KR102565219B1 - Diversified investment recommendation system based on big data-based investment propensity analysis results - Google Patents

Abstract

This specification relates to a diversified investment recommendation system based on big data-based investment propensity analysis results and a diversified investment method using the same. The investment method according to the system includes the steps of adding an online investment-linked financial product selected by an investor to a wish list; When the investor decides to invest in an online investment-linked financial product selected from among online investment-linked financial products included in the wish list, receiving an investment decision command from the investor from an investor terminal; generating an investment contract session for concluding an investment contract for all of the selected online investment-linked financial products; concluding an investment contract for all of the selected online investment-linked financial products collectively through the investment contract session; and completing an investment by performing an investment procedure on all online investment-linked financial instruments for which the investment contract has been concluded.

Description

A diversified investment recommendation system based on big data-based investment propensity analysis results

This specification relates to an investment system for online investment-linked financial products and an investment method using the same. It is about how to recommend.

Recently, unlike the loan system of the existing financial sector, online investment-linked finance, which provides loans to individuals by investing an unspecified number of money based on the Internet, is spreading.

Companies that provide online investment-linked financial services sell loan products with loan conditions based on the evaluation results of the loan applicant's collateral and financial credit quality to loan applicants, and to investors by selling investment products based on the loan conditions of loan products. We provide a service that connects loans and investments by injecting the investment money received from customers into loan funds for loan products.

As an example of an invention related to the above-mentioned online investment-linked financial service, "a third party collateral-based loan provision system, method and program through online" (Korean Patent Registration No. 10-1662014) has been filed.

Korean Patent Registration No. 10-1662014, 2016.10.04.

However, when investing in multiple online investment-linked financial products for diversified investment, investors have the inconvenience of having to individually write an investment contract for each investment product and apply for investment. Therefore, the need for a distributed investment service that solves these inconveniences to investors is increasing.

In addition, companies providing online investment-linked financial services recommend or provide investment products according to investors' investment propensity, so there is an increasing need for services that can provide investment products suitable for investors.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

This specification presents a diversified investment method using a diversified investment recommendation system based on big data-based investment propensity analysis results. The method includes adding an online investment-linked financial product selected by an investor to a wish list; When the investor decides to invest in an online investment-linked financial product selected from among online investment-linked financial products included in the wish list, receiving an investment decision command from the investor from an investor terminal; generating an investment contract session for concluding an investment contract for all of the selected online investment-linked financial products; concluding an investment contract for all of the selected online investment-linked financial products collectively through the investment contract session; and completing an investment by performing an investment procedure on all online investment-linked financial instruments for which the investment contract has been concluded.

The diversified investment method and other embodiments using the diversified investment recommendation system according to the big data-based investment propensity analysis results may include the following features.

According to an embodiment, the method may further include collecting product information about an online investment-linked financial product in which the investment is completed; analyzing the investment propensity of the investor by inputting the collected product information into an investment propensity analysis engine; and recommending an investment product to the investor based on the analysis result of the investment propensity.

According to an embodiment, the step of analyzing the investment propensity of the investor by inputting the collected product information into the investment propensity analysis engine includes the investment experience questionnaire collected from the investor, the investment propensity diagnosis questionnaire, and the investor's personality propensity. The method may include determining an investment propensity of the investor based on at least one of the survey results.

According to an embodiment, the step of recommending an investment product to the investor based on the investment propensity analysis result may include the loan applicant's collateral value, the reliability of the loan applicant's collateral value evaluation result, the loan applicant's credit score, the loan applicant Classifying the investment product according to the risk level based on the annual income, investment period, and delinquency rate and insolvency rate of the same existing investment product; and recommending investment products classified according to the degree of risk based on a result of determining the investment propensity of the investor.

According to an embodiment, the method may further include calculating an investment success rate between the recommended investment products; and adjusting weights of analysis elements of the investment propensity analysis engine based on the investment success rate.

On the other hand, this specification presents a device for recommending diversified investment according to the results of analyzing big data-based investment propensity. The device for recommending diversified investment according to the big data-based investment propensity analysis result includes a communication unit; storage unit; and a control unit functionally connected to the communication unit and the storage unit, wherein the control unit invests in the selected online investment-linked financial product after the investor puts the online investment-linked financial product in the wish list created in the storage unit. When a decision command is issued, the investment decision command is received from the investor's terminal, and an investment contract session is created for creating an investment contract for all of the selected online investment-linked financial products among the online investment-linked financial products stored in the wish list. , Investment contracts for all selected online investment-linked financial products are concluded collectively through the investment contract session, and the investor can complete the investment by performing the investment procedure for all online investment-linked financial products selected from the wish list. .

The diversified investment recommendation device according to the big data-based investment propensity analysis result and other embodiments may include the following features.

According to an embodiment, the control unit collects product information about the online investment-linked financial product for which the investment has been completed, inputs the collected product information to an investment propensity analysis engine to analyze the investment propensity of the investor, and An investment product may be recommended to the investor based on the investment propensity analysis result.

Depending on the embodiment, the control unit may also determine the investment propensity of the investor based on at least one of an investment experience questionnaire collected from the investor, an investment propensity diagnosis questionnaire, and a result of a personality trait survey of the investor.

According to an embodiment, the controller may also determine the loan applicant's collateral value, the reliability of the loan applicant's collateral value evaluation result, the loan applicant's credit score, the loan applicant's annual income, the investment period, and the delinquency rate and insolvency rate of the same existing investment product Based on the above, investment products may be classified according to risk levels, and investment products classified according to the risk levels may be recommended based on the result of determining the investment propensity of the investor.

Depending on the embodiment, the control unit may also calculate a success rate of investment among the recommended investment products, and adjust a weight of an analysis element of the investment propensity analysis engine based on the success rate of investment.

According to the embodiments disclosed in this specification, when an investor makes a diversified investment in a plurality of online investment-linked financial products, there is an effect of removing the inconvenience of having to individually conclude an investment contract.

In addition, according to the embodiments disclosed herein, there is an effect that a suitable investment product can be recommended according to the investor's investment propensity.

On the other hand, the effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

The following drawings attached to this specification illustrate preferred embodiments of the present invention, and serve to further understand the technical idea of the present invention together with specific details for carrying out the invention, so the present invention is described in such drawings should not be construed as limited to
1 shows a schematic configuration of a distributed investment recommendation system according to big data-based investment propensity analysis results according to an embodiment.
2 is a block diagram of an apparatus for recommending distributed investment according to big data-based investment propensity analysis results according to an embodiment.
3 is a signal flow diagram illustrating a diversified investment method of an investor by a diversified investment recommendation system based on big data-based investment propensity analysis results according to an embodiment.
4 shows a perceptron structure to which the method proposed in this specification can be applied.
5 shows a multilayer perceptron structure to which the method proposed in this specification can be applied.
6 shows an example of a deep neural network to which the method proposed in this specification can be applied.
7 shows an example of a convolutional neural network to which the method proposed in this specification can be applied.
8 illustrates a filter operation in a convolutional neural network to which the method proposed in this specification can be applied.
9 illustrates a neural network structure in which a circular loop to which the method proposed in this specification can be applied exists.
10 shows an operating structure of a recurrent neural network to which the method proposed in this specification can be applied.
11 is an example of a DNN model to which the method proposed in this specification can be applied.
12 is a block diagram of an AI device according to an embodiment of the present invention.

The technology disclosed in this specification can be applied to a diversified investment recommendation system based on big data-based investment propensity analysis results using a wish list and a diversified investment recommendation method using the same. However, the technology disclosed in this specification is not limited thereto, and may be applied to all devices and methods to which the technical spirit of the technology may be applied.

