor system according to an embodiment of the present invention.
FIG. 6 is a flowchart illustrating a process of providing analysis query information to a client terminal in an investment propensity analysis server according to an embodiment of the present invention.
FIG. 7 is a flowchart illustrating a process in which an investment propensity analysis server according to an embodiment of the present invention generates investment propensity information of a customer.
FIG. 8 is a flowchart illustrating a process of creating a recommended portfolio using the robo advisor system according to an embodiment of the present invention.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to provide a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, this is merely an example and the present invention is not limited thereto.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intention or custom of the user, the operator, and the like. Therefore, the definition should be based on the contents throughout this specification. The terms used in the detailed description are intended only to describe embodiments of the invention and should in no way be limiting.

1 is a network configuration diagram showing an overall connection relationship of a robo advisor system according to an embodiment of the present invention and includes a customer terminal 100, a financial company interworking platform 200, a robot advisor system 300, and the like have. At this time, the robot advisor system 300 and the customer-side terminal 100 may be connected through a wired or wireless communication network (not shown).

As shown in FIG. 1, the financial institution interworking platform 200, which is interlocked with the robot advisor system 300 and provides data on financial products, may include a securities company API, a bank API, an insurance API, , A securities company, a bank, an insurance company, and other financial institutions, and then provide the corresponding data to the robot adviser system 300. [

The customer-side terminal 100 can access the robot advisor system 300 and display and receive data corresponding to the investment propensity analysis and the financial product recommendation service.

Such a client-side terminal 100 is a computing device that can be connected to a network and can execute various types of applications. Examples of the computing device include a notebook computer, a personal computer, a smart phone, and the like. Specifically, the computing device corresponding to the customer-side terminal 100 is provided with a communication circuit 102 capable of supporting connection to the robo advisor system 300, and various services through interlocking with the robo advisor system 300 At least one processor 106 for executing applications, a display unit 108 for displaying various services in an identifiable form by the user, various inputs and outputs And an input / output circuit 110 for processing the input / output signals.

The robot advisor system 300 analyzes a customer's investment propensity and analyzes a propensity for a financial product, and generates a recommended portfolio composed of at least one financial product matching the investment propensity of the customer through matching between the analyzed information , And can provide the generated recommended portfolio to the customer-side terminal 100.

Also, the robo advisor system 300 checks the responsiveness of the financial product based on the investment-related information collected by the external information collection server 308, and updates the tendency analysis information of the financial product according to the responsiveness It is possible to change the financial products in the recommended portfolio through the re-matching process between the tendency analysis information of the financial product and the investment propensity of the customer.

The robot adviser system 300 is based on a process of deriving a customer investment propensity, that is, a process of analyzing query information and response information provided to a customer, a process of selecting a financial product according to a customer's investment propensity, Investment logic can be created.

Meanwhile, the robot advisor system 300 verifies the investment logic through the analysis of the yield of the financial products in the recommended portfolio as a predetermined period, for example, 3, 6, 9 and 12 months, and then, based on the result of the verification, Financial products can be changed. Specifically, the robo advisor system 300 not only checks the responsiveness of the external information to the financial products having low returns, but also reflects the opinions of the experts, It is possible to detect an error part of the investment logic by judging the degree of mismatch between information and to select a financial product that can complement the detected error part and use it to change the financial product in the recommended portfolio.

Hereinafter, the detailed configuration and operation of the robot advisor system 300 will be described with reference to FIG.

2 is a block diagram showing the detailed configuration of the robot advisor system 300 according to the embodiment of the present invention.

2, the robot advisor system 300 includes an investment propensity analysis server 302, a financial product analysis server 304, a recommended goods matching server 306, an external information collection server 308, Server 310 and automated teller machine support server 312 and may additionally include a membership information database 314, a financial product database 316 and an investment logic database 318.

First, the investment propensity analysis server 302 provides analytic query information in a plurality of query / response formats to the client side terminal 100, receives answer information therefrom, and based on the received response data, The investment propensity information of the customer can be generated by analyzing the investment propensity. 3, the investment propensity analysis server 302 includes a query information database 322, a query information provider 324, a score calculation unit 326, and an investment propensity information generation unit 328, , ≪ / RTI >

As shown in FIG. 4, the query information database 322 stores a plurality of query information for analysis connected to one main area 320a and to which a plurality of evaluation areas 320b are connected. Here, the main area 320a connected to the analysis query information is any one of a risk management tendency, an investment period, health, income stability, consumption status, asset level, asset type, The evaluation area 320b may include all of the data corresponding to the main area 320a. At this time, the evaluation area 320b corresponding to the main area 320a connected to the analysis query information among the plurality of evaluation areas 320b may be the main evaluation area 320b. In other words, any one of the plurality of evaluation areas 320b may be the same data as the main area 320a. For example, the analysis query information linked to the main area 320a of the risk management tendency may be connected to a plurality of evaluation areas 320b including the evaluation area 320b corresponding to the risk management tendency.

