KR20190081658A - A System Providing Automatic Payment Service After Repayment Based on Credit Evaluation Using Multi-Media Big Data Matching Pattern Analysis - Google Patents

Abstract

The present invention relates to a credit assessment based pre-distribution reservation post-redemption automatic payment service providing system using multimedia big data matching pattern analysis. The system includes: a user information input module receiving user information from a user and authenticating the user information; a credit assessment interface interlocking with a payment gateway (PG) server and an affiliated company server to receive external information including multimedia payment information, which includes a payment history of any one of a card company and a telecommunication company, and affiliated company information, which includes a deposit history of the affiliated company, and evaluating credit of the user based on the external information; a redemption condition setting module setting a payment method, with which the amount of goods is paid when pre-distributing goods to the user, and a redemption period and an automatic payment day of the amount; a pre-distribution managing module pre-distributing eligible goods to the user; and a redemption managing module recognizing whether the amount is redeemed from the user and performing automatic payment to have the amount redeemed. According to the present invention, it is possible to analyze exact and effective purchase patterns by performing a survey to identify purchase patterns of users.

Description

[0001] The present invention relates to a system for providing an automatic payment service after a prepayment based on credit evaluation using a multi-media big data matching pattern analysis,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a system for providing a prepayment automatic payment service after prepayment based on a credit evaluation using a multi-media big data matching pattern analysis. More specifically, The system is a system that prepares the goods based on the evaluation of the goods and selects and reserves the specific repayment date and time and then makes an automatic payment. It is a system that not only receives the goods without being affected by the economic environment, The present invention relates to a system for providing a prepayment automatic payment service based on credit evaluation using a multi-media pattern analysis.

Generally, people pay the amount and receive the goods. This method is a socially accepted method, and it is the most convenient and reliable method since the goods provider can receive the stable amount of money and the recipient can receive the goods with stability. However, for those who are not economically comfortable, for example, housewives, students, and job seekers, there is a limit to the types of products that can be paid according to the amount. In addition, a system has been developed in which, for convenience of payment, a settlement means or payment information is registered in advance and a payment for a commodity is automatically paid when a commodity is paid.

In this regard, Korean Patent Registration No. 10-1791743 (entitled " Open Simple Payment System ") proposes a system in which an automatic payment is made using previously registered payment information.

According to the present invention, the PIN of the merchant is input from the merchant to the online merchandise purchased through the user's mobile terminal, and the PIN is registered, The user is automatically settled to use or admit the online merchandise, and the transaction details are pushed in real time to the merchant terminal.

In addition, Korean Patent Registration No. 10-1682888 entitled " Postpaid Claim Method Using Mobile Prepaid Card, Prepaid Card Server Server and Postpaid Card Company Server < / RTI > A method of payment using a mobile prepaid card for postpaid use of a used service or a used amount, a prepaid card company server and a post card company server and system for supporting the payment method.

The payment system for performing payment requested by the mobile prepaid card verifies validity of the payment card from the prepaid card company server, the prepaid card information received from the prepaid card company server, the payment card, and the unique information of the mobile terminal, When the approval of the first approval limit is received from the card issuer server, the amount of use of the payment card is deducted by the amount corresponding to the requested initial approval limit, and the first approval limit requested by the prepaid card company server is approved. To the prepaid card company server, to the prepaid card company server.

However, the above two inventions have disadvantages in that the reliability of the above-described invention can not be secured because the user's credit is not evaluated. In order to overcome these drawbacks, we analyze the big data of the patterns and histories of the users in the multimedia (card companies, telecom companies, online shopping malls, content service companies, etc.) There is a need for development of an ongoing system.

The present invention has been devised in order to overcome the problems of the above-described technology, and it is an object of the present invention to provide a service providing system and a method for providing a payment service, History and living standard) history of the user, and then based on this, the main goal is to automatically settle the goods to the user after prepayment.

Another object of the present invention is to conduct a survey to analyze a user's purchase pattern in order to evaluate a user's credit. Surveys can provide quality of goods and services, push marketing, and matching services by securing various pattern and user tendencies.

It is still another object of the present invention to provide a system and a method for managing a user's ability to repay a user's goods before a user prepares the goods based on the user's credit rating, do.

It is a further object of the present invention to allow the user to selectively display the goods by grasping the tendency of the user to select the goods through investigation of the latter after the user receives the goods.

In order to achieve the above object, there is provided a system for providing a prepayment automatic payment service after a prepayment reservation based on a credit evaluation using a multi-media big data matching pattern analysis according to the present invention. The system includes a user who receives user information from a user, An information input module; A payment gateway (PG) server and affiliate server, receives external information including multi-media settlement information including payment details of a card company and a communication company, and affiliate information including a payment history of an affiliate, A credit evaluation interface for evaluating credit of the user based on the credit rating; A redemption condition setting module for setting a redemption period and an automatic redemption date of the money, a payment means for receiving an amount of the goods at the time of pre-payment of the goods to the user; An advance payment management module for prepaying the goods to be prepaid to the user; And a redemption management module for recognizing whether or not the amount of money is redeemed from the user and executing automatic settlement so as to receive the redemption amount.

The credit evaluation interface may further include: a multi-media analyzing unit for extracting user settlement information, which is payment details of the user among the user information, from the multi-media settlement information and deriving a multi-media analysis result by analyzing the payment pattern of the user; And a payment pattern information generation module for generating payment pattern information based on the result of the determination; a purchase pattern checking module for generating a purchase pattern information by analyzing a purchase pattern of a user by providing a survey to the user; A credit rating calculation module for calculating a credit rating obtained by digitizing the credit of the user based on the settlement pattern information and the purchase pattern information; and a suitability determination module for determining a prepayment qualification of the goods of the user based on the credit rating .

Further, the payment pattern analysis module extracts additional payment information, which is the payment history of the user, of the user information from the affiliate information, analyzes the payment pattern of the user by integrating the additional payment information and the payment pattern information, Wherein the settlement pattern information generation unit generates the integrated settlement pattern information by applying the additional pattern information to the settlement pattern information.

In addition, the purchasing pattern inquiry module may include a questionnaire providing unit for generating a questionnaire for inquiring purchase intention and purchase habits of the user and providing the questionnaire to the user, an answer input unit receiving the answer sentence of the questionnaire from the user, A term grouping unit for deriving a valid term through a grouping factor obtained by grouping the terms in accordance with the frequency of appearance of the extracted terms; And a purchase pattern information generation unit for generating the purchase pattern information by grasping the purchase pattern of the user based on the representative term.

A system for providing a prepayment automatic payment service after prepayment based on a credit rating using analysis of a large-volume big data matching pattern according to the present invention,

1) Before the goods are dispensed, the reliability of the system of the present invention can be secured by evaluating the user's credit through the payment pattern and the purchase pattern of the user,

2) It is possible to conduct an accurate and efficient purchase pattern analysis by conducting surveys that can identify users' purchase patterns,

3) Providing the effect of further improving the stability of the system of the present invention by managing the repayment of the amount of money after goods are paid to the user,

4) By examining the user's post-payment for the pre-paid goods, the user can identify the tendency to select the goods and present the effect of selectively displaying the goods to the user based on the tendency.

1 is a conceptual diagram showing a schematic configuration of a system of the present invention;
2 is a block diagram showing a detailed configuration of a system of the present invention;
Figure 3 is a flow chart illustrating the process of the system of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The accompanying drawings are not drawn to scale and wherein like reference numerals in the various drawings refer to like elements.

Fig. 1 is a conceptual diagram showing a schematic configuration of the system 1 of the present invention.

Referring to FIG. 1, a credit reimbursement automatic payment settlement service providing system based on a credit evaluation based on the analysis of a large-volume big data matching pattern according to the present invention is provided for securing the stability of the system 1, (Payment gateway, electronic payment service) server and a server of an affiliate affiliated with the system 1 of the present invention to grasp the payment tendency and purchase tendency of users using various media, that is, multi-media big data matching, And evaluates the creditworthiness of the user based on the obtained credit rating.

Further, the system (1) of the present invention determines whether or not the user can use the system (1) based on the credit of the user evaluated through the credit evaluation model, prepares the goods to the user, Function. Here, the system (1) of the present invention is a system (1) in which a user receives a prepayment when there is insufficient payment amount or purchase cost when purchasing or using virtual money, service, commodity and cyber money in online and offline media, (1) is a system (1) in which a payment is made after a reservation is made using a means that an individual can settle in order to prevent a case where the person can not be repaid, and the payment is automatically made when the payment is exceeded.