It should be noted that technical terms used in this specification are only used to describe specific embodiments and are not intended to limit the spirit of the technology disclosed in this specification. In addition, technical terms used in this specification should be interpreted in terms commonly understood by those of ordinary skill in the field to which the technology disclosed in this specification belongs, unless specifically defined otherwise in this specification. It should not be interpreted in an overly comprehensive sense or in an excessively reduced sense. In addition, when the technical terms used in this specification are incorrect technical terms that do not accurately express the spirit of the technology disclosed in this specification, it is a technical term that can be correctly understood by those of ordinary skill in the field to which the technology disclosed in this specification belongs. should be replaced with In addition, general terms used in this specification should be interpreted as defined in advance or according to context, and should not be interpreted in an excessively reduced sense.

Terms including ordinal numbers such as first and second used herein may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention.

Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted.

In addition, in describing the technology disclosed in this specification, if it is determined that a detailed description of a related known technology may obscure the gist of the technology disclosed in this specification, the detailed description will be omitted. In addition, it should be noted that the accompanying drawings are only intended to facilitate understanding of the spirit of the technology disclosed in this specification, and should not be construed as limiting the spirit of the technology by the accompanying drawings.

Hereinafter, a schematic configuration of an online investment-linked financial product investment system according to an embodiment of the present invention will be described with reference to FIG. 1 attached.

The meaning of “credit” used in this specification is an economic term that means the ability to pay for the goods transacted in the future. can mean a word that refers to

The online investment-linked financial product investment system disclosed in this specification may refer to a system that relays loans and investments by connecting a loan applicant and an investor who wants to obtain investment returns by investing loan funds for the loan online.

1 shows a schematic configuration of a distributed investment recommendation system according to big data-based investment propensity analysis results according to an embodiment.

Referring to FIG. 1, the distributed investment recommendation system 1000 according to the big data-based investment propensity analysis results disclosed in this specification includes a borrower credit evaluation data providing server 100, and a distributed investment recommendation device according to the big data-based investment propensity analysis results. 200, a loan applicant terminal 300, and an investor terminal 400 may be configured, and the components may exchange data with each other through a wired or wireless communication network.

The loan applicant accesses the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result through the loan applicant terminal 300 and provides personal information necessary for loan application, personal information use agreement, and information on collateral goods, At the same time, the apparatus 200 for recommending distributed investment according to the results of applying for a loan and analyzing the investment propensity based on big data provides the personal information related to the loan of the loan applicant received through the loan applicant terminal 300 and the lender credit evaluation data server 100 Based on the financial credit data of the loan applicant and the collateral data for the collateral received from the loan applicant, creditworthiness of the loan may be evaluated.

As used herein, "collateral data" may refer to data on a collateral object provided as collateral for a loan. For example, if the collateral data is real estate, it means to include all data related to the value of real estate collateral referred to as real estate collateral loan evaluation data, valuation and adjustment of value, in addition to market price information or transaction information of real estate provided as collateral can be interpreted as

The lender credit evaluation data providing server 100 provides the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result for various financial credit data for evaluating the loan credit of the loan applicant and related to collateral for collateral evaluation. As a server for providing data, it may include a financial credit data providing server 110, a collateral data providing server 120, a real estate collateral evaluation data providing server 130, and the like.

Based on the big data-based investment propensity analysis result, the distributed investment recommendation device 200 is based on the personal information provided by the loan applicant, consent to use personal information, and information on collateral, received from the borrower credit evaluation data providing server 100 After collecting the financial credit data and collateral data of the loan applicant, evaluating the loan credit of the loan applicant by processing the collected data with the provided loan credit evaluation model, the result can be transmitted to the loan applicant terminal 300 there is.

In addition, the diversified investment recommendation device 200 according to the results of analyzing the big data-based investment propensity provides investors with product information on various types of online investment-linked financial products designed based on loan products for loan applicants, and the investor terminal It is possible to receive an investment decision order or instructions for online investment-linked financial products from 400 . The diversified investment recommendation device 200 according to the results of the big data-based investment propensity analysis concludes an investment contract according to the investment decision order determined by the investor, receives the investment money, and provides it to loan applicants according to the loan conditions of the predetermined loan product. By doing so, it is possible to invest in online investment-linked financial products consisting of loans and investments.

The loan applicant terminal 300 is a personal terminal such as a smartphone, tablet, PC, etc. used by the loan applicant to apply for a loan, and the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result through a wired and wireless communication network After accessing, necessary data such as the loan application amount entered by the user, personal information, and information on real estate collateral are delivered, and the loan credit evaluation output from the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result You can receive results.

The investor terminal 400 is a personal terminal used by an investor to proceed with an investment, and a terminal identical to or similar to the aforementioned loan applicant terminal 300 may be used. After accessing the diversified investment recommendation device 200 according to the results of big data-based investment propensity analysis through wired and wireless communication networks through these investor terminals 400, data necessary for information on online investment-linked financial products and execution of investment contracts, etc. can send and receive.

The wired/wireless communication network (network) disclosed in this specification includes, for example, a wireless network, a wired network, a public network such as the Internet, a private network, a Global System for Mobile communication network (GSM) network, and a general purpose. General Packet Radio Network (GPRN), Local Area Network (LAN), Wide Area Network (WAN), Metropolitan Area Network (MAN), Cellular Network, Public Switched Telephone Network (Public Switched Telephone Network; PSTN), Personal Area Network, Bluetooth, Wi-Fi Direct, Near Field communication, Ultra-Wide band, these , or any other network, but is not limited thereto.

Hereinafter, the configuration of the apparatus 200 for recommending diversified investment according to the big data-based investment propensity analysis result according to an embodiment will be described in detail with reference to FIG. 2 .

2 is a block diagram of an apparatus for recommending distributed investment according to big data-based investment propensity analysis results according to an embodiment.

Referring to FIG. 2 , the apparatus 200 for recommending diversified investment according to the big data-based investment propensity analysis result may include a communication unit 210, a control unit 220, and a storage unit 230. The illustrated components are not essential, so the diversified investment recommendation device 200 according to the big data-based investment propensity analysis result having more components or fewer components may be implemented, and these components may be implemented in hardware or software, or through a combination of hardware and software.

The communication unit 210 may receive information about real estate to be provided as collateral and a personal information use agreement including personal information necessary for evaluation of loan credit for a loan request applied from the loan applicant terminal 300 of the loan applicant.

The control unit 220 connects to the external borrower credit evaluation data providing server 100 through the communication unit 210 based on the loan applicant's personal information use agreement and information on real estate to be provided as collateral, and then provides the borrower credit evaluation data Financial credit data for the loan applicant and collateral data for the collateral may be collected from the providing server 100 . When the collateral provided by the loan applicant is real estate, the collateral data may include at least one of market price information for real estate collateral, an adjustment factor for real estate collateral, and right relationship information for real estate collateral. As described above, the lender credit evaluation data providing server 100 may include the financial credit data providing server 110, the collateral data providing server 120, the real estate collateral evaluation data providing server 130, and the like, and the control unit 220 ) Collects financial credit data of the loan applicant and collateral data for the provided real estate collateral by using methods such as downloading from the borrower credit evaluation data providing server 100, providing third party information, data scraping, or using the right to request transfer can do. Here, the right to transfer request means the right to transfer the personal information of the loan applicant, and means the right to bring the personal information of the loan applicant managed by one service provider or to request the transfer to another service provider. For example, in order for a loan applicant to receive a real estate mortgage loan from a company that executes a non-face-to-face real estate mortgage loan, a credit card company or bank holding his or her personal information to a company that executes the above-mentioned non-face-to-face real estate mortgage loan personal information may mean the right to request transmission. In addition, the controller 220 may collect data that can be used for evaluating real estate collateral on the network using a data scraping method. The aforementioned data may be collected through an application programming interface (API) provided by the corresponding institution, received as separate data from the corresponding institution, or directly collected from a data providing web page of the corresponding institution.