On the other hand, the analytical query information is divided into a customer's personal information group, an import / expense group, and an asset management group, and each group can include a plurality of query information for analysis.

The query information providing unit 324 selects at least one or more of the plurality of query information for analysis stored in the query information database 322 according to the investment molding analysis request of the client terminal 100 and transmits the selected query information to the customer terminal 100 .

The query information providing unit 324 searches the query information database 322 for one query information for analysis and provides the query information to the client terminal 100. After the response information is received from the client terminal 100, And the analysis-use query information is retrieved from the query information database 322 and provided to the client-side terminal 100 by providing the query information in association with the answer information. Specifically, the query information providing unit 324 receives a score for a plurality of evaluation regions connected to the query information for analysis calculated by the score calculating unit 326 from the score calculating unit 326, The user can search the query information database 322 for analysis query information linked to the selected main area and select the main area based on the score for the area, and provide the search query information to the client side terminal 100. [ For example, if the score of the evaluation area related to the investment period is high among the scores of the plurality of evaluation areas calculated based on the answer information after providing the query information for analysis related to the risk management tendency, Can search the query information database 322 for analysis query information having the investment period as the main domain and provide the query information to the client terminal 100. [

On the other hand, the query information providing unit 324 counts the number of query information for analysis provided for each main area whenever the query information for analysis is provided to the client terminal 100, and provides the information to the investment propensity information generating unit 328 .

The score calculation unit 326 provides analysis query information to the client side terminal 100 by the query information provision unit 324 and then transmits the analysis query information to the client side terminal 100 using the response information received from the client side terminal 100 And may provide the calculated score to the query information providing unit 324 or the investment propensity information generating unit 328. [

The investment incentive information generating unit 328 evaluates the investment incentive information based on the scores of the plurality of evaluation areas for analysis use query information provided from the score calculating unit 326 after the predetermined analytical query information is provided to the client terminal 100 It is possible to calculate the score for each area.

The investment propensity information generating unit 328 may calculate the evaluation score by reflecting the number of analytical query information for each main area provided from the query information providing unit 324 when calculating the evaluation score for each evaluation area. Specifically, the investment propensity information generation unit 328 calculates the investment propensity information based on the answer information for the analysis query information when the number of analysis query information for a specific main area exceeds a predetermined number (exceeds a predetermined threshold value) It is possible to calculate an evaluation score for an evaluation area corresponding to a specific main area by calculating a predetermined weight to a score associated with the evaluation area corresponding to the specific main area among the scores of the plurality of evaluation areas. For example, when the evaluation score for the evaluation area of the risk management tendency is calculated while the analytic query information linked to the risk management tendency is provided to the customer side terminal 100 exceeding the predetermined number, (For example, a weight below the reference value) to the score for the evaluation area of the risk management tendency calculated from the analysis query information connected to the main area other than the risk management tendency, The score for the risk management tendency can be calculated by calculating the score for the evaluation area of the risk management tendency calculated from the calculated scores and the quality information for analyzing the risk management tendency.

When calculating the evaluation score for each evaluation area using the answer information for the analysis query information, if the evaluation area is the same as the main area of the analysis query information, the investment tendency information generation unit 328 If the evaluation area is different from the main area of the analytical query information, the weight is given to the evaluation score and a score below the reference value is calculated, and the score is calculated to calculate the evaluation score of the evaluation area Can be calculated. For example, when calculating the evaluation score for the evaluation area of the asset type, the score of the evaluation area of the asset type obtained from the answer information of the analysis query information connected to the main area of the asset type is given a weight that is larger than the reference value, And calculating a score by assigning a weight below the reference value to the score of the evaluation area of the asset form acquired from the answer information of the analysis query information connected to the main area other than the main area of the asset form The evaluation score for the evaluation area of the asset type can be calculated.