Here, the multi-media big data matching method is a method of matching and analyzing the common points, differences, and problems of the big data patterns including purchase, payment, and service use in various media such as card companies, communication companies, shopping malls, and offline shops. The Big Data is collectively referred to as various data stored in the PG server and the affiliate server, which will be described later, for managing various data in unison. Thus, in the system of the present invention, the user's credit is evaluated based on the multi-media big data matching method.

To this end, the system (1) of the present invention comprises a user who wants to receive goods in advance, a user terminal used by the user, a PG server in which data for evaluating the user's credit is stored, It is made up of stored affiliate servers.

A user is a user who wants to receive goods, and can also be referred to as a user terminal that can select goods to be pre-paid. Here, the term 'goods' refers to at least one of goods, services, virtual money, and cyber money. At this time, the user terminal does not limit the types of terminals such as a smart phone, a PC, a tablet PC, and a notebook. In order to use the system 1 of the present invention, the user must evaluate the creditworthiness prior to the advance payment of goods. To this end, the user's information including the user's personal information, account and user terminal number, . The user who has been pre-paid for the goods based on the credit evaluated by the system (1) of the present invention reimburses the price corresponding to the goods, that is, the price of the goods. A text message or an e-mail indicating that a redeemable date is coming can be sent to a user's smartphone, a PC or a tablet PC. Further, after the repayment period is over, the payment is automatically made through the user's credit card or account. At this time, the automatic payment can be performed in cooperation with the user terminal.

The PG server is a subject having payment information including all payment details between an individual and a company using the electronic payment service. The PG server extracts only the payment information of the user from the PG server and transmits the payment information to the system 1 of the present invention as data . Here, the payment information includes the name of the company that the user made a payment, the amount the user paid, and the time point the user made payment. Furthermore, based on this payment information, the system 1 of the present invention may provide a basis upon which a user's credit can be assessed.

An affiliate server is a server of a business affiliated to use the system (1) of the present invention, wherein the affiliate is a business affiliated with a shopping mall, a game, a school, a playground, An affiliate server can be an affiliate server for an individual PC for a DB server or an aggregate of individual PCs for a server provided in an affiliate, or if the server can be operated through a tablet PC or a smart phone provided by an affiliate, Of course. The affiliate server has affiliate information including an industry type of an affiliate company using the system 1, a usage period of the system 1, and a payment history of an amount by a payment means. Such affiliate information provides a basis for evaluating the user's credit based on whether or not the user transfers the amount of the goods sent to the affiliate when the user wants to receive the goods of the affiliate, and a detailed description thereof will be described later.

In other words, the system 1 of the present invention defines a server serving as a module as an application server, a server storing data as a database server, and interworking with the application server and the database server, And may have modules in the context of clarifying configurations that may be included in the system 1 of the present invention.

The detailed configuration that can be included in the system 1 of the present invention will be described as follows. Here, in the system 1 of the present invention, a module is a subject that performs specific functions in cooperation with software based on hardware (including a CPU or a memory) having an arithmetic and analysis unit.

Fig. 2 is a block diagram showing the detailed configuration of the system 1 of the present invention, and Fig. 3 is a flowchart showing the process of the present invention.

2 and 3, the system 1 of the present invention includes a user information input module 100, a credit evaluation interface 200, a redemption condition determination module, a prepayment management module 400, a redemption management module 500 as a basic configuration, and may further include a pre-advance re-managing module 600.

The user information input module 100 receives user information from a user who intends to use the system 1 of the present invention. The user information includes a user name, an age, an address, a mobile phone number, account information, , Workplace, affiliation, and so on. Since the system 1 of the present invention receives a reimbursement of the amount of money after prepaid of the user determined to have a high credit, the stability of the system 1 is very important and accurate authentication of the user information is required. Therefore, the user information input module 100 may additionally include a user information authentication unit 110. [ In addition, the system 1 of the present invention may further include an electronic signature input unit 120 for receiving a digital signature for payment from the user, since the system 1 proceeds with payment after the user prepares the goods. Since the system 1 of the present invention receives user information and digital signatures that require security, it is desirable that the user's agreement on their use be premised.

The credit evaluation interface 200 evaluates the credit of the user to determine whether the user is a suitable person to use the system 1 of the present invention in order to secure the stability of the system 1 of the present invention . To this end, the credit evaluation interface 200 includes an external information input module 210, a payment pattern analysis module 220, a purchase pattern inquiry module 230, a credit rating calculation module 240, a credit rating derivation module 250, A compliance determination module 260, and an approval determination module 270. [

The external information input module 210 receives input from the PG server and the affiliate server and generates external information as external information. The external information input module 210 can evaluate the user's credit by grasping the purchasing and payment habits of the user based on the generated external information . Here, the external information means that the external information includes multimedia payment information, which is payment and payment information inputted from a server of a multi-media including a card company and a communication company using PG. Further, the affiliate information including affiliate information including the duration of use of the system (1) of the affiliate, the history of payment and amount of money by the payment means, purchase and deposit history by period, installment history, delinquency / accident history, . ≪ / RTI > In other words, the affiliate information includes the amount of money deposited by the user through bank transfer or direct debit, rather than payment through card or mobile. That is, since the external information includes the payment details of the user and the fee payment history, it is possible to know the field of the goods mainly used by the user, the amount of payment for the specific goods and the method thereof, and the user's credit can be evaluated based on this.

The settlement pattern analysis module 220 provides a function of extracting user settlement information corresponding to the user information from the external information and analyzing the settlement pattern of the user. That is, the payment information of the user including the place, the amount, and the payment means where the user made the settlement is extracted from the multi-media settlement information included in the external information, and the settlement pattern of the user, that is, . For example, payment information related to payment of a card used by a user and payment of a mobile phone charge can be extracted from external information to analyze a payment pattern of a user. For this, the payment pattern analysis module 220 may include a multimedia analysis unit 221, an additional analysis unit 222, and a payment pattern information generation unit 223.

The multi-media analyzing unit 221 provides a function of extracting and analyzing user payment information, which is payment information of the card company or the communication company, or payment of the fee among the multi-media payment information. For example, if it is desired to analyze the settlement pattern through the mobile phone charge details of the user input from the communication company used by the user first of all the multi-media settlement information, the current number usage period and payment history of the user among the user payment information, The result of the analysis of the multimedia is analyzed by analyzing the settlement pattern of the user, the payment history of the terminal charge, additional information use and settlement, the pattern analysis, the history of the mobile settlement use, the overdue delinquency and the accident, and the credit rating of the carrier itself.

And a settlement pattern information generation unit 223 for generating settlement pattern information based on the result of the multi-media analysis. The settlement pattern information generating unit 223 generates the settlement pattern information of the user based on the result derived from the multimedia analyzer 221. A user's credit can be evaluated based on the generated payment pattern information, and a detailed description of a method for evaluating a user's credit will be described later.

In addition, it may further include an additional analyzer 222 for analyzing the payment information of the user once more to more accurately analyze the payment pattern of the user. Here, if the multimedia analysis unit 221 derives the results of the user's multimedia analysis based on the multimedia payment information, the additional analysis unit 222 may add the additional payment information based on the additional payment information, which is information on the amount of goods deposited by the user on the affiliate side The results of the analysis are derived. Further, the additional payment information and the settlement pattern information are integrated and analyzed, and conversely, it is also possible that the additional payment information and the settlement pattern information are integrated.

By analyzing the payment pattern of the user twice in this way, the payment habits of the user can be more accurately derived, and the user's repayment ability can be estimated through the payment amount and the delinquency.

The existing credit rating method was based on only the financial aspects such as education, occupation, and property, but it did not take into account the financial aspects (for example, human psychology and economic flow) . Thus, we can conduct a more advanced credit evaluation by surveying the consumer's psychology and reflecting it.

Accordingly, the purchasing pattern inquiry module 230 grasps the pattern of purchasing goods by the questionnaire, and provides the role of conducting the credit evaluation reflecting the user's consumer psychology. At this time, the user inputs the questionnaire answer directly through the user terminal, and the purchase pattern inquiry module 230 provides the function of generating the purchase pattern information that grasps the purchase pattern of the user through its analysis. The purchasing pattern inquiry module 230 includes a questionnaire providing unit 231, an answer input unit 232, a term extracting unit 233, a term grouping unit 234, a representative term setting unit 235, (236).