The financial credit data providing server 110 is a server that provides financial credit data of credit information concentration institutions and credit information companies that provide data for evaluating the credit status of loan applicants in terms of finance, and provides financial credit data. The server 110 has public institutions such as the Health Insurance Corporation and the National Tax Service that can check the four major insurance payment details and tax payment details that can help evaluate/determine the credit status of the loan applicant in terms of finance. Data providing servers of various private financial information providing institutions may be further included.

The collateral data providing server 120 is a server providing market price information, actual transaction price, public price, and evaluation results of the real estate collateral provided by the loan applicant, or a data providing server of an institution providing a value evaluation result for real estate collateral. It may include data providing servers of public institutions and various private appraisal corporations. An example of market price information of real estate collateral provided by the collateral data providing server 120 may be KB apartment market price information provided by KB Kookmin Bank.

The above public institutions and various private appraisal corporations can be replaced with collateral appraisal agencies, etc. to perform the method proposed in this specification, and can be replaced with other institutions, companies, etc. that perform the same or similar functions. .

The real estate collateral evaluation data providing server 130 is a server that provides information/data related to real estate collateral used when evaluating the value of real estate collateral. The information related to real estate collateral may include information such as a rate of change in land price, regional factors, environmental factors, and rights that affect the value of real estate collateral. The control unit 220 evaluates the value of the real estate collateral provided by the loan applicant using the evaluation data collected by the real estate collateral evaluation data providing server 130, and adjusts the evaluated value to determine the accurate value of the real estate collateral can do. Here, the real estate collateral evaluation data providing server 130 is not limited to only one server of any one institution and may include servers of various institutions that provide data used for evaluating the value of real estate collateral.

The real estate collateral evaluation data providing server 130 includes a data providing server for adjustment factors of real estate prices of the Korea Real Estate Agency. It is possible to provide data related to time point correction data to reflect the rate of change in land price, regional factors and individual factors that directly or indirectly affect the evaluation results of real estate collateral. Among the various data reflected to precisely measure the price of real estate, the rate of change in land price is included. The rate of change in land price refers to a numerical value expressing the change in the state of land prices nationwide over a certain period of time as a percentage. Correction of timing means that if there is a change in the price level due to a discrepancy between the transaction time of the transaction case and the reference time of the object to be collateral, the price of the transaction case is modified to the price level of the reference time. Therefore, in calculating the evaluation result of the real estate provided as collateral, it is possible to obtain an accurate evaluation result by reflecting the rate of change in the land price between the transaction time of the case and the reference time of the collateral object.

Regional factors refer to natural, social, economic, and administrative factors that affect the formation of the real estate price level in the region to which the object subject to collateral belongs. For example, the occupational group of residents in the region, the school district in the region, and , sewage, city gas, etc. may be included as regional factors. Individual factors refer to individual factors unique to the subject matter that affect the specific value of the subject matter insured. factors can be included. Regional factors and individual factors can be divided into detailed items such as access conditions, environmental conditions, and restrictions, respectively. Meanwhile, detailed items of these regional factors and individual factors can be classified into different items for each zone, such as commercial zones, residential zones, industrial zones, and agricultural zones, but in this specification, a residential zone will be taken as an example to explain the items. First, the access conditions as regional factors include the distance from the city center, the state of transportation facilities, the layout of shopping malls, and the layout of public and convenience facilities. Environmental conditions as regional factors include weather conditions, natural environment, social environmental conditions, It includes the condition of the lot, the condition of supply and treatment facilities, etc., and restrictions as a regional factor include the proximity of dangerous and hateful facilities, the risk of disasters, the degree of pollution, and the degree of administrative regulation. . Next, access conditions as individual factors include accessibility to transportation facilities, accessibility to shopping malls, and accessibility to public and convenience facilities, and environmental conditions as individual factors include sunlight, natural environment, nearby environment, supply facilities and treatment facilities. It includes the condition of the property, the existence of dangers and hateful facilities, and the restrictions as an individual factor include the area, contact surface, width, depth, shape, direction, height, condition of the contact road, level of administrative regulations, etc. contains These regional and individual factors are factors that affect the value evaluation result of the collateral object. They do not directly affect the collateral evaluation, but accessibility to public and convenience facilities and transportation facilities are indirectly in the collateral evaluation. When registering a loan product, it can be used as content that provides interest in the loan product to investors who provide loan funds and enhances persuasiveness for investment.

In addition, the real estate collateral evaluation data providing server 130 includes a data providing server of an Internet registry office, which is a public institution, a government civil application portal such as Government 24, etc. can provide information. In addition, the relationship of rights to the real estate collateral is the land use plan confirmation document, land ledger, cadastral map, general building ledger, certificate of all building registration items, transfer-in household inspection board, lease contract, certificate of all registered building items, and collective building ledger It may be determined based on at least one document related to the real estate security. Some of the aforementioned documents can be obtained from Internet registries, government civil application portals, and Ministry of Land, Infrastructure and Transport portals, and others can be obtained from servers of public institutions that provide related information. Structure of real estate provided as collateral through these documents, floor space, location, lot number, land category, area, grade, use, owner information, current status of ownership, matters concerning ownership other than ownership, whether or not mortgage/root mortgage and chonsegwon have been established, Since the individual official land price, whether the loan applicant and the actual owner of the real estate provided as collateral match, and the authenticity of the collateral object can be confirmed, these documents can be used to determine the value of the real estate provided as collateral, calculate the amount of priority payment, It can be used to calculate the risk of a collateral item or to calculate a loan limit.

The control unit 220 calculates an evaluation score according to predetermined evaluation criteria for each predetermined configuration item for the collected financial credit data and collateral data, and assigns a weight to a specific configuration item to evaluate the loan credit for the loan applicant's loan application results can be drawn.

The controller 220 provides financial credit data for the loan applicant received from the financial credit data providing server 110, collateral data for real estate collateral received from the collateral data providing server 120, and real estate collateral evaluation data providing server 130 Based on the evaluation data on the real estate collateral received from the loan applicant, a credit evaluation model capable of calculating detailed loan conditions such as loan amount and loan interest rate may be included.

The credit evaluation model may assign preset scores and weights for each component of the financial credit data and the collateral data, and based on the components of the financial credit data and the collateral data, the loan applicant's expected income and the A collateral credit risk for real estate collateral may be predicted, and a maximum loan-to-value ratio (LTV) may be calculated based on the collateral data. The value of real estate collateral can be evaluated by a predetermined method based on the above-mentioned real estate statistical information, time point correction data to reflect the rate of change in land price, regional factors and individual factors for real estate collateral, and information on the relationship of rights to real estate collateral. can Thereafter, the credit evaluation model determines the loan applicant based on a score and weight preset for each component of the financial credit data and the collateral data, the predicted income, the collateral credit risk, and the maximum loanable limit compared to the real estate collateral value. The loan credit evaluation result for can be calculated.

In other words, the control unit 220 assigns an evaluation score for each predetermined evaluation element from the evaluation data collected according to a predetermined evaluation criterion, that is, a credit evaluation model created in advance, and assigns a positive weight to a specific evaluation element, or assigns a positive weight to another specific evaluation element. The credit rating of the loan applicant can be evaluated by assigning negative weights to the evaluation factors.

For example, when a loan applicant using an apartment as collateral provides information on real estate collateral along with an agreement to provide personal information through the loan applicant terminal 300, the control unit 220 transmits credit information through the communication unit 210 Access to the financial credit data providing server 110 provided by centralized institutions and credit information companies, collect financial credit data for loan applicants, access the collateral data providing server 120 of KB Kookmin Bank, and loan applicants By connecting to the market price information of the apartment provided by the Korea Real Estate Agency, the market price information of the apartment provided by the loan applicant as collateral by accessing the collateral data providing server 120 of the Korea Real Estate Agency, and the collateral data providing server 120 of the Ministry of Land, Infrastructure and Transport, the loan applicant Connects to the real transaction price information of the apartment provided as collateral or the collateral data providing server 120 of the National Tax Service, collects information on the standard market price of the apartment provided as collateral by the loan applicant, and provides real estate collateral evaluation data providing server 130 ) to collect information that can identify the rights relationship, such as a certificate of all collective building registration items for the apartment provided as collateral. In addition, the control unit 220 collects the collective building ledger provided by the loan applicant through the government civil application portal and uses it to determine the value of the apartment provided as collateral.