On the other hand, the investment propensity information generating unit 328 can calculate the evaluation score based on the number of pieces of analysis query information for each main region provided to the customer-side terminal 100 in calculating the evaluation score for each evaluation region. Specifically, when the number of analytical query information from which a specific evaluation region for which an evaluation score is to be calculated is greater than or equal to a predetermined number, the investment tendency information generation section 328 calculates the investment tendency information of analytical query information The score of the specific evaluation area derived from the answer information, and the score of the specific evaluation area derived from the answer information of the analytical query information of the specific evaluation area and the main area are assigned different weights, Can be calculated.

On the other hand, the investment propensity information generation unit 328 calculates the evaluation score for each evaluation area by the above-described method, and generates the investment propensity information of the customer through analysis of the calculated evaluation score for each evaluation area.

For example, as shown in FIG. 5, the investment propensity information generating unit 328 calculates the score calculated based on the answer information on the query information for analysis related to health to calculate the evaluation score, It is possible to generate analysis summary information such as customer's psychology, income stability and consumption propensity through analyzing the score. Also, if analysis query information related to family history is further included in the quality query information related to health, Psychology, income stability, and consumption propensity. Then, based on this information, customers' investment propensity information can be generated.

In the embodiment of the present invention, the investment propensity information may be information in the form of a combination of short-term, medium-term, long-term, offensive (dangerous), stable, For example, information on investment propensity can be information such as investment access needs such as medium-term attack, medium-term stabilization, short-term attack, and short-term stabilization.

Meanwhile, the investment incentive analysis server 302 according to an embodiment of the present invention may interwork with an external information providing server, for example, a hospital server (not shown), a bank server (not shown) Asset status, expenditure status, etc., and can be used to generate investment propensity analysis information. Specifically, the investment propensity analysis server 302 can receive hospital history information, asset status, expenditure status, and the like of a customer and its family by being connected to a hospital server, a bank server, and the like based on the agreement of the customer.

The answer information and the investment propensity information analyzed by the investment propensity analysis server 302 as described above can be stored in the member information database 314.

On the other hand, the investment propensity analysis server 302 provides an interface for receiving customer search information to the customer-side terminal 100, collects site information, query word information, etc., used by the customer through the interface, Can be changed. Specifically, the investment propensity analysis server 302 monitors the behavior of the customer on the customer side terminal 100, that is, on the customer side terminal 100 registered at the time of membership registration, when the customer who is a member of the robot advisor system 300 For example, a web beacon to the customer-side terminal 100, collects site information, search word information, and the like used by the customer through the provided web beacon, derives a customer interest field based on the collected site information, The investment propensity information can be updated based on the field.

A process of analyzing a customer's investment propensity by the investment propensity analysis server 302 having the above-described configuration will be described with reference to FIGS. 6 and 7. FIG.

6 is a flowchart illustrating a process of providing analytic query information to the client side terminal 100 in the investment propensity analysis server 302 according to an embodiment of the present invention.

6, when a service subscription or investment propensity analysis request is first received from the customer terminal 100, the investment propensity analysis server 302 analyzes an analysis (analysis) connected to one of the main areas in the query information database 322, In operation S600, the number of times of providing the query information for analysis is counted in operation S602.

Then, when answer information for a plurality of analytical query information is received from the client side terminal 100 (S604), the investment tendency analysis server 302 calculates a plurality of evaluation regions (S606).

Thereafter, the investment propensity analysis server 302 determines whether the number of times of provision has reached a predetermined number (S608).

If it is determined in step S608 that the provided number of times has not reached the preset number of times, the investment propensity analysis server 302 selects the main area based on the score for the plurality of evaluation areas (S610), and then proceeds to step S600 . That is, the investment propensity analysis server 302 searches the query information database 322 for analysis query information linked to the selected main area, and provides the query information to the customer terminal 100.

On the other hand, if it is determined in step S608 that the number of times of provision has reached the preset number of times, the investment propensity analyzing server 302 may calculate the number of times of providing the evaluation domain by using the scores of the plurality of evaluation areas, (S612), and then generates the investment propensity information of the customer based on the score. This will be described with reference to FIG.

7 is a flowchart illustrating a process of generating investment propensity information of a customer by the investment propensity analysis server 302 according to an embodiment of the present invention.

7, first, the investment propensity analysis server 302 calculates a score of a plurality of evaluation regions calculated on the basis of the answer information for analytical query information connected to the first main region provided to the customer-side terminal 100 (Hereinafter referred to as " same area score ") of the evaluation area corresponding to the first main area among the plurality of evaluation areas (S700).