The questionnaire providing unit 231 generates a questionnaire and provides the questionnaire to the user to investigate the purchase tendency and purchase habits of the user. At this time, the questionnaire may be divided into a multiple-choice question and a short-answer question, but a plurality of questionnaires are provided in a complex manner so that the user can freely input the purchase propensity and the buying habit into the system (1) Algorithm. As an example of the questionnaire, there are two types of questionnaires: 'What kind of product do you mainly buy?', 'What kind of service do you mainly use', 'When do you want to buy the goods or services you listed', ' Such as 'what is to buy together with the purchased product or service', 'what is the purpose of purchasing a specific product', and the like, to the user. In addition, a method of providing a questionnaire may include a method of transmitting a questionnaire to a user terminal in the system 1 of the present invention, a method of transmitting a text message in which a URL address of a site generating the questionnaire is written to a user terminal, Section, and the like.

The answer input unit 232 receives an answer to a questionnaire from a user who has been provided with the plurality of questionnaires through a user terminal. It is a matter of course that the form of the answer is naturally a sentence. Thus, the answers thus obtained are analyzed through the following configuration to provide a function of grasping the purchase pattern of the user.

The answer entered by the user is composed of a plurality of answer sentences including a number of terms. Therefore, a kind of text analysis process is needed in order to understand the fact that the user is interested in what goods and the reason why the goods are purchased.

The term extraction unit 233 extracts a term from the sentence of the answer inputted through the answer input unit 232. [ One sentence is composed of a plurality of terms, and functions to extract only the terms after dividing the terms included in the answers into units of words. The term extracting unit 233 divides the texts included in the answer sentence into phrases, and then renders the sentence using the sentence text, such as a noun having a specific meaning and emotional text indicating emotion, And it is a role to make it a term. This is one of the processes that can be performed as a basic operation in the general text analysis process. The term extraction unit 233 is a factor that can be changed to a user's purchase pattern to be finally grasped according to how small the terms are divided into Of course.

However, we do not intend to complicate the criteria for dividing the terms here, but merely divide the terms based on a single word, and since the word may contain the search or affix, it is recognized as a term and excluded It is of course possible to have a configuration for utilizing it.

When the terms can be judged in the answer sentence, the term grouping unit 234 can derive a valid term from the grouping factor grouped according to the frequency and proximity of the terms. For example, it is understood that the word 'cosmetics' most frequently appears in the answer sentence or the word 'travel' repeatedly appears in a specific sentence among the answer sentences. In the case of proximity, the terms of 'travel' and 'enjoyment', or 'travel' and 'want to go' are likely to be repeated, although they may vary somewhat depending on the method of application. At this time, other terms are located at a certain distance between "travel" and "pleasure" or "travel" and "want to go" such as "the last trip was very pleasant" or " is average. Thus, since these books have different degrees of proximity such as book / purchased and notes / purchased, when these samples are collected in various ways, comparing these proximities can actually determine how much the two words are related. In the case of this proximity judgment, various data mining engines that are used in work such as word mining in the big data field are disclosed and it is possible to utilize them. However, the proximity and appearance frequencies are compared with each other Is an inefficient way to deal with a lot of data.

Therefore, as a method for solving such a problem, the term grouping unit 234 groups the terms using a grouping factor. The grouping factor is a method of repeatedly comparing two or more terms to the frequency of occurrence and proximity, The importance of each of the included terms is to be grasped to derive effective terms to be used in actual analysis. In other words, it is possible to judge the level of the validity of the terms through the grouping factor.

This grouping factor is a configuration that allows a user to preliminarily extract a term that is expected to be meaningful rather than to compare and judge all the terms at the time of terminology analysis, and to utilize the extracted terms. In other words, the frequency and proximity of each term can be used as an effective reference value.

It is assumed that the content of the answer sentence of the specific user is as follows.

I am a 27 year old female worker. I like traveling once a week. I had a great vacation trip to Jeju Island and Busan this summer. Especially, Jeju Island with a panoramic view was very good. Also, Busan, where the blue sea spreads, was also very good. However, my monthly salary is often difficult to cover. "

In these sentences, terms such as 'Jeju Island', 'Busan' and 'I like it' can be found in sets of words repeatedly appearing in the answer sentence. These terms are high when compared to other terms, and Jeju Island and Busan are located closer than Jeju Island / I was good. Of course, in the case of good, there is no proximity of other terms except for these three terms, so it can be confirmed that it is formed higher than 0, so if the effective term is derived through the grouping factor, It is possible to judge that a term can be derived.

When applying the grouping factor, it is possible to quickly extract terms because the extraction time of effective terms is short in comparison with other existing technologies. If the answer sentence length is long or the answer sentence is input from a large number of users, Even if you have the same hardware, you can do more analysis. It is also possible to provide a system 1 capable of simultaneously extracting valid terms more sharply.

In other words, the term grouping unit 234 groups the extracted terms and makes it possible to grasp the relationship between the extracted terms. At this time, preferably, a plurality of recognized terms can be grouped by a grouping factor (GF) to speed up the grouping of terms. At this time, the grouping factor used for grouping is calculated by the following equation (1).

Equation 1,

는 키워드 x와 근접한 키워드 y의 누적 합,

은 키워드 x와 근접한 전체 키워드의 누적 합,

는 키워드 y와 근접한 키워드 x의 누적 합,

은 키워드 y와 근접한 전체 키워드의 누적 합,

는 키워드 x, y 상호 간의 그룹화 팩터를 나타낸다.here,

Is the cumulative sum of the keyword y close to the keyword x,

Is a cumulative sum of all keywords close to the keyword x,

Is a cumulative sum of the keyword x and a keyword x close to the keyword y,

Is a cumulative sum of all keywords close to the keyword y,

Represents a grouping factor between the keywords x and y.

In other words, if the preceding answer sentence is continued to be used, the terms 'Jeju Island' and 'Busan' are used closely and frequently. Therefore, if grouping is performed, 'Jeju Island' and 'Busan' are likely to be grouped into one group.

By grouping a plurality of terms using the grouping factor calculated as described above, it is possible to effectively grasp not only the relationship among the terms but also the frequency at which the same term appears in the same sentence several times. In addition, in the grouping factor calculation, it is possible to calculate a grouping factor based on a relation between a specific term, that is, a relation between similar terms (for example, 'Jeju Island' and 'Busan' or 'Jeju Island' Grouping is possible.

The representative term setting unit 235 sets any one of the terms as a representative term based on the result grouped by the term grouping unit 234 described above. Grouping is performed by the grouping factor, Is set as a representative term. For example, supposing that a sentence contains both the goods purchased by the user and the place where they were bought, there is one term for goods, three terms for goods, five terms for notes, When two terms of stationary are calculated, it can be said that the most frequently used terms and the terms which are the main points of relation are the representative terms.

The purchase pattern information generation unit 236 generates the purchase pattern information of the user by recognizing what goods the user purchases mainly based on the representative terms set by the representative term setting unit 235. [ In other words, it is possible to grasp that the goods which the user frequently purchases often are notes by means of the representative term of notes as exemplified above. The purchase pattern information may be generated based on the identified representative terms and used to evaluate the user's credit, which will be described in detail later.

A method for evaluating the user's credit based on the settlement pattern of the user analyzed above and the purchase pattern of the user investigated through the purchase pattern inquiry module 230 will be described in detail.

The credit rating calculation module 240 provides a function of calculating a credit rating of a user based on the result of the combined analysis of the settlement pattern information and the purchase pattern information. More specifically, Quot ;, and compares the result with the score given to each of the terms included in the " score " To this end, the integrated analysis module includes a pattern value calculation unit 241, a comparison calculation unit 242, a comparison processing unit 243, an eigenvalue calculation unit 244, a significance calculation unit 245, and a credit rating calculation unit 246 .

The pattern value calculation unit 241 receives the payment pattern information and the purchase pattern information from the payment pattern information generation unit 223 and the purchase pattern information generation unit 236 and converts each information into a numeric value. More specifically, a payment pattern value can be calculated by assigning a numerical value to the payment means in the payment pattern information. For example, payment methods include credit card, account transfer and cash settlement. Credit card payment is the best way to prove the user's ability to repay. Therefore, it is possible to derive the settlement pattern value by giving the credit card a maximum value (for example, 100 points out of 100 points) and assigning the value different according to the degree of proof of the user's repayment ability. The purchase pattern information generation unit 236 may also assign numerical values to the representative terms digitized to calculate purchase pattern values.