The control unit 220 evaluates the financial credit of the loan applicant based on the collected information or data, predicts future income, and determines the value of the apartment and the risk of the value. The control unit 220 calculates the maximum loanable limit against the real estate collateral value based on the financial credit of the loan applicant, the evaluation result of the apartment value, the future income prediction result, and the risk determination result.

The storage unit 230 may temporarily or permanently store data used for loan credit evaluation, such as data collected for loan credit evaluation, evaluation results, and a credit evaluation model.

Meanwhile, the communication unit 210 may communicate with the investor terminal 400 of the investor, transmit investment product information to the investor terminal 400 according to the investor's request, and transmit contract-related information for an investment contract.

When an investor accesses the investment terminal 400, the control unit 220 may provide investment product-related information, conclude an investment contract according to the investor's investment decision order, and proceed with the investment. The control unit 220 provides information on a pre-designed online investment-linked financial product, loan applicant information linked to the product, collateral information, loan credit evaluation results, etc. to the investor terminal 400, and after confirming the investment product information , Investment can be carried out by concluding an investment contract with an investor who has decided to invest, receiving the investment, and providing the received investment to the loan applicant. Thereafter, the controller 220 receives interest and loan principal from the loan applicant according to predetermined loan conditions and procedures, and pays interest to the investor according to predetermined investment conditions and procedures.

In addition, the storage unit 230 may store data for the conclusion of an investment contract, and the wish list 231 generated in the storage unit 230 stores online investment-linked financial products that investors are interested in. can Investors can access the diversified investment recommendation device 200 according to the results of big data-based investment propensity analysis through the investor terminal 400, search for investment products through the user interface screen, select the investment product you like, and wish It can be stored in the list 231. The contents of the wish list 231 are output on the display of the investor terminal 400, and the investor can determine investment by viewing the investment products in the wish list 231 at a glance and selecting only a specific investment product.

Hereinafter, a diversified investment method of an investor performed by the diversified investment recommendation apparatus 200 according to the big data-based investment propensity analysis result according to an embodiment will be described with reference to FIGS. 2 and 3 .

3 is a signal flow diagram illustrating a diversified investment method of an investor by an online investment-linked financial product investment system according to an embodiment.

The online investment-linked financial product investment system according to an embodiment of the present invention can perform diversified investment by a method including the processes described below.

Referring to FIGS. 2 and 3 , the controller 220 transmits information on various online investment-linked financial products to the investor terminal 400 and guides investors to the online investment-linked financial products (S300). After the investor checks the information on various online investment-linked financial products received from the diversified investment recommendation device 200 according to the big data-based investment propensity analysis result, a command to put the product to invest in the wish list 231 is transmitted ( S310).

The control unit 220 stores the online investment-linked financial product selected by the investor in the wish list 231 of the storage unit 230 (S320). The wish list 231 may be created in the storage unit 230, and an online investment-linked financial product selected by an investor according to an investor's order may be stored in the wish list 231 according to each investor's account.

When an investor checks the investment information of online investment-linked financial products contained in the wish list 231 at the investor terminal 400 and then selects several investment products and makes an investment decision, the controller 220 receives the information from the investor terminal 400. An investor's investment decision command is received (S330).

When the control unit 220 receives an investment decision command from the investor terminal 400, for the conclusion of an investment contract for all specific online investment-linked financial products selected from among several investment products included in the investor's wish list 231, That is, after creating an investment contract session for preparing investment contracts for all online investment-linked financial products selected by the investor from the wish list 231 (S340), the online selected from the wish list 231 through the investment contract session. An investment contract for all investment-linked financial products is collectively concluded (S345). Here, the process of concluding an investment contract for all online investment-linked financial products selected from the wish list 231 in a lump sum makes the entire investment easy without concluding individual investment contracts for a plurality of online investment-linked financial products selected by the investor. It is the process by which a contract can be concluded. Therefore, through such a package investment contract, it is not necessary for an investor to individually conclude each investment contract in the process of diversified investment by selecting a plurality of online investment-linked financial products, which has the effect of facilitating diversified investment for investors.

Next, the controller 220 completes the investment by executing the investment procedure for all online investment-linked financial products for which the investment contract has been concluded (S350).

Next, the control unit 220 collects product information on the online investment-linked financial product for which the investment has been completed (S360), and inputs the collected product information to the investment propensity analysis engine to analyze the investor's investment propensity (S370). This investment propensity analysis may be called behavior-based investment propensity analysis.

Investment products are classified into five types according to their risk levels: ultra-high risk products, high-risk products, medium-risk products, low-risk products, and ultra-low-risk products, and products corresponding to each type are grouped to form an investment product group DB. The control unit 220 classifies the investment propensity based on the number of investments by product group actually invested, the number of rejections of recommended products, and the total amount of investment for each product group based on the log data on the wish list, and the investment by product group actually invested The investment propensity analysis engine is trained using the number of times and total amount of investment for each product group.

The control unit 220 generates a graph of the number of investments for each product group according to the investment risk and the total amount of investment for each product group according to the investment risk as an image, and then uses the image as learning data to construct a CNN included in the investment propensity analysis engine (Convolution Neural Network) based image analysis model is trained. For example, a graph with each product group according to investment risk as the horizontal axis and the number of investments as the vertical axis, and a graph with each product group according to investment risk as the horizontal axis and the total amount of investment as the vertical axis are created as images. , By training the image analysis model with the generated images, the investment propensity analysis engine inputs the two investment results of a specific investor (the number of investments by product group actually invested and the total amount of investment by each product group). Since it is possible to analyze the image and calculate which investment type the analysis result belongs to as a probability value, the average of the probability values for the two investment results determines which product group was invested the most and the investment risk level Investors can be divided into investment types by determining which of the product groups according to which the largest amount was invested.

In addition, the investment propensity analysis engine determines the investment propensity of the investor based on at least one of the investment experience questionnaire, the investment propensity diagnosis questionnaire, and the general personality trait survey results collected from the investor as well as the product information in which the investor has invested as described above. It is possible to determine (S380). This investment propensity analysis may be referred to as a survey-based investment propensity analysis. The controller 220 may recommend various investment products to the investor based on the analysis result of the investor's propensity to invest (S390).

The investment experience questionnaire or investment propensity diagnosis questionnaire is a questionnaire-based investment propensity analysis method that classifies investors’ investment propensity into five types: stable type, stability-seeking type, risk-neutral type, active investment type, and aggressive investment type. ) analyzes the investment propensity of the investor to one of the above five types based on the survey answers received from the investor, and analyzes the investor's investment propensity based on the aforementioned wish list-based actual investment results (S370) Depending on the analysis results and the analysis results of the investment experience questionnaire or the investment propensity diagnosis questionnaire, online investment-linked financial products suitable for the investor's investment propensity, that is, the expected return and risk desired by the investor, can be recommended.

The general personality traits survey is a Big 5 personality factor analysis or a five factor model analysis that classifies an individual's personality into five types: openness, conscientiousness, extroversion, agreeableness, and neuroticism. As a personality propensity analysis method, investors are again classified into three categories: 'openness', 'will to achieve', and 'self-control' according to the investor's personality factors, and the number of investment items of the investor is calculated according to the classification result. The controller 220 may recommend an online investment-linked financial product suitable for the investor's personality propensity according to the analysis result of the investor's personality propensity.