Then, the investment propensity analysis server 302 calculates an evaluation domain corresponding to the first main area among the scores of the plurality of evaluation domains calculated on the basis of the answer information on the analysis query information linked to the main domain other than the first main domain, (Hereinafter referred to as " another area score ") (S702).

Thereafter, the investment propensity analysis server 302 calculates a first evaluation value by assigning a first weight value (e.g., a value larger than a reference value) to the same area score (S704), and assigns a second weight value (Step S706), the first and second evaluation values are calculated to calculate the evaluation score for the first main area (i.e., the first evaluation area) (S708) .

The above steps are repeatedly performed to calculate the evaluation score for the evaluation area corresponding to each main area.

Thereafter, the investment propensity analysis server 302 generates the investment propensity information of the customer based on the evaluation score of each evaluation area (S710).

Before performing the step of weighting, the investment propensity analysis server 302 determines that only the main region whose number of analysis query information provided to the client side terminal 100 is greater than or equal to a predetermined number is provided to the client side terminal 100 for each main region, The evaluation score can be calculated using the grant method. In this case, the investment propensity analysis server 302 can calculate the score of the evaluation area by calculating a score for the evaluation area when the counted number is less than a predetermined number.

According to the embodiment of the present invention described above, the evaluation score for each evaluation area is calculated by the weighting method, but it is also possible to calculate the evaluation score for each evaluation area through correlation analysis. In this case, when a correlation between the average value of the scores of the evaluation region and the score of the evaluation region, for example, a correlation function is applied and the resultant value exceeds a predetermined value, for example, 0.5 to 0.7 (a high correlation value) -0.7, and an evaluation score for the evaluation area can be calculated using scores other than the excluded scores.

The financial product analysis server 304 receives financial data, such as financial product data, in cooperation with a securities company server, a bank server, and an insurance company server through a financial company interworking platform 200 and analyzes the received financial product data, And store it in the financial product database 316. [0040] FIG. Specifically, the financial product analysis server 304 collects reputation information of a manager using manager information for managing financial products using financial data, determines the maturity level of the financial product investment, Etc., and generate tendency information on the financial product based on the analysis.

The tendency information of the financial product generated by the financial product analysis server 304 according to the embodiment of the present invention has a form of combination of the medium term, short term, long term, attack type (risk type), insurance type, And may be stored in the financial product database 316.

On the other hand, the propensity information of the financial product analyzed by the financial product analysis server 304 may be changed according to external environmental factors. Specifically, the financial product analysis server 304 checks the responsiveness of each financial product based on the environmental factor information, and may change the tendency information of the financial product if the responsiveness is equal to or greater than a predetermined threshold value.

To this end, the robo advisor system 300 according to the embodiment of the present invention may further include an external information collection server 308. [

The external information collection server 308 can collect environmental factor information related to the investment in cooperation with various external servers such as a search server, a rate providing server, a price management server, a media information providing server, and the like. Herein, the environmental factor information includes big data analysis information on the search keyword, price data, news information recently issued, interest rate information, and the like. Big data analysis information includes a search keyword by age group, a search keyword distribution status Information.

The recommended goods matching server 306 searches at least one financial product in the financial product database 316 based on the investment propensity information of the customer and can generate an investment portfolio composed of the retrieved financial products. Specifically, the recommended goods matching server 306 searches the financial product database 316 for a financial product matching the investment propensity information, for example, a financial product related to the insurance property, a safe asset type financial product, a high risk asset type financial product, A recommended asset portfolio type recommended portfolio and a high risk asset type recommended portfolio according to the investment propensity of the customer, and generates a comprehensive recommendation portfolio based on the created portfolio. The generated portfolio is transmitted to the client side terminal 100, Lt; / RTI >

Also, the recommended product matching server 306 generates investment logic based on information on the matching process, for example, a process of calculating investment propensity information and a financial product matching process, and transmits the generated investment logic to the investment logic database 318. [ Lt; / RTI >

The investment logic verification server 310 is a device that operates after a predetermined time elapses after providing a comprehensive recommendation portfolio to a customer. The investment logic verification server 310 calculates a return on a financial product in a comprehensive recommendation portfolio, The return rate set by the adviser system or the return rate set for the financial product) is compared with the standard rate of return, it is possible to check the investment logic through the verification process of the information of the customer's investment propensity and the tendency of the financial product. Specifically, the investment logic verification server 310 not only checks the responsiveness of external information to the financial products whose yields are lower than the standard rate of return among the financial products in the recommendation portfolio, but also reflects the opinions of the experts, The degree of mismatch between the tendency analysis information of the financial product is determined to detect an error portion of the investment logic, and the investment logic stored in the investment logic database 318 is updated based on the detected error portion. In addition, After selecting a financial product, it can be used to change the financial product in the comprehensive recommendation portfolio.