The comparison calculation unit 242 compares the settlement pattern value derived from each settlement pattern information with the purchase pattern value derived from the purchase pattern information to calculate a comparison value. Therefore, it is possible to consider factors that make payment at the same time as purchasing by judging how much the user's payment propensity matches with the purchase propensity. In the calculation of such a comparison value, the following equation can be used.

Equation (2)

는 비교수치,

는 구매패턴수치,

는 결제패턴수치를 의미하며,

는 구매패턴수치에 대한 구매패턴가중치로서 그 범위는

이다.here,

Is a comparative value,

The purchase pattern value,

Means a settlement pattern value,

Is the purchase pattern weight for the purchase pattern value,

to be.

At this time, the purchasing pattern weight for purchasing pattern information is considered as the most important factor, which is the most important factor that the user purchases in order to evaluate the credit of the user. The purchase pattern weight is a value that can be set in advance, and the user can receive the purchase pattern weight in advance, or the specific purchase pattern information in the purchase pattern information is important information for evaluating the user's credit, A high purchase pattern weight may be preliminarily designated on the system 1 only for the purchase pattern information.

는 결제패턴수치에 대한 결제패턴가중치로 해석할 수 있으며, 즉 이는 유저가 결제 수단을 통해 재화를 구매할 때 그 금액을 잘 납부했는지를 고려하는 요소이다. 결국 상기 수학식 2는 결제패턴수치와 구매패턴수치에 각각의 결제패턴중요도 및 구매패턴중요도를 곱한 것의 합이 바로 비교수치라고 해석할 수 있다.In Equation 2,

Can be interpreted as a settlement pattern weight for the settlement pattern value, that is, it is an element that considers whether the user has paid the amount well when purchasing the goods through the settlement means. In Equation (2), the sum of the settlement pattern value and the purchase pattern value multiplied by the importance of the settlement pattern and the importance of the purchase pattern can be interpreted as a comparative value.

The comparison processing unit 243 performs a function of comparing the respective comparison values to generate a comparison matrix. In this case, when the comparison value is input, the preparation for determining the importance of each comparison value is performed. At this time, since the importance determination is performed through the comparison matrix analysis, a comparison matrix is generated. The comparison matrix generated at this time is as follows.

procession

은 상기 비교수치 각각을 비교한 비교행렬로서

의 정사각행렬,

는 행렬

의 원소,

는 상기 비교수치

개 중

번째 요소의 상대중요도,

은 비교수치의 총 개수이다.here,

Is a comparison matrix in which each of the comparison values is compared

Square matrix,

The matrix

Of element,

Lt; / RTI >

Of

The relative importance of the ith element,

Is the total number of comparison numbers.

In this case, the importance degree is a value obtained by comparing the settlement pattern value and the purchase pattern value with each other and determining the importance thereof. For example, if the importance is input by the user, or if it is already known on the system 1 that the information is important among the payment pattern information or the purchase pattern information, only a high importance is assigned to the system 1 ). ≪ / RTI >

The eigenvalue computation unit 244 computes a eigenvalue of the comparison matrix generated through the comparison processing unit 243, and when the comparison matrix is generated, the eigenvalue is calculated from the matrix. In order to calculate the eigenvalue, a eigenvalue can be calculated by creating a eigenvalue and finding a solution to the eigenvalue. In order to obtain the solution of the eigenvalue, a determinant is used. Can be calculated. These eigenvalues may appear as a single value or as multiple values. At this time, the eigenvalue can be calculated by finding the solution of the following equation.

≪

는 행렬

의 고유치,

은

단위행렬,

은 비교수치의 총 개수를 의미한다.here,

The matrix

The eigenvalue,

silver

Unit matrix,

Means the total number of comparison values.

In this manner, the eigenvalue can be calculated by deriving an answer of the determinant based on the above-described mathematical expression.

The importance calculating unit 245 calculates the importance from the comparison matrix based on the eigenvalues. If the eigenvalues are represented by a plurality of values, the importance is calculated based on the maximum value of the eigenvalues. The importance calculated is the importance of each comparison value, and the importance of each comparison value is calculated based on the eigenvalue.

The credit rating calculator 246 calculates the credit rating of the user based on the credit rating determined by the credit rating calculator 245. The credit rating calculator 246 calculates the credit rating of the user based on the credit rating obtained by the credit rating calculator 245. As described above, . ≪ / RTI > In addition, an additional weight may be added to the credit rating. The weight is a credit weight of the user, and is a weight of how much the user has paid the settlement price or the fee. Or may be input by the administrator of the system 1. Or a reimbursement weight calculated through a prepayment re-management module 600 to be described later (this will be described later in detail). A method of calculating the final numerical value is performed by the following equation (3).

Equation 3,

는 상기 유저의 신용평점,

는 상기 비교수치,

는 비교행렬

의 고유치 중 최댓값,

은 유저의 신용기반가중치를 의미한다.here,

The credit rating of the user,

Lt; RTI ID = 0.0 >

The comparison matrix

Of the eigenvalues,

Means the user's credit-based weight.

Based on the calculated credit rating, not only can the user determine whether the system 1 of the present invention can be used, that is, whether the user is capable of repaying the amount of money, The stability of the system 1 of the present invention can be ensured.

The credit rating derivation module 250 derives the credit rating of the user calculated by the credit rating calculation unit 246 by converting the credit rating into a credit rating. At this time, the criterion for deriving the credit rating is divided into points And the method of assigning from A to F grades can be set, and the method is not limited. It is needless to say that the criterion for dividing the credit rating can be set by the system 1 of the present invention and can be arbitrarily set by the manager of the system 1 of the present invention. For example, if the administrator of the system (1) sets a credit rating standard, the credit rating of A is 90 points from 100 points, the rating of B is 90 points from 80 points, You can set the rating criteria.

The suitability determination module 260 provides a function to determine whether the user is suitable to use the system 1 of the present invention based on the credit rating of the user derived by the credit rating derivation module 250. [ For example, when the user's credit rating is A-rated by the credit rating derivation module 250, the user can recognize that the credit rating is high enough to have the repayment ability. Therefore, this user can be said to be a user suitable for using the system 1 of the present invention. As another example, an example of a criterion for making a suitability determination for use of the system 1 of the present invention may be that a user who has received a credit rating from A to C may determine that the system 1 is suitable for use have. At this time, the criterion for judging compliance and nonconformity can be set by the system 1 of the present invention like the standard setting of the credit rating, and can be set by the manager of the system 1 of the present invention.

The approval decision module 270 is for authorizing the user who has been judged by the conformity judgment module 260 to use the system 1 of the present invention. As soon as the system 1 of the invention is approved for use, the system 1 of the present invention can be used, i.e., the goods can be pre-paid.

The redeployment condition setting module 300 sets a condition for repayment of the amount of goods after the user who has been approved by the approval decision module 270 receives the prepaid goods. In other words, when the designated repayment period passes, the user performs a function of setting the condition so that the amount is repaid through the payment means selected by the user. For this, the redeem condition setting module 300 may include a payment means selection unit 310, a redemption period designation unit 320, and an automatic payment determination unit 330.

The payment means selection unit 310 allows the user to select a payment means to be automatically settled when the automatic payment is made because the amount is not redeemed over the redemption period. The payment means may be a credit card, a bank transfer, There is no restriction as long as it is a means by which the user can make payment. In addition to this, a template for selecting a payment means may be formed and displayed on the display of the user terminal, so that the user can provide the function so that the user can select the payment means through the user terminal.

The reimbursement period designation unit 320 designates a reimbursement period for the user who has been pre-paid by the system 1 of the present invention to repay the amount for the goods, Allow the user to repay the amount within the period. For example, if the user receives the advance payment of goods on December 20, 2017, the repayment period can be set to 5 or 7 days from the day following the prepayment date. At this time, the method for setting the repayment period can be set by the user, and the system 1 of the present invention can set the repayment period.

When the user does not repay the amount within the designated repayment period, the automatic settlement journal unit 330 receives an automatic settlement date in which the amount is automatically paid through the payment means selected by the payment means selection unit 310 And performs the function of designating the position. For example, if the end of the redemption period is December 27, 2017, the automatic redemption date may be designated as the following day, December 28, 2017. In this case, since the system 1 of the present invention can ensure the stability of the present invention when the automatic settlement date is not automatically redeemed within the repayment period, the system 1 of the present invention automatically releases the day after the end of the repayment period It is preferable to designate the settlement date.

The prepayment management module 400 provides the environment for prepaid goods to the user who is approved for use of the system 1 of the present invention. For this purpose, the prepaid management module 400 includes a goods database 410, a goods display unit 420, (430), and a material supplier (440).