In addition, the control unit 220 calculates the loan applicant's collateral value calculated based on the information provided at the time of loan application collected from the loan applicant by the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result, and the loan applicant's Loan information on loans constituting the online investment-linked financial product, such as reliability of the collateral value evaluation result, credit score of the loan applicant, and annual income of the loan applicant, as well as the loan period for the corresponding loan product, that is, the investment period and the same Based on the delinquency rate and insolvency rate of the borrower for the investment (loan) product, the online investment-linked financial products are classified according to the risk level, and the investment product classified according to the risk level can be recommended according to the result of the investor's investment propensity. That is, the controller 220 may recommend a suitable online investment-linked financial product by matching the risk level of the investment product and the investment propensity according to the investment stability of the investor. For example, when a loan applicant applies for a loan with a house as collateral, factors affecting the risk of housing mortgage loans (LTV, DTI/DSR, home mortgage value, annual income, occupation (categorized into earned income and others)) are variables. is set to calculate the risk level of the investment product.

On the other hand, the control unit 220 calculates the actual investment success rate between the recommended investment products, and then adjusts the weight of the analysis elements constituting the investment propensity analysis engine based on this, so that the recommendation rate of the investment product according to the risk of the investment product. can be adjusted.

For example, the controller 220 uses a method of extracting features of a product using a content-based filtering method as a recommendation method algorithm, finding and recommending a product similar to a product actually invested by an investor using the feature, and using a different recommendation method algorithm. As a method, we use the collaborative filtering method, which is an algorithm that recommends products that investors can invest in using product-investor interaction information. This is a method of recommending products by examining investors' indirect preference using information such as products included in the wish list and products for which investment was canceled. As a disadvantage of the recommendation method algorithm, product recommendation is difficult due to lack of initial data, or product recommendation is made based on the investor's past record, so there is a problem of continuously recommending similar products. To solve this problem, if the initial data is insufficient, the control unit 220 supplements it with an investor type questionnaire obtained from the investor, an investor personality type test, etc., and the problem of continuously recommending similar products is the actual investment in the product recommended by the system to the investor. The ratio value is input into the investment propensity analysis engine to supplement the recommendation method. The control unit 220 compares the ratio of actual investment compared to the product recommended by the investor through the product recommended to the investor through the investment propensity analysis engine and the log result of the product the investor actually invested in and the product log result of the investor canceling the investment. It analyzes the impact of the product recommended by the investment propensity analysis engine on the investor, and adjusts the values of the investor's propensity analysis factors to recommend products that the investor prefers more.

In the foregoing description, steps or actions S310 to S390 may be further divided into additional steps, procedures or actions, or combined into fewer steps, procedures or actions, depending on the implementation of the invention. It can be. In addition, some steps, processes or operations may be omitted if necessary, or the order of steps, processes or operations may be switched. In addition, each step or operation included in the diversified investment method using the online investment-linked financial product investment system described above can be implemented as a computer program and stored in a computer-readable recording medium, and each step and process can be implemented by a computer device. Alternatively, the action may be executed.

Hereinafter, artificial intelligence learning that can be applied to the controller 220 of the present invention will be described with reference to the accompanying drawings 4 to 12.

Artificial Intelligence (hereafter referred to as AI) can use numerous analyzes to determine how complex target tasks are to be performed. In other words, AI can increase efficiency and reduce processing delays. Time-consuming tasks such as analyzing large amounts of data can be performed instantly by using AI.

Hereinafter, machine learning, a type of AI, will be looked at in more detail.

Machine learning refers to a set of actions that train a machine to create a machine that can do tasks that humans can or cannot do. Machine learning requires data and a running model. In machine learning, data learning methods can be largely classified into three types: supervised learning, unsupervised learning, and reinforcement learning.

Neural network training is aimed at minimizing errors in the output. Neural network learning repeatedly inputs training data to the neural network, calculates the output of the neural network for the training data and the error of the target, and backpropagates the error of the neural network from the output layer of the neural network to the input layer in a direction to reduce the error. ) to update the weight of each node in the neural network.

Supervised learning uses training data in which correct answers are labeled in the learning data, and unsupervised learning may not have correct answers labeled in the learning data. That is, for example, learning data in the case of supervised learning related to data classification may be data in which each learning data is labeled with a category. Labeled training data is input to the neural network, and an error may be calculated by comparing the output (category) of the neural network and the label of the training data. The calculated error is back-propagated in a reverse direction (ie, from the output layer to the input layer) in the neural network, and connection weights of each node of each layer of the neural network may be updated according to the back-propagation. The amount of change in the connection weight of each updated node may be determined according to a learning rate. The neural network's computation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently according to the number of iterations of the learning cycle of the neural network. For example, a high learning rate is used in the early stages of neural network learning to increase efficiency by allowing the neural network to quickly achieve a certain level of performance, and a low learning rate can be used in the late stage to increase accuracy.

The learning method may vary depending on the characteristics of the data. For example, in a case where the purpose of the receiver is to accurately predict data transmitted by the transmitter in a communication system, it is preferable to perform learning using supervised learning rather than unsupervised learning or reinforcement learning.

The learning model corresponds to the human brain, and the most basic linear model can be considered. called deep learning).

The neural network cord used as a learning method is largely divided into deep neural networks (DNN), convolutional deep neural networks (CNN), and recurrent neural networks (RNN). there is.

An artificial neural network is an example of connecting several perceptrons.

4 shows a perceptron structure to which the method proposed in this specification can be applied.

Referring to FIG. 4, when the input vector x = (x1, x2, ..., xd) is input, each component is multiplied by the weight (W1, W2, ..., Wd), and after summing up the results, The entire process of applying the activation function σ(·) is called a perceptron. The huge artificial neural network structure may extend the simplified perceptron structure shown in FIG. 1 and apply input vectors to different multi-dimensional perceptrons. For convenience of description, an input value or an output value is referred to as a node.

Meanwhile, the perceptron structure shown in FIG. 4 can be described as being composed of a total of three layers based on input values and output values. An artificial neural network in which H number of (d + 1) dimensional perceptrons exist between the 1st layer and the 2nd layer and K number of (H + 1) dimensional perceptrons between the 2nd layer and the 3rd layer can be expressed as shown in FIG. 5 .

5 shows the structure of a multi-layer perceptron to which the method proposed in this specification can be applied.

The layer where the input vector is located is called the input layer, the layer where the final output value is located is called the output layer, and all the layers located between the input layer and the output layer are called hidden layers. In the example of FIG. 5 , three layers are disclosed, but when counting the number of layers of an actual artificial neural network, the number of layers is counted excluding the input layer, so a total of two layers can be considered. The artificial neural network is composed of two-dimensionally connected perceptrons of basic blocks.

The above-described input layer, hidden layer, and output layer can be jointly applied to various artificial neural network structures such as CNN and RNN, which will be described later, as well as multi-layer perceptrons. As the number of hidden layers increases, the artificial neural network becomes deeper, and a machine learning paradigm that uses a sufficiently deep artificial neural network as a learning model is called deep learning. In addition, the artificial neural network used for deep learning is called a deep neural network (DNN).

6 shows an example of a deep neural network to which the method proposed in this specification can be applied.

The deep neural network shown in FIG. 6 is a multi-layer perceptron consisting of 8 hidden layers + 8 output layers. The multilayer perceptron structure is expressed as a fully-connected neural network. In a fully-connected neural network, there is no connection relationship between nodes located on the same layer, and a connection relationship exists only between nodes located on adjacent layers. DNN has a fully-connected neural network structure and is composed of a combination of multiple hidden layers and activation functions, so it can be usefully applied to identify the correlation characteristics between inputs and outputs. Here, the correlation characteristic may mean a joint probability of input and output.

On the other hand, depending on how a plurality of perceptrons are connected to each other, various artificial neural network structures different from the aforementioned DNN can be formed.

7 shows an example of a convolutional neural network to which the method proposed in this specification can be applied, and FIG. 8 shows a filter operation in the convolutional neural network to which the method proposed in this specification can be applied.