In addition, the investment logic verification server 310 receives the evaluation information for each of the financial products in the comprehensive recommendation portfolio from the client side terminal 100 after a predetermined time has elapsed after providing the comprehensive recommendation portfolio, , The error rate between the propensity information for each of the financial products in the comprehensive recommendation portfolio and the investment propensity information for the customer can be calculated and the financial product in the comprehensive recommendation portfolio can be changed based on the error rate.

The automated teller machine support server 312 provides the customer's terminal 100 with a comprehensive recommendation portfolio, and then provides customer information to a service server (not shown) that sells the financial instrument selected by the customer, And so on.

A process of generating a recommendation photopolis using the robot advisor system 300 having the above-described configuration will be described with reference to FIG.

FIG. 8 is a flowchart illustrating a process of generating a recommended portfolio using the robo advisor system 300 according to an embodiment of the present invention.

8, the robot advisor system 300 is connected to a securities company, a bank, an insurance company, and other financial companies through a financial company interworking platform 200, receives data on a plurality of financial products, The manager's reputation information, the market maturity invested by the financial product, the investment scale of the financial product, and the like, and generates trend information for a plurality of financial products based on the analyzed information (S800).

Thereafter, the robot advisor system 300 matches trend information to a plurality of financial product names and stores them in the financial product database 316 (S802).

Next, the robot advisor system 300 collects environmental factor information that influences the investment (S804), and checks the degree of reactivity to each of the financial products according to the collected environmental factor information, that is, the degree of influence on the financial product (S806). If the checked reactivity exceeds the predetermined threshold, the tendency information on the financial product exceeding the threshold is re-analyzed to update the tendency information of the financial product (S808). Specifically, the robot advisor system 300 can change the trend information of the financial product using the checked reactivity.

Then, the robot advisor system 300 provides a plurality of query information for analysis to the client terminal 100 in response to a request of a customer, that is, a request for membership of a customer, (S810).

The robot advisor system 300 then determines the risk management tendency, the investment period, the income status, the income stability, the health, the consumption propensity, the asset level, the asset type and the like of the customer based on the answer information received from the customer- At least two of the types of the debt are analyzed to generate the investment propensity information of the customer (S812).

Then, the robo advisor system 300 searches the financial product database 316 for a financial product having propensity information matching the investment molding information, generates a recommended portfolio composed of the retrieved financial products, To the customer-side terminal 100 (S814). As a result, customers can invest in financial products based on a recommended portfolio.

When a predetermined period of time has elapsed since the investment in the financial product, the robo advisor system 300 calculates a return rate for each of the financial products in the recommended portfolio (S816), and calculates the return rate and the standard rate of return Or a financial product having a return rate lower than the standard rate (step S818).

As a result of the determination in S818, if there is a financial product, the robo advisor system 300 checks the investment logic through the verification process for the propensity information on the financial product having the return rate lower than the standard return rate and the information about the customer's investment propensity (S820 ). Specifically, the investment logic verification server 310 not only checks the reactivity of the financial products in the recommended portfolio with respect to the external information for the financial products whose yield is lower than the standard rate of return, but also reflects the opinions of the experts, (Step S822). In step S822, the financial product in the recommended portfolio is changed using the selected financial product to detect the error part of the investment logic by determining the degree of mismatch between the tendency analysis information of the product, .

It will be understood by those of ordinary skill in the art that the foregoing description of the embodiments is for illustrative purposes and that those skilled in the art can easily modify the invention without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

It is intended that the present invention covers the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents. .