The goods database 410 stores goods information including goods names and goods images and goods prices for a plurality of goods to be prepaid. Here, the term "goods information" refers to the name of the goods, the representative image of the goods, the image of the goods, which is an image containing the example of the use of goods, and other information about the additional goods And the like.

The goods information may be registered in the goods database 410. The affiliate company may store the goods information in the goods database 410 to register the goods in the goods database 410. The system 1 may directly sell the goods It is needless to say that the manager of the system 1 of the present invention may input goods information.

The goods display unit 420 functions to create and display a goods selection interface composed of a plurality of selection areas on the display of the user terminal held by the user. Such a goods selection interface has a selection region in which the goods information is posted for performing goods selection and a template for displaying the goods information to the user as the goods information is clicked by arranging the goods information in the selection region . At this time, preferably, the goods information, the goods image, and the goods price are basically arranged in the selection area, and the information about the goods is transmitted to the user. Further, the selection area may additionally display other information on the goods (detailed description including the characteristics of the goods and usage) in addition to the goods information.

The goods selection unit 430 performs a function of displaying a template showing the goods information on the display of the user terminal so that the user can select the goods to be pre-paid. In other words, if the goods are provided by the goods display unit 420, the user can select the goods to be pre-paid through the user terminal.

The goods supplier 440 performs the function of providing the selected prepaid target goods to the user. If the prepaid target goods is a specific commodity, the commodity supplier 440 may send the commodity to the user, or if the prepaid target commodity is a specific service The service can be provided to the user.

The system 1 of the present invention can directly perform the goods prepayment role for the user through the configuration of the prepayment management module 400. In this case, according to the selection of the goods of the user, It is possible to grasp the amount of the goods in a more efficient manner, and the distribution margin can be expected to be reduced.

The redemption management module 500 manages the reimbursement of the user for the amount of money, and if the redemption has not been made, the redemption management module 500 provides a function of automatically performing settlement processing through the payment means selected by the payment means selection unit 310, The reimbursement reminder unit 510, the reimbursement determination unit 520, the automatic payment request unit 530, and the automatic payment cancellation unit 540 may be included for this purpose.

The reminder providing unit 510 provides a notification to inform the user that the end of the repayment term comes, and the notification can be transmitted to the user terminal in the form of a text message or in the system 1 of the present invention, There is no limit to the method, for example, by letting the system (1) sound a notification when the end of the repayment period arrives. At this time, the time when the notification is provided can be set to two days before or three days before the last day of the repayment period, and it is needless to say that the user or the administrator of the system 1 can set the timing at which the notification is provided.

The reimbursement determination unit 520 identifies whether the user has repaid the amount of the goods that have been pre-paid. Whether or not the reimbursement is recognized can be determined by the system 1 of the present invention or linked to the affiliate server It can be determined whether the amount has been repaid. Specifically, the reimbursement determination unit 520 may determine whether the reimbursement amount of the goods that the user has pre-paid is redeemed by credit card or account transfer in cooperation with the PG server, or may be correlated with the affiliate server, The amount of money can be directly repaid to the affiliate side, and the details thereof can be grasped in the system of the present invention by itself.

The automatic payment request unit 530 requests the PG server to automatically settle the amount through the payment means selected by the payment means selection unit 310 when the user does not repay the amount despite the arrival of the repayment period Role. Specifically, if the user does not redeem the money until the end of the reimbursement period, the user sends a request to the PG server in order to automatically settle the amount to the payment means selected by the user on the automatic settlement day, which is the day after the last day of the reimbursement period.

The automatic payment cancellation unit 540 sends a cancellation request for the automatic payment to the PG server within a predetermined period after the repayment if the user repays within the repayment period. In other words, if the user repays the amount within the repayment period, the information on the automatic payment stored in the system 1 of the present invention is lost, and the revocation of the automatic payment is requested to the PG server. At this time, a text message indicating that the automatic payment has been canceled can be transmitted to the user.

The prepayment re-management module 600 provides a role of re-managing the goods if the user desires to receive the goods through the system 1 of the present invention again after the repayment of the prepaid goods is completed after the automatic payment is made. In other words, if the user re-uses the system 1 of the present invention to select the same or different goods and receive prepayment, the user is informed of how much the user has been reimbursed in the past and the user using the prepaid goods You can choose the goods to do. The prepayment re-managing module 600 includes a redeeming information generating unit 610, a redeeming factor determining unit 620, a redeeming weight calculating unit 630, and a blind executing unit 660, 640 and a late investigation performing unit 650. [0054] FIG.

The redemption information generating unit 610 generates a redemption information using the redemption date, the redemption amount, the number of times the user repays the redemption amount within the redemption period, and the redemption date until the redemption date The reimbursement information including the reimbursement required period of the reimbursement information. It provides the role of calculating the repayment sincerity that can grasp how well the user has been repaid.

The redeemability determining unit 620 determines the redeemability of the user based on the redeeming information. The degree of integrity is determined by the number of times the user repays the amount of money in the redeeming period with respect to the total number of redeemed goods to the user And there is no restriction on how to judge sincerity.

The reimbursement weight calculation unit 630 provides a function of calculating a reimbursement weight based on the reimbursement information obtained by the reimbursement determination unit 620 and the reimbursement information generated by the reimbursement information generation unit 610 . At this time, the repayment weight can be calculated by the following equation.

Equation.

은 상환 가중치,

는 유저가 선지급 받은 재화의 금액을 상환한 총 횟수,

는

번째 선지급 시 유저의 상환 성실도,

는

번째 선지급 시 유저가 금액을 상환한 상환일에서 재화를 선지급 받은 일을 뺀 상환소요기간의 역수,

는

번째 선지급 시 유저가 금액을 상환한 횟수를 의미한다.here

Is the repayment weight,

Is the total number of times the user repaid the amount of the good that was pre-paid,

The

The reimbursement of the users'

The

The reciprocal of the period required for repayment, minus the advance payment of the goods from the repayment date on which the user repays the amount,

The

Means the number of times the user repaid the amount at the pre-payment time.

The credit rating of the user may be revalued on the basis of the reward weights thus calculated. In other words, the credit rating of the user can be calculated by further reflecting the repayment weight in the above Equation (3). Further, the reimbursement weights may be given additional weight by other factors. A detailed description of a method of reimbursing the reimbursement weights through the different weights will be described later.

Furthermore, the user can leave a post about prepaid goods. In the latter period, not only the user's consumption patterns and purchase patterns can be reasonably grasped, but the reimbursement weights can be reassessed based on the result of analyzing the latter. That is, since the testimony of the input of the pre-paid goods includes the user's psychology, the reward weights reflecting the user's psychology can be calculated.

For this, a configuration of the late investigation performing unit 650 may be provided. The late investigation performing unit 650 is for investigating the late period of the pre-paid goods to the user after the goods are pre-paid. The late investigation performing unit 650 includes a late entry part 651 for inputting the latter period from the user, An effective keyword extracting part 652 for deriving a valid keyword through a grouping factor grouped according to the frequency of occurrence and proximity of the keyword, and an effective keyword extracting part 652 for comparing the effective keyword with the goods database 410 And a target goods selection part 653 for selecting the goods to be prepaid to be prepaid to the user.

A late entry part 651 may be included to investigate the latter, which provides an input space so that the user can freely express satisfaction or inconvenience after using the pre-paid goods, Enter the latter. In this case, since the received sentence is a sentence freely written by the user, it is preferable to extract and analyze the keyword included in the sentence sentence in order to analyze the inputted sentence sentence.

Here is an example of the latter sentence entered by the user.

'System. The goods I chose were usually the ones I usually buy. I was very satisfied with my trip to Jeju and Busan through the purchased travel products. The trip to Jeju Island was satisfactory because I could see the panoramic view. I was also pleased to see the beautiful sea in Busan. Next time I get another advance payment, I want to re-select the travel products that I was satisfied with. "

The effective keyword extracting part 652 has a function of extracting a valid keyword (for example, a keyword that indicates the satisfaction of the user with respect to the pre-paid goods) out of the keywords included in the later sentence inputted to the later input part 651 . The principle of extracting the effective keyword is the same as that of the term extraction unit 233 and the term grouping unit 234, and therefore, a detailed description thereof will be omitted. That is, if the effective keyword of the latter sentence is extracted on the same principle as that of extracting the term of the answer sentence in the purchase pattern inquiry module 230, it is 'Jeju Island', 'Busan', 'Satisfied'.