In DNN, nodes located inside one layer are arranged in a one-dimensional vertical direction. However, in FIG. 7, it can be assumed that the nodes are two-dimensionally arranged with w nodes horizontally and h nodes vertically (convolutional neural network structure of FIG. 7). In this case, since a weight is added for each connection in the connection process from one input node to the hidden layer, a total of h×w weights must be considered. Since there are h×w nodes in the input layer, a total of h2w2 weights are required between two adjacent layers.

The convolutional neural network of FIG. 7 has a problem in that the number of weights increases exponentially according to the number of connections, so instead of considering all mode connections between adjacent layers, it is assumed that there is a filter with a small size, and FIG. 8 As shown in , weighted sum and activation function calculations are performed for overlapping filters.

One filter has weights corresponding to the number of filters, and learning of weights can be performed so that a specific feature on an image can be extracted as a factor and output. In FIG. 8 , a 3×3 filter is applied to the top left 3×3 region of the input layer, and the weighted sum and activation function calculations are performed on the corresponding node, and the resulting output value is stored in z22.

While scanning the input layer, the filter moves by a certain distance horizontally and vertically, performs weighted sum and activation function calculations, and places the output value at the position of the current filter. This operation method is similar to the convolution operation for images in the field of computer vision, so the deep neural network of this structure is called a convolutional neural network (CNN), and the hidden layer generated as a result of the convolution operation is called a convolutional layer. In addition, a neural network having a plurality of convolutional layers is referred to as a deep convolutional neural network (DCNN).

In the convolution layer, the number of weights can be reduced by calculating a weighted sum by including only nodes located in a region covered by the filter from the node where the current filter is located. This allows one filter to be used to focus on features for a local area. Accordingly, CNN can be effectively applied to image data processing in which a physical distance in a 2D area is an important criterion. Meanwhile, in the CNN, a plurality of filters may be applied immediately before the convolution layer, and a plurality of output results may be generated through a convolution operation of each filter.

Meanwhile, there may be data whose sequence characteristics are important according to data attributes. Considering the length variability and precedence relationship of these sequence data, input each element on the data sequence one by one at each time step, and input the output vector (hidden vector) of the hidden layer output at a specific time point together with the next element on the sequence A structure in which this method is applied to an artificial neural network is called a recurrent neural network structure.

9 illustrates a neural network structure in which a circular loop to which the method proposed in this specification can be applied exists.

Referring to FIG. 9, a recurrent neural network (RNN) inputs an element (x1(t), x2(t), ..., xd(t)) of any gaze t on a data sequence into a fully connected neural network In the process, the immediately preceding time point t-1 applies the weighted sum and activation function by inputting the hidden vector (z1 (t-1), z2 (t-1), ... , zH (t-1)) It is a structure that The reason why the hidden vector is transmitted to the next time point in this way is that information in the input vector at previous time points is regarded as being accumulated in the hidden vector of the current time point.

10 shows an operating structure of a recurrent neural network to which the method proposed in this specification can be applied. Referring to FIG. 10 , the recurrent neural network operates in a predetermined sequence of views with respect to an input data sequence.

The hidden vector (z1(1), z2(1), ... when the input vector (x1(t), x2(t), ... , xd(t)) at time point 1 is input to the recurrent neural network , zH(1)) is input together with the input vector (x1(2), x2(2), ... , xd(2)) at time point 2, and the vector of the hidden layer (z1(2) ), z2(2), ... , zH(2)). This process is repeated until time point 2, time point 3, ..., time point T.

Meanwhile, when a plurality of hidden layers are arranged in a recurrent neural network, it is referred to as a deep recurrent neural network (DRNN). Recurrent neural networks are designed to be usefully applied to sequence data (eg, natural language processing).

As a neural network core used as a learning method, in addition to DNN, CNN, and RNN, Restricted Boltzmann Machine (RBM), deep belief networks (DBN), and Deep Q-Network It includes various deep learning techniques such as computer vision, voice recognition, natural language processing, voice/signal processing, and derivation of optimal values through large-capacity data processing.

11 is an example of a DNN model to which the method proposed in this specification can be applied.

A deep neural network (DNN) is an artificial neural network (ANN) composed of several hidden layers between an input layer and an output layer. Deep neural networks can model complex non-linear relationships, just like regular artificial neural networks.

For example, in a deep neural network structure for an object identification model, each object may be represented as a hierarchical composition of basic image elements. In this case, the additional layers may consolidate the characteristics of the gradually gathered lower layers. This feature of deep neural networks allows complex data to be modeled with fewer units (units, nodes) compared to similarly performed artificial neural networks.

As the number of hidden layers increases, the artificial neural network is called “deep.” The machine learning paradigm that uses a sufficiently deep artificial neural network as a learning model is called deep learning. In addition, a sufficiently deep artificial neural network used for such deep learning is collectively referred to as a deep neural network (DNN).

In the present invention, data required to train the POI data generation model can be input to the input layer of the DNN, and meaningful data usable by the user can be generated through the output layer while passing through the hidden layers.

Although the artificial neural network used for this deep learning method is collectively referred to as DNN in the specification of the present invention, other deep learning methods can be applied as long as meaningful data can be output in a similar way.

12 is a block diagram of an AI device according to an embodiment of the present invention.

The AI device 20 may include an electronic device including an AI module capable of performing AI processing or a server including the AI module. In addition, the AI device 20 is included as at least a part of the configuration of the control unit 220 of the distributed investment recommendation device 200 according to the big data-based investment propensity analysis result shown in FIG. 2, CNN learning, deep learning, image It may also be provided to perform at least a part of AI processing such as recognition. The AI processing of the AI device 20 will include all operations related to the control of the distributed investment recommendation device 200 according to the big data-based investment propensity analysis results shown in FIG. 2 and all operations through artificial intelligence learning. can

The AI device 20 may include an AI processor 21, a memory 25 and/or a communication unit 27.

The AI device 20 is a computing device capable of learning a neural network, and may be implemented in various electronic devices such as servers, desktop PCs, notebook PCs, and tablet PCs, or implemented as a single chip. In the present invention, the AI device 20 may be a device for recommending distributed investment according to a big data-based investment propensity analysis result implemented in any one of the various electronic devices.

The AI processor 21 may learn a neural network using a program stored in the memory 25 . In particular, the AI processor 21 may learn a neural network for recognizing device-related data. Here, the neural network for recognizing device-related data may be designed to simulate the structure of a human brain on a computer, and may include a plurality of network nodes having weights that simulate neurons of the human neural network. A plurality of network modes may transmit and receive data according to a respective connection relationship so as to simulate synaptic activity of neurons that transmit and receive signals through synapses. Here, the neural network may include a deep learning model developed from a neural network model. In the deep learning model, a plurality of network nodes may exchange data according to a convolution connection relationship while being located in different layers. Examples of neural network models include deep neural networks (DNN), convolutional deep neural networks (CNN), recurrent neural networks (RNN), Restricted Boltzmann Machine (RBM), deep trust It includes various deep learning techniques such as deep belief networks (DBN) and deep Q-network, and can be applied to fields such as computer vision (CV), voice recognition, natural language processing, and voice/signal processing. can

Meanwhile, the processor performing the functions described above may be a general-purpose processor (eg, CPU), or may be an AI-only processor (eg, GPU) for artificial intelligence learning.

The memory 25 may store various programs and data necessary for the operation of the AI device 20 . The memory 25 may be implemented as a non-volatile memory, a volatile memory, a flash-memory, a hard disk drive (HDD), or a solid state drive (SDD). The memory 25 is accessed by the AI processor 21, and reading/writing/modifying/deleting/updating of data by the AI processor 21 can be performed. In addition, the memory 25 may store a neural network model (eg, the deep learning model 26) generated through a learning algorithm for data classification/recognition according to an embodiment of the present invention.

Meanwhile, the AI processor 21 may include a data learning unit 22 that learns a neural network for data classification/recognition. The data learning unit 22 may learn criteria regarding which training data to use to determine data classification/recognition and how to classify and recognize data using the training data. The data learning unit 22 may acquire learning data to be used for learning and learn the deep learning model by applying the obtained learning data to the deep learning model.