100: Customer side terminal
200: Interlocking platform of financial companies
300: RoboAdvisor System
302: Investment propensity analysis server
304: financial instrument analysis server
306: Recommended product matching server
308: External information collection server
310: Investment logic validation server
312: Automated marketing support server
314: Member information database
316: financial product database
322: Query information database
324: Query information provisioning
326: score calculation unit
328: Investment propensity information generating unit

Claims (10)

Wherein the main region is connected to one of main regions of a customer's risk management tendency, investment period, income status, income stability, health, consumption propensity, asset level, asset type and debt type, A query information database in which a plurality of analysis query information linked to the evaluation area is stored,
A query information providing unit for selecting at least one of the plurality of query information for analysis according to a request for analysis of investment propensity of the client terminal and providing the selected query information to the client terminal;
A score calculation unit for calculating a score for a plurality of evaluation areas connected to the provided query information for analysis, when the answer information for the analysis query information provided to the client terminal is received;
And an investment propensity information generating section for calculating an evaluation point for each evaluation area by using the score for each evaluation area calculated by the score calculating section and generating investment propensity information of the customer based on the calculated evaluation score A Customer Investment Propensity Analysis System Using Multiple Point - based Data Mining of Indicators.
The method according to claim 1,
The query information providing unit,
Selecting one analysis query information from the query information database and providing the analysis query information to the client terminal, and receiving answer information from the client terminal, the analysis query information correlated with the received answer information is stored in the query information An apparatus for analyzing customer investment propensity using multiple point - based data mining of multiple indexes providing multiple analytic query information in a manner that is searched and provided in a database.
3. The method of claim 2,
Wherein the investment propensity information generating unit comprises:
The number of analysis query information for each main area is counted as the number of analysis query information provided to the client terminal reaches a predetermined number and is calculated as answer information according to the counted number of analysis query information for each main area An apparatus for analyzing customer investment propensity using multi - point - based data mining that calculates evaluation scores for each evaluation area by calculating a weighted score.
3. The method of claim 2,
Wherein the investment propensity information generating unit comprises:
A weighting unit that assigns a weight greater than a reference value to a score of the same evaluation area as the main area connected to the analysis query information and assigns a weight smaller than a reference value to scores of the evaluation areas other than the main area, An apparatus for analyzing customer investment propensity using multi - point - based data mining that computes scores given small weights and calculates evaluation scores for the evaluation area.
The method according to claim 1,
Wherein the investment propensity information generating unit comprises:
Calculating a mean value for each of the evaluation areas by summing the scores of the evaluation areas, and averaging the scores, analyzing a correlation between the average value and the score, and analyzing a correlation between the average value and the score, A system for analyzing customer investment propensity using multi - point - based data mining that calculates score for the evaluation area using the remaining scores after excluding the score.
Wherein the main region is connected to one of main regions of a customer's risk management tendency, investment period, income status, income stability, health, consumption propensity, asset level, asset type and debt type, A method for analyzing a customer's investment propensity using a query information database in which a plurality of analysis query information linked to an evaluation region is stored,
Selecting at least one of the plurality of analytical query information according to a request for analyzing investment propensity of the client terminal, and providing the at least one analytical query information to the client terminal;
Calculating a score for a plurality of evaluation regions connected to the provided query information using the received answer information when the response information for the analysis query information provided to the client terminal is received;
Calculating an evaluation score for each evaluation area by using the score for each of the evaluation areas calculated;
And generating the investment propensity information of the customer based on the calculated evaluation score.
The method according to claim 6,
Wherein the step of providing to the customer-
Selecting one analysis query information from the query information database and providing the query information to the client terminal;
Searching the query information database for analysis query information correlated with the received answer information as response information is received from the client terminal, and providing the search query information to a plurality of search query information A Methodology for Analyzing Customer Investment Propensity Using Multiple Point - based Data Mining of Indicators.
8. The method of claim 7,
The step of calculating the evaluation score includes:
Counting the number of analysis query information for each main area as the number of analysis query information provided to the client terminal reaches a predetermined number;
A step of calculating an evaluation score for each evaluation area by assigning weights to the scores calculated by the answer information according to the counted number of analysis query information for each main area, How to analyze customer propensity using.
8. The method of claim 7,
The step of calculating the evaluation score includes:
Assigning a first weight greater than a reference value to the score of the evaluation area that is the same as the main area connected to the analysis query information;
Calculating a score assigned to the evaluation area by calculating a score to which the first or second weight is assigned after assigning a second weight smaller than the reference value to the score of the evaluation area different from the main area; Analysis of customer investment propensity using multi - point - based data mining.
The method according to claim 6,
The step of calculating the evaluation score includes:
Calculating a mean value for an evaluation area by summing scores of any one of the evaluation areas and averaging the scores;
Analyzing a correlation between the average value and a score for one of the evaluation regions, and then excluding a score exceeding a predetermined first value or less than a predetermined second value as a result of the analysis;
And calculating an evaluation score for the evaluation area by using scores other than the excluded scores.

Images