When grouping is performed in this manner to extract a valid keyword, the plurality of goods information that are compared with the goods information included in the goods database 410 again through the effective keyword are selected as prepara- tion target goods . In other words, in the case of Jeju Island and Busan, detailed travel information of Jeju Island and Busan such as sightseeing spot, image of sightseeing spot, admission fee of sightseeing spot and operating hours can be derived. Such information uses a sort of similarity comparison. In the case of the similarity comparison, a commercial engine or a related analysis method widely used in the field of big data analysis (especially text analysis) can be applied, and a detailed explanation thereof is omitted .

In analyzing the input sentence, the effective keyword is derived based on Equation (1). The number of occurrences of the keywords included in the sentence sentence, the degree of proximity, and the number of effective keywords The additional weight can be calculated.

The additional weight can be calculated by the following equation (4).

Equation 4.

는 추가 가중치,

는

번째 선지급 시 상기 유효 키워드의 등장빈도,

는

번째 선지급 시 상기 유효 키워드의 근접도,

는 상기 후기문장에 포함된 전체 단어의 개수 대비 상기 유효 키워드의 개수,

은 총 선지급한 횟수를 의미한다.here,

Is an additional weight,

The

The occurrence frequency of the effective keyword at the time of the first pre-

The

The proximity of the effective keyword at the first pre-

The number of the effective keywords, the number of the effective keywords,

Means the total number of prepayments.

That is, the additional weight is a weight considering the effective keyword included in the later sentence, and can be re-calculated to calculate the reward weight again. The recalculated reimbursement weight can be calculated by the following equation (5).

Equation (5)

은 상환 가중치,

는 유저가 선지급 받은 재화의 금액을 상환한 총 횟수,

는

번째 선지급 시 유저의 상환 성실도,

는

번째 선지급 시 유저가 금액을 상환한 상환일에서 재화를 선지급 받은 일을 뺀 상환소요기간의 역수,

는 i번째 선지급 시 유저가 금액을 상환한 횟수,

는 상기 추가 가중치를 의미한다.here,

Is the repayment weight,

Is the total number of times the user repaid the amount of the good that was pre-paid,

The

The reimbursement of the users'

The

The reciprocal of the period required for repayment, minus the advance payment of the goods from the repayment date on which the user repays the amount,

Is the number of times the user repaid the amount at the i-th advance payment,

Quot; means the additional weight.

은 상기 상환기간지정부(320)에 의해 설정된 상환기간, 유저의 상환 성실도, 유저가 금액을 상환한 날을 기반으로 산출할 수 있다. 즉, 유저가 상환기간 내에 상환을 했다 하더라고 상환기간의 말일에 가깝게 상환(즉,

값이 작음)한 것이 아닌 상환기간의 초일에 가깝게 상환했다면(즉,

값이 크다면) 이에 대한 가중치를 부여하는 것이다. 다시 말해, 상환기간 내에 상환을 마치기만 해도 유저의 상환 성실도는 증가할 수 있으나, 말일에 상환하는 것보단 초일에 상환하는 것이 유저의 상환 성실도를 더욱 증가시키는 요소가 된다. 이에 더하여, 상기 추가 가중치

를 더 반영하여, 유저가 입력한 후기, 즉 유저의 소비 패턴이나 합리적인 구매 패턴을 고려한 상환 가중치를 산출할 수도 있다.That is, the reward weights calculated through Equation (5) are calculated by reflecting the satisfaction of the user's goods included in the latter sentence. To describe Equation 5 above in more detail,

Can be calculated on the basis of the repayment period set by the repayment period designation unit 320, the user's redeemable state, and the date on which the user repays the amount. That is, even if the user has made a repayment within the repayment period, the repayment is made close to the end of the repayment period (that is,

If the repayment is close to the first day of the repayment period (ie, the value is small)

If the value is large). In other words, even if the repayment is completed within the repayment period, the repayment of the user may be increased, but the repayment on the first day rather than the repayment on the last day is a factor that further increases the repayment of the user. In addition,

To calculate a repayment weight in consideration of a later input by the user, that is, a consumption pattern of the user or a reasonable purchase pattern.

For example, suppose that the period for which the amount of prepaid goods must be repaid is from December 18, 2017 to December 22, 2017. The first day of the repayment period is 18 days and the last day is 22 days. If the user repays the above amount on the 18th day, it can prove the user's loyalty rather than redeem it on the 22nd day. Therefore, when the user makes a repayment as soon as possible within the repayment period, the value of the repayment weight can be increased.

Furthermore, when the keywords included in the later sentence are listed in a line, the late investigation performing unit 650 may further include an interval grasp part 654 for grasping the interval, which is the number of keywords located between the listed keywords. By applying the interval to the proximity and using the interval weight according to the size of the interval, the concept of proximity is further extended when the grouping factor is obtained, And a blind keyword extracting part for extracting a blind keyword that is a keyword that does not match the extended valid keyword.

For example, in the case of Jeju Island and Busan, we used the example of the previous explanation, and repeatedly appeared twice. Also, Jeju Island / Satisfied Na Pusan / Satisfied the case of 2 times repeatedly appeared together, but there is more than one term located between them, and it is seen that the interval of the number of terms appears differently when appearing.

Of course, in the case of 'satisfactory', it is generally possible to define it as an emotional term in the text analysis process. Since it is a keyword that can determine the desired content, it can not be directly compared with the goods database 410 of the present invention Although it is a keyword, it can be judged that the proximity degree of Jeju Island / Busan is lower than that of Jeju Island / Busan, assuming that it is possible to compare with the goods information included in the goods database 410 in the "satisfactory" place Do. The closer the interval is, the higher the distance, and the lower the interval weight can be used.

Therefore, in order to actually select the goods to be prepaid to the user by the system 1 of the present invention, it is necessary to consider the interval interest proximity difference of these keywords. Therefore, it is possible to further include the structure of the goods tendency grasp part 656 which can grasp the tendency of the goods selected by the user through the grouping factor grouped with the interval weights applicable to the proximity according to the size of the intervals . However, since the interval weights can be changed according to the user's writing habits, it may be considered to control the application range of the interval weights through a method of previously conducting a pre-examination process to determine the user's writing habits in advance Of course.

And a goods propensity grasp part 656 for grasping the extended goods information, which is the goods information of the goods that the user wishes to be pre-paid based on the extended effective keyword derived as above, to generate the goods propensity information. This is for blind processing of goods that do not coincide with the goods propensity of the user. The goods propensity identification part 656 identifies the provided goods identification part 640, the target goods selection part 653, and the extended keyword derivation part 655) to identify the tendency of the goods that the user wishes to be pre-paid and to generate them as the goods propensity information. More specifically, the user can grasp the goods information of the pre-paid goods while using the system 1 of the present invention through the provided goods identification unit 640, and at the same time, By comparing the goods information of the selected goods and generating the goods incentive information, it is possible to know which goods the user is pre-paid mainly, and thereby, the user's goods tendency can be grasped.

Further, a blind good selection part 658 for comparing the extended effective keyword and the good propensity information to designate the good propensity information determined to be inconsistent as blind good information may further be provided. Before describing the equation, a formula for calculating the extended grouping factor and a formula for shipping the reward weights are described.

Here, a method of calculating an extended grouping factor for deriving an extended valid keyword is expressed by Equation (6) below.

≪ / RTI >

는 확장 그룹화 팩터,

은 전체 그룹화 횟수,

는

번째 그룹화에서 키워드

에 대한 키워드

의 그룹화 팩터,

은

번째 그룹화에서 키워드

와 키워드

의 간격,

는 키워드

와 키워드

의 최대 간격을 의미한다.here,

Is an extended grouping factor,

The total number of grouping,

The

Keywords in the ith grouping

Keywords for

Lt; / RTI >

silver

Keywords in the ith grouping

And Keywords

Spacing,

Keywords

And Keywords

.

In this way, when the keyword is extracted through the extended grouping factor derived from Equation (6), the intervals of the respective keywords can be compared. In this case, if the later sentences are arranged in a line to compare the intervals of the derived keywords with respect to the total number of keywords, it means an average value of the keyword intervals, and the correction weights can be calculated through this.

The correction weight is calculated based on the derived grouping factor, and the above-described reward weight can be calculated by reflecting the calculated correction weight. The reward weight can be calculated by Equation (7).