The data learning unit 22 may be manufactured in the form of at least one hardware chip and mounted on the AI device 20 . For example, the data learning unit 22 may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or manufactured as a part of a general-purpose processor (CPU) or a graphics-only processor (GPU) for the AI device 20. may be mounted. Also, the data learning unit 22 may be implemented as a software module. When implemented as a software module (or a program module including instructions), the software module may be stored in a computer-readable, non-transitory computer readable recording medium (non-transitory computer readable media). In this case, at least one software module may be provided by an operating system (OS) or an application (application program).

The data learning unit 22 may include a training data acquisition unit 23 and a model learning unit 24 .

The training data acquisition unit 23 may acquire training data required for a neural network model for classifying and recognizing data. For example, the learning data acquisition unit 23 may acquire image data and/or sample data for transmitters on a road infrastructure to be input to a neural network model as learning data.

The model learning unit 24 may learn to have a criterion for determining how to classify predetermined data by using the obtained training data. In this case, the model learning unit 24 may learn the neural network model through supervised learning using at least some of the learning data as a criterion. Alternatively, the model learning unit 24 may learn the neural network model through unsupervised learning in which a decision criterion is discovered by self-learning using learning data without guidance. In addition, the model learning unit 24 may learn the neural network model through reinforcement learning using feedback about whether the result of the situation judgment according to learning is correct. In addition, the model learning unit 24 may train the neural network model using a learning algorithm including error back-propagation or gradient decent.

When the neural network model is learned, the model learning unit 24 may store the learned neural network model in memory. The model learning unit 24 may store the learned neural network model in a memory of a server connected to the AI device 20 through a wired or wireless network.

The data learning unit 22 further includes a training data pre-processing unit (not shown) and a learning data selection unit (not shown) to improve the analysis result of the recognition model or save resources or time required for generating the recognition model. You may.

The learning data pre-processing unit may pre-process the acquired data so that the acquired data can be used for learning for situation determination. For example, the learning data pre-processing unit may process the acquired data into a preset format so that the model learning unit 24 can use the acquired learning data for learning for image data recognition of the transmission unit.

In addition, the learning data selector may select data necessary for learning from among the learning data acquired by the learning data acquisition unit 23 or the learning data preprocessed by the preprocessor. The selected learning data may be provided to the model learning unit 24 . For example, the learning data selection unit may select only data included in a specific field as training data by recognizing a specific field among data sets collected through a network.

In addition, the data learning unit 22 may further include a model evaluation unit (not shown) to improve the analysis result of the neural network model.

The model evaluation unit inputs evaluation data to the neural network model, and when an analysis result output from the evaluation data does not satisfy a predetermined criterion, it may cause the model learning unit 22 to learn again. In this case, the evaluation data may be predefined data for evaluating the recognition model. For example, the model evaluator may evaluate that the predetermined criterion is not satisfied when the number or ratio of evaluation data for which the analysis result is not accurate among the analysis results of the learned recognition model for the evaluation data exceeds a preset threshold. .

The communication unit 27 may transmit the AI processing result by the AI processor 21 to an external electronic device.

On the other hand, the AI device 20 shown in FIG. 12 has been functionally divided into an AI processor 21, a memory 25, a communication unit 27, etc., but the above-described components are integrated into one module and the AI module Note that it can also be called

As used herein, the term “unit” (eg, a controller) may refer to a unit including one or a combination of two or more of, for example, hardware, software, or firmware. “Unit” may be used interchangeably with terms such as, for example, unit, logic, logical block, component, or circuit. A "unit" may be a minimum unit of an integrally constituted part or a part thereof. A “unit” may be a minimal unit or part thereof that performs one or more functions. A “unit” may be implemented mechanically or electronically. For example, a "part" is an application-specific integrated circuit (ASIC) chip, field-programmable gate arrays (FPGAs), or programmable-logic device that performs certain operations, known or developed in the future. may contain at least one.

At least some of the devices (eg, modules or functions thereof) or methods (eg, operations) according to various embodiments may be stored on computer-readable storage media in the form of, for example, program modules. It can be implemented as a command stored in . When the command is executed by a processor, the one or more processors may perform a function corresponding to the command. The computer-readable storage medium may be, for example, a memory.

Computer-readable storage media / computer-readable recording media include hard disks, floppy disks, magnetic media (eg magnetic tape), optical media (eg CD-ROM (compact) disc read only memory), digital versatile disc (DVD), magneto-optical media (e.g. floptical disk), hardware devices (e.g. read only memory (ROM), random access memory), or flash memory, etc. In addition, program commands may include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to act as one or more software modules to perform the operations of various embodiments, and vice versa.

A module or program module according to various embodiments may include at least one or more of the aforementioned elements, some may be omitted, or additional elements may be further included. Operations performed by modules, program modules, or other components according to various embodiments may be executed in a sequential, parallel, repetitive, or heuristic manner. Also, some actions may be performed in a different order, omitted, or other actions may be added.

As used herein, the term "a" is defined as one or more than one. Also, the use of introductory phrases such as “at least one” and “one or more” in a claim means that the same claim includes introductory phrases such as “at least one” and “one or more” and ambiguous phrases such as “an”. If any, be construed to mean that the introduction of another claim element by the ambiguous phrase "an" limits any particular claim containing the so-introduced claim element to an invention containing only one such element. It shouldn't be.

Unless otherwise specified, terms such as "first" and "second" are used to arbitrarily distinguish the elements they describe. Thus, these terms are not necessarily intended to indicate a temporal or other order of priority of such elements, and the mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to advantage. . Accordingly, these terms are not necessarily intended to indicate a temporal or other priority of such elements. The mere fact that certain measures are cited in different claims does not indicate that a combination of these measures cannot be useful.

Arrangements of components to achieve the same function are effectively "related" so that the desired function is achieved. Thus, any two components that are combined to achieve a particular functionality may be considered "related" to each other such that the desired function is achieved, regardless of structure or intervening components. Similarly, two components so associated can be considered "operably connected" or "operably coupled" to each other to achieve a desired function.

Further, those skilled in the art will recognize that the boundaries between the functionality of the foregoing operations are exemplary only. A plurality of actions may be combined into a single action, a single action may be distributed into additional actions, and the actions may be executed at least partially overlapping in time. Also, alternative embodiments may include multiple instances of a particular operation, and the order of operations may be changed in various other embodiments. However, other modifications, variations and alternatives are also possible. Accordingly, the detailed description and drawings are to be regarded in an illustrative rather than a limiting sense.

The phrase “may be X” indicates that condition X can be met. This phrase also indicates that condition X may not be met. For example, a reference to a system that contains a specific component should also include a scenario where the system does not contain that specific component. For example, a reference to a method that includes a specific action must also include scenarios in which the method does not include that specific component. But to take another example, a reference to a system configured to perform a specific action should also include a scenario in which the system is not configured to perform the specific action.

The terms "comprising", "having", "consisting of", "consisting of" and "consisting essentially of" are used interchangeably. For example, any method may include at least the operations included in the drawings and/or specifications, and may include only the operations included in the drawings and/or specifications. Alternatively, the word "comprising" does not exclude the presence of the recited elements or acts in a claim.

It should be understood that the boundaries between logical blocks are merely illustrative, and that alternative embodiments may merge logical blocks or circuit elements or impose alternative decompositions of functionality on various logical blocks or circuit elements. will recognize that there is Accordingly, it should be understood that the architecture shown herein is merely illustrative, and in fact many other architectures that achieve the same functionality may be implemented.

Also, for example, in one embodiment, the illustrated examples may be implemented on a single integrated circuit or as circuitry located within the same device. Alternatively, the above examples may be implemented as any number of discrete integrated circuits or discrete devices interconnected with each other in any suitable manner, and other alterations, modifications, variations and alternatives are also possible. Accordingly, the specification and drawings are to be regarded in an illustrative rather than restrictive sense.