Equation 7,

은 상환 가중치,

는 유저가 선지급 받은 재화의 금액을 상환한 총 횟수,

는

번째 선지급 시 유저의 상환 성실도,

는

번째 선지급 시 유저가 금액을 상환한 상환일에서 재화를 선지급 받은 일을 뺀 상환소요기간의 역수,

는 i번째 선지급 시 유저가 금액을 상환한 횟수,

는 상기 추가 가중치,

는 상기 보정 가중치를 의미한다.here

Is the repayment weight,

Is the total number of times the user repaid the amount of the good that was pre-paid,

The

The reimbursement of the users'

The

The reciprocal of the period required for repayment, minus the advance payment of the goods from the repayment date on which the user repays the amount,

Is the number of times the user repaid the amount at the i-th advance payment,

Lt; RTI ID = 0.0 > weight,

Quot; means the correction weight.

번째 선지급 시 후기문장의 키워드의 간격을 비교하면, 간격이 가까울 때에는 가중치가 커지고 간격이 멀면 가중치가 작아짐을 알 수 있다. 다시 말해, 키워드가 서로 가까울수록 문장 내에서 중요한 역할을 한다는 의미이고,

번째 선지급 시점이 중요한 시점이라고 판단할 수 있다.Equation (7) is a mathematical expression for rewriting the reweighted weights by reflecting the correction weights instead of the additional weights, unlike Equation (5). Reflection of the correction weights based on the extended grouping factor means that the reimbursement weight is calculated by reflecting the extended valid keyword, i.e., the psychology inputted at the later stage of the user, in the reimbursement weight. That is, if the credit rating of the user is re-calculated by reflecting the re-calculated weighted value through the formula 7, the credit rating of the user that more accurately reflects the consumption pattern and purchase pattern of the user can be calculated. Further, through Equation (7)

If we compare the intervals of the keywords in the second sentence, we can see that the weights become larger when the intervals are close to each other and the weights become smaller when the intervals are longer. In other words, the closer the keywords are to each other,

It can be concluded that the time point of the first pre-payment is important.

The credit rating of the user can be calculated again by reflecting the calculated reward weights, and the user can perform the role of selectively showing the goods to be prepaid to the user based on the reward weights.

If the derived value is greater than a certain criterion as the extended validity keyword, the weight is applied to each item of information even if it does not appear repeatedly. Therefore, unlike the previous effective keyword It can be said that the proximity interest interval can be applied to the grouping factor as a factor that can be derived as an extended effective keyword and helps grasp a more detailed user's propensity to select goods.

After analyzing the latter sentence as described above, the user grasps the patterns of the goods selected by the user. Then, the goods that are far from the user's tendency are blind processed so that they are not displayed on the display of the user terminal, And so on.

In addition, the prepayment re-managing module 600 may further include a providing-goods acquiring unit 640 for acquiring information on the pre-paid goods to the user. This is to grasp the pre-paid goods to the user in cooperation with the above-mentioned material supplier 440, so that it is possible to know what field of the goods which the user has mainly pre-paid. Thus, based on the user's pre-paid goods and the user's input, the user can know the tendency of the goods that are pre-paid. Details of how the provided goods identification unit 640 interlocks with other configurations will be described later.

Also, the pre-paid re-managing module 600 may further include a blind good selection part 658, which compares the extended effective keyword and the good propensity information to determine that they are inconsistent And to designate the goods propensity information as blind merchandise information, so that the goods that do not match the user's goods propensity are selected as blind merchandise. Further, such blind goods can designate the goods information as blinds not only according to the user's goods propensity but also according to the user's repayment reliability. Specifically, if the user is sincere in repaying, he can provide expensive goods, and if he has not made the repayment, he will blind the expensive goods and display only low-priced goods. It is needless to say that the criterion for blind processing of the goods according to the price based on the repayment degree of the user can be arbitrarily set by the administrator of the system 1 of the present invention and can be set in the system 1 itself.

In addition, the pre-advancement management module 600 further comprises a blind execution unit 660 for selectively blinding the goods not to be prepaid to the user among the goods to be prepaid according to the size of the re-calculated credit rating . More specifically, the blind process performs blind processing on the blind material. In this case, the blind process includes blind processing of the blind material in the selected area, or blind processing of the blind material so as not to be visible in the selected area . Therefore, it is possible to prevent the main server of the system 1 of the present invention from reading the goods information of the blind goods, thereby preventing the goods information designated by the blind goods from being displayed on the user terminal. For example, if the credit rating recalculated by the repayment weight is high, the number of goods to be blind processed will be small, and if the re-calculated credit rating is low, the number of goods to be blind processed will be large.

Furthermore, the blind executing unit 660 receives the reference reimbursement amount and the reference reimbursement goodness degree from the manager of the system 1 of the present invention and generates a release criterion setting part 661, and a blind release portion 662 that allows the blind to be released by the release criteria.

The release criterion setting part 661 receives and sets the release criterion from the manager of the system 1 of the present invention and generates the release criterion information. Can be set to be released so that the blind is canceled when the user's redeemability is more than a specific value. At this time, the blind can be set to be released based on the credit rating of the re-calculated user in addition to the redeemability of the user. Or the blind is released when the user repays a predetermined amount or more.

The blind cancel part 662 can release the blinds of the blind processed goods by setting based on the user's repayment reliability, the user's credit rating, or the user's written review. The credit rating, the credit rating, or the user's input, to cancel the blind process for the blind good. The user can then re-enter the redeemable credit rating or the credit rating If you recover, or if you leave a positive testimonial about the good that has been pre-paid, you can unblind related goods.

As described above, in the system 1 of the present invention, the goods information is received from the goods database 410 and the goods are paid to the user. As the number of blind goods increases, the amount of goods that can be actually paid by the user can be reduced . According to the prepayment re-managing module 600, since the setting of the blind good is based on the redeemable sincerity, the goods pre-paid to the user, and the user's input, the user who has few blind items Can be seen to have done well in repayment.

As described above, the structure and the operation of the system for providing the automatic payment service after prepayment based on the credit evaluation using the analysis of the multi-media big data matching pattern according to the present invention are shown in the above description and drawings, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

1: System 2: User
3: PG server 4: Affiliate server
100: user information input module 110: user information authentication part
120: digital signature input unit 200: credit evaluation interface
210: external information input module 220: settlement pattern analysis module
221: Multicast analysis unit 222: Additional analysis unit
223: payment pattern information generation unit 230: purchase pattern inspection module
231: Questionnaire preparation 232: Answer draft
233 Term extraction unit 234 Term grouping unit
235: Representative term setting unit 236: Purchase pattern information generating unit
240: credit rating calculation module 241: pattern value calculation part
242: comparison calculation section 243: comparison processing section
244: eigen value calculating unit 245: importance calculating unit
246: credit rating calculation unit 250: credit rating derivation module
260: conformity determination module 270:
300: Repayment condition setting module 310:
320: Redemption period administration 330: Automatic payment journal administration
400: prepayment management module 410: goods database
420: goods display unit 430: goods selection unit
440: Commercialization 500: Repayment management module
510: Redemption notification providing unit 520:
530: Automatic payment request unit 540: Automatic payment cancellation unit
600: prepaid re-management module 610: re-
620: Repayment Guarantee Division 630: Repayment Weight Calculator
640: Provided goods identification unit 650:
651: late entry part 652: valid keyword extraction part
653: Selection of the target goods Part 654:
655: Extracting extended keywords Part 656: Identifying the nature of goods
657: Blind information extraction part 658: Blind material selection part
660: Blind execution part 661: Release criterion setting part
662: Blind Release Part

Claims (14)