Also, for example, any of the above examples or portions thereof may be implemented as a software or code representation of a physical circuit or a logical representation translatable to a physical circuit, such as any suitable type of hardware description language.

In addition, the present invention is not limited to physical devices or units implemented with non-programmable hardware, but generally includes mainframes, mini computers, servers, workstations, personal computers, notepads, referred to herein as 'computer systems'. Programmable devices capable of performing desired device functions by operating in accordance with appropriate program code, such as personal digital assistants (PDAs), electronic games, automobiles and other embedded systems, mobile phones and various other wireless devices. Or it can be applied to units as well.

A system, apparatus or device referred to in this specification includes at least one hardware component.

Connections as described herein may be any type of connection suitable for transmitting a signal from or to a respective node, unit or device via an intermediate device, for example. Thus, unless implicitly or otherwise stated, a connection may be, for example, a direct connection or an indirect connection. A connection may be described or described with reference to a single connection, multiple connections, unidirectional connections, or bidirectional connections. However, different embodiments may change the implementation of connectivity. For example, separate unidirectional connections can be used instead of bidirectional connections, and vice versa. Also, multiple connections may be replaced with a single connection that transmits a plurality of signals sequentially or in a time multiplexed manner. Likewise, a single connection carrying multiple signals can be broken into multiple connections carrying subsets of these signals. Therefore, many options exist for transmitting signals.

In the above, a preferred embodiment of the technology of this specification has been described with reference to the accompanying drawings. Here, terms or words used in this specification and claims should not be construed as being limited to ordinary or dictionary meanings, but should be interpreted as meanings and concepts consistent with the technical spirit of the present invention. The scope of the present invention is not limited to the embodiments disclosed herein, and the present invention can be modified, changed, or improved in various forms within the spirit of the present invention and the scope described in the claims.

100: Borrower credit evaluation data providing server
110: financial credit data providing server
120: collateral data providing server
130: real estate collateral evaluation data providing server
200: Distributed investment recommendation device based on big data-based investment propensity analysis results
210: communication department
220: control unit
230: storage unit
231: wish list
300: Loan applicant terminal
400: investor terminal
1000: Online investment-linked financial product investment system

Claims (10)

In the diversified investment method using the diversified investment recommendation system according to the results of big data-based investment propensity analysis,
Adding an online investment-linked financial product selected by a single investor to the wish list;
When the single investor decides to invest in a plurality of online investment-linked financial products selected from among online investment-linked financial products included in the wish list, receiving an investment decision command from the single investor from an investor terminal;
generating an investment contract session for concluding an investment contract for all of the selected plurality of online investment-linked financial products;
concluding an investment contract for all of the selected plurality of online investment-linked financial products collectively through the investment contract session;
Completing an investment by performing an investment procedure for all online investment-linked financial products for which the investment contract has been concluded;
Collecting product information on a plurality of online investment-linked financial products for which the investment is completed;
Determining the investment propensity of the single investor based on the investment propensity analysis result of the single investor for the collected product information through the investment propensity analysis engine and the investment experience questionnaire or investment propensity diagnosis questionnaire result collected from the single investor , The investment propensity analysis engine is trained to classify the investment propensity of investors based on the number of investments by product group actually invested, the number of rejections of recommended products, and the total amount of investment by product group based on the log data of each investor's wish list. Learning data consisting of an image for a graph with the classification of the product group as the horizontal axis and the number of investments for each product group as the vertical axis and an image for the graph with the classification of the product group as the horizontal axis and the total investment amount for each product group as the vertical axis. Including the learned CNN (Convolution Neural Network)-based image analysis model, when the number of investments by product group and the total amount of investment by product group actually invested by a specific investor are entered, the image of the investment result is analyzed, and the analysis result is After calculating which investment type belongs to a probability value, the average of the probability values is used to determine the product group in which the specific investor has invested the most and the largest amount among the product groups according to the investment risk, and based on the determination result Based on this, the investment propensity analysis result of the specific investor is derived, and the online investment-linked financial products are classified into ultra-high risk products, high-risk products, medium-risk products, low-risk products, and ultra-low-risk products according to the investment risk, and each of the classified products Online investment-linked financial products corresponding to the type of investment risk are grouped to form a product group; and
Based on the investment propensity determination result, the investment product is recommended to the single investor, but the loan applicant's collateral value, the reliability of the loan applicant's collateral value evaluation result, the loan applicant's credit score, the loan applicant's annual income, the investment period and the existing Classifying investment products according to investment risk based on the delinquency rate and insolvency rate of the same investment product, and recommending an investment product whose investment risk level of the investment product matches the investment propensity determination result. delete delete delete According to claim 1,
Calculating an investment success rate for the recommended investment product through a product log in which the single investor actually invested in the recommended investment product and a product log result value in which the single investor did not invest; and
The method further comprising adjusting a recommendation ratio of an investment product according to an investment risk by adjusting a weight of an analysis element of the investment propensity analysis engine based on the investment success rate. communications department;
storage unit; and
Including a control unit functionally connected to the communication unit and the storage unit, wherein the control unit,
When a single investor places an online investment-linked financial product in the wish list created in the storage unit and then issues an investment decision order for a plurality of selected online investment-linked financial products, the investment decision command is received from the terminal of the single investor ,
Creating an investment contract session for concluding an investment contract for all of the selected plurality of online investment-linked financial products among the online investment-linked financial products stored in the wish list;
Through the investment contract session, an investment contract for all of the selected plurality of online investment-linked financial products is collectively concluded,
Complete the investment by performing the investment procedure for all online investment-linked financial products for which the above investment contract has been concluded;
Collect product information on a plurality of online investment-linked financial products for which the investment has been completed, and analyze the investment propensity of the single investor on the collected product information through the investment propensity analysis engine and the investment experience questionnaire collected from the single investor. Alternatively, the investment propensity of the single investor is determined based on the results of the investment propensity diagnosis survey, but the investment propensity analysis engine determines the number of investments by product group actually invested based on the log data for each investor's wish list, the number of times of rejection of recommended products, and It is learned to classify the investment propensity of investors based on the total investment amount for each product group, but the horizontal axis is the classification of the product group and the number of investments by product group is the vertical axis. The number of investments in each product group and the investment amount for each product group that a specific investor actually invested in, including a CNN (Convolution Neural Network)-based image analysis model learned with training data consisting of images of a graph with the total investment amount for each product group as the vertical axis When the total amount data is input, the image of the investment result is analyzed, and after calculating the probability value to which investment type the analysis result belongs to, the specific investor invests the most among the product groups according to the investment risk with the average of the probability value A product group and the product group in which the largest amount is invested are judged, and based on the judgment result, the investment propensity analysis result of the specific investor is derived. It is classified into medium-risk products, low-risk products, and ultra-low-risk products, and online investment-linked financial products corresponding to each classified investment risk type are grouped to form a product group, and
Based on the investment propensity determination result, the investment product is recommended to the single investor, but the loan applicant's collateral value, the reliability of the loan applicant's collateral value evaluation result, the loan applicant's credit score, the loan applicant's annual income, the investment period and the existing Based on the delinquency rate and insolvency rate of the same investment product, the investment product is classified according to the investment risk level, and the investment product that matches the investment risk level of the investment product with the investment propensity judgment result is recommended.
A device for recommending diversified investment according to the results of big data-based investment propensity analysis. delete delete delete The method of claim 6, wherein the control unit,
In the recommended investment product, the investment completion rate for the recommended investment product is calculated through the product log that the single investor actually invested in and the product log result value that the single investor did not invest in, and
Based on the investment success rate, adjusting the weight of the analysis element of the investment propensity analysis engine to adjust the recommendation rate of investment products according to the investment risk
Distributed investment recommendation device according to the results of big data-based investment propensity analysis, characterized in that.

Images