A system for providing a prepayment automatic payment service after prepayment based on a credit evaluation using a multi-media big data matching pattern analysis,
A user information input module for authenticating the user information after receiving user information from the user;
A payment gateway (PG) server and affiliate server, receives external information including multi-media settlement information including payment details of a card company and a communication company, and affiliate information including a payment history of an affiliate, A credit evaluation interface for evaluating credit of the user based on the credit rating;
A redemption condition setting module for setting a redemption period and an automatic redemption date of the money, a payment means for receiving an amount of the goods at the time of pre-payment of the goods to the user;
An advance payment management module for prepaying the goods to be prepaid to the user;
And a redemption management module for recognizing whether or not the amount of money is redeemed from the user and executing automatic settlement so as to receive the redemption amount. Automatic Repayment Service System. The method according to claim 1,
The credit evaluation interface comprises:
A multi-media analyzing unit for extracting user settlement information, which is payment details of a user among the user information, from the multi-media settlement information and deriving a multi-media analysis result by analyzing a payment pattern of the user; A settlement pattern analysis module including a settlement pattern information generation section for generating a settlement pattern,
A purchasing pattern inquiry module for providing a questionnaire to the user and analyzing a purchase pattern of the user to generate purchase pattern information,
A credit rating calculation module for calculating a credit rating obtained by digitizing the user's credit based on the payment pattern information and the purchase pattern information;
And a conformity determining module for determining a pre-financing qualification of the goods of the user based on the credit rating. The system for providing a reimbursement automatic payment service based on a credit evaluation based on a multi-media big data matching pattern analysis. 3. The method of claim 2,
The payment pattern analysis module includes:
An additional analyzing unit for extracting additional payment information as a payment history of the user among the user information from the affiliate information, analyzing the payment pattern of the user by integrating the additional payment information and the payment pattern information, and generating additional pattern information Including,
Wherein the settlement pattern information generating unit comprises:
And the integrated payment pattern information is generated by applying the additional pattern information to the payment pattern information. The system for providing a prepaid post-payment automatic payment service based on a credit evaluation using a multi-media big data matching pattern analysis. 3. The method of claim 2,
Wherein the purchase pattern checking module comprises:
A questionnaire form for surveying the purchasing propensity and purchase habits of the user and providing the questionnaire to the user;
An answer input unit for inputting an answer sentence of the questionnaire from the user;
A term extraction unit for extracting terms included in the answer sentence;
A term grouping unit for deriving a valid term through a grouping factor obtained by grouping the terms according to the frequency and proximity of the extracted terms,
A representative term setting unit that sets any one of the valid term as a representative term,
And a purchase pattern information generation unit for generating purchase pattern information by grasping the purchase pattern of the user on the basis of the representative term. Automatic payment service provision system. 5. The method of claim 4,
The term grouping unit,
Wherein the grouping factor is calculated using Equation (1): " (1) "
Equation 1

(here,

Is the cumulative sum of the term y, which is close to the term x,

Is the cumulative sum of the terms in close proximity to the term x,

Is the cumulative sum of the term x, which is close to the term y,

Is the cumulative sum of the terms in close proximity to the term y,

Is the grouping factor between the terms x and y) 3. The method of claim 2,
Wherein the credit rating calculation module comprises:
A pattern value calculation unit for calculating a settlement pattern value and a purchase pattern value by receiving the settlement pattern information and the purchase pattern information,
A comparison calculation unit for comparing the payment pattern value and the purchase pattern value to calculate a comparison value;
A comparison processor for comparing the comparison values to generate a comparison matrix,
An eigenvalue calculation unit for calculating eigenvalues for the comparison matrix;
An importance calculating unit for calculating an importance from the comparison matrix based on the eigenvalues,
And a credit rating calculation unit for calculating the credit rating based on the importance. The system for providing a prepayment automatic payment service after a prepayment based on a credit rating using a multi-media big data matching pattern analysis. The method according to claim 6,
Wherein the comparison-
The comparison value is calculated by the following equation (2)
(2)

(here,

Is a comparative value,

The purchase pattern value,

The payment pattern value,

Is the purchase pattern weight for the purchase pattern value,

)
The comparison processing unit,
Generates the comparison matrix through the following matrix,
procession

(here,

Is a comparison matrix in which each of the comparison values is compared

Square matrix,

The matrix

Of element,

Lt; / RTI >

Of

The relative importance of the ith element,

The total number of comparative figures)
The credit rating calculator calculates,
Wherein the credit rating is calculated through the following Equation (3): < EMI ID = 4.0 >
(3)

(here,

The credit rating of the user,

Lt; RTI ID = 0.0 >

Lt; RTI ID = 0.0 >

Of the eigenvalues,

Is a credit-based weight value of the predetermined user calculated on the basis of the importance, The method according to claim 1,
The prepayment management module,
A goods database storing goods information including a goods name, a goods image and an amount of goods for a plurality of the goods to be prepaid,
A goods display unit for generating and displaying a goods selection interface including a plurality of selection areas on a display of a user terminal owned by the user,
A goods selection unit displaying a template showing the goods information and inputting the selected goods from the user,
And a goods providing unit for providing the selected goods to the user. A system for providing a prepayment automatic payment service after a prepayment based on a credit rating using a multi-media big data matching pattern analysis. The method according to claim 1,
The system comprises:
A reimbursement information generating unit for generating reimbursement information including the reimbursement date, the reimbursement amount, the reimbursement amount, and the reimbursement period from the date of the user's pre-payment to the reimbursement date from the reimbursement management module; A reimbursement degree determining unit for determining a reimbursement degree based on a comparison between the total number of times the user has paid the goods and the reimbursement number based on the reimbursement information and the reimbursement information based on the reimbursement information and the reimbursement information, And a prepayment re-administration module including a reimbursement weight calculation unit,
The credit rating calculator calculates,
Re-calculating the credit rating based on the repayment weight,
The pre-payment management module includes:
Further comprising a blind execution unit for selectively blind processing the goods not to be pre-paid to the user among the goods to be prepaid according to the re-calculated high and low scores of the credit rating. Based payment prepayment system. 10. The method of claim 9,
The pre-payment management module includes:
By checking the advance payment period of the goods that have been pre-paid to the user,
A later input part for inputting a later sentence from the user,
A valid keyword extracting part for extracting a valid keyword through grouping factors grouped according to the frequency of occurrence and proximity of the keywords included in the later sentence,
Further comprising a late search performing unit for calculating additional weights based on the frequency of occurrence, the proximity, and the number of effective keywords,
The reimbursement weight calculation unit may calculate,
Wherein the system is configured to calculate the reimbursement weights by reflecting the additional weights, wherein the reimbursement weights are calculated by using the multi-media big data matching pattern analysis. 11. The method of claim 10,
The additional weight may be expressed as:
Is calculated through Equation (4)
Equation 4.

(here,

Is an additional weight,

The

The occurrence frequency of the effective keyword at the time of the first pre-

The

The proximity of the effective keyword at the first pre-

The number of the effective keywords, the number of the effective keywords,

The total number of prepayments)
The re-
The system comprising: a system for providing a prepayment automatic payment service based on a credit evaluation based on analysis of a large-volume big data matching pattern.
Equation (5)

(here

Is the repayment weight,

Is the total number of times the user repaid the amount of the good that was pre-paid,

The

The reimbursement of the users'

The

The reciprocal of the period required for repayment, minus the advance payment of the goods from the repayment date on which the user repays the amount,

Is the number of times the user repaid the amount at the i-th advance payment,

Lt; RTI ID = 0.0 > weight) 11. The method of claim 10,
Wherein the late inspection performing unit comprises:
An interval grasp part for arranging the keywords included in the later sentence in a line and grasping the interval of each of the keywords,
An extended keyword derivation part for deriving an extended effective keyword through an extended grouping factor obtained by further applying an interval weight according to the interval of the intervals,
A goods propensity grasp part for grasping the goods to be pre-paid by the user on the basis of the extended valid keyword and generating goods propensity information,
The reimbursement weight calculation unit may calculate,
Wherein the system further calculates the reimbursement weights by further reflecting the correction weights calculated through the extended grouping factor. The system for providing reimbursement automatic payment services after a prepayment reservation based on credit evaluation using a multi-media big data matching pattern analysis. 13. The method of claim 12,
Wherein the extended grouping factor comprises:
Is calculated by the following equation (6)
≪ / RTI >

(here,

Is an extended grouping factor,

The total number of grouping,

The

Keywords in the ith grouping

Keywords for

Lt; / RTI >

silver

Keywords in the ith grouping

And Keywords

Spacing,

Keywords

And Keywords

Lt; / RTI >
The re-
The system comprising: means for calculating a credit amount based on a credit amount of the credit card;
Equation (7)

(here

Is the repayment weight,

Is the total number of times the user repaid the amount of the good that was pre-paid,

The

The reimbursement of the users'

The

The reciprocal of the period required for repayment, minus the advance payment of the goods from the repayment date on which the user repays the amount,

The

The number of times the user repaid the amount at the first pre-

Lt; RTI ID = 0.0 > weight,

≪ / RTI > 13. The method of claim 12,
Wherein the blind executing unit comprises:
A release criterion setting part for receiving the reference reimbursement amount and the reference reimbursement goodness degree from the manager of the system and generating the release criterion information to set the release criterion,
The prepayment re-managing module receives any one of the redeem information and the redeemable information of the user to generate comparison information, compares the comparison information with the release criterion information, Further comprising a blind cancel part for releasing the blind processed in said blind good if said amount exceeds said amount or said redeemable good exceeds said reference redeemable goodness rating, Based payment prepayment system.

Images