KR20220121422A - Customer-tailored store linkage discount rate and store recommendation system and method - Google Patents

Abstract

A customized store-linked discount rate and store recommendation system according to one embodiment of the present invention includes a store recommendation application which is executed in a customer terminal possessed by a customer to provide a discount service linked between stores and receives pre-registered discount linked stores such that through a management server linked through a communication network, the discount stores can be searched. The store recommendation application provides a discount service linked between stores through the management server such that a customer can receive a discount on store products when using a linked store linked to a specific store. The store recommendation application is configured to collect statistical information from the customer terminal, provide and recommend a linked discount store suitable for the customer based on the statistical information using an artificial intelligence recommendation algorithm, optimize a linked discount rate of the linked discount store by using big data based on the statistical information, and provide the optimized linked discount rate. The statistical information used when recommending the linked discount store includes at least one among customer information and linked discount provision details, movement information according to the movement of a customer location, a discount rate linked between stores, information on linked discount stores near the corresponding location, and customer information. According to the present invention, both customers and stores can maximize their profits through linked discounts.

Description

{Customer-tailored store linkage discount rate and store recommendation system and method}

The present invention relates to a customer-customized store-linked discount rate and store recommendation system and method, and more particularly, to a system and method for recommending a store-linked discount rate and store optimized to a customer using AI.

Local currency that is being used recently is a virtual currency that people make with their own hands and used only in a certain area. It is common to issue points in place of legal currency for the purpose of revitalizing the local economy and community through mutual aid in the form of mutual aid between members or volunteer work. It has various names such as Green Money and Songpa Money to suit the characteristics of the local community and is traded between members.

However, although most local currency transactions are used in the region, various discount links are not provided, so there is a need to provide a linked discount service that can activate transactions between local stores.

On the other hand, machine learning is a method of learning a computer using data. Machine learning can be divided into supervised learning, unsupervised learning, and reinforcement learning. Among them, supervised learning is a method of training a computer in a state where labels (exact correct answers) for data are given. In general, the performance of a machine learning model improves as it is trained with a large amount of data. Meanwhile, a convolutional neural network (CNN), one of the machine learning models, is showing excellent performance in the field of image detection. Since CNNs have hundreds of thousands of parameters, they must be trained with a sufficient number of training images.

Therefore, linkage between local currency and stores is used to activate local community transactions, but in order to improve the efficiency and economic feasibility of discount linked to local stores, an optimized linked discount rate and optimized linkage discount rate by using big data-based machine learning There is a need to develop a customer-tailored linked discount store recommendation system that recommends stores to customers.

Republic of Korea Patent Publication No. 10-2019-0143565 (published on December 31, 2019) Republic of Korea Patent Publication No. 10-2019-0142656 (published on December 27, 2019)

An object of the present invention is to utilize the linkage between local currency and stores, but to provide information on the optimized linkage discount rate to which the linkage discount is applied by an artificial intelligence algorithm based on the collected data, and through this, the optimized linkage discount It is to provide a customer-customized store-linked discount rate and a store recommendation system and method for recommending a store to a customer.

A customer-customized store-linked discount rate and store recommendation system according to an embodiment of the present invention is executed in a customer terminal owned by a customer to provide a store-linked discount service, and is pre-registered through a management server linked through a communication network. It includes a store recommendation app that is provided to be able to inquire a linked discount store, and the store recommendation app is a link discount between stores so that, through the management server, a customer can receive a discount on a store product when using a linked store linked to a specific store. The service is provided, but statistical information is collected from the customer terminal, and an artificial intelligence recommendation algorithm is used to recommend a linked discount store suitable for the customer based on the statistical information, and big data based on the statistical information Recommends linked discount stores that provide optimized linked discount rates by optimizing the linked discount rates of linked discount stores using It is characterized in that at least one of the movement information, the discount rate linked between stores, information on the linked discount stores around the corresponding location, and customer information is included.

When a customer visits a store, the store recommendation app is a store recommendation app installed on the customer's terminal when the clerk completes customer verification by looking up a preset QR code or phone number for the store product. A discount coupon is automatically issued, and a list of at least one linked discount shop recommended by an artificial intelligence algorithm is provided to the discount coupon of the linked discount store.

When the store recommendation app comes close to the location (coordinate) where the customer terminal is pre-registered and the set linked discount store is located, the linked discount shop discount coupon is automatically issued, and the augmented reality/virtual reality based map in the store recommendation app It is characterized in that the map or GUI of the linked discount store is implemented and displayed, and when the customer terminal is near the location (coordinate) where the linked discount store is located, the store discount coupon is issued to the customer terminal with a pop-up notification.

The management server receives customer information to receive the inter-store discount service from the customer terminal and registers, or registers the store to provide the inter-store discount service to the customer in order to register store products in the store terminal. register; a linked discount unit that applies the inter-store discount rate set for the product or service when the payment approval unit requests a payment; a payment approval unit that receives a barcode or QR code of local currency registered in the customer terminal and performs payment approval according to whether the code is approved; After collecting statistical information on customer information and linked discount offer history, it can be provided to recommend linked discount stores to the customer using an artificial intelligence recommendation algorithm. Big data and artificial intelligence algorithm based on statistical information a recommendation unit that optimizes the linked discount rate of the linked discount store using It further includes a security management unit that distributes, stores, and manages information on customer information, store products, payment, and linked discount rates in various systems based on a block chain using a block chain server.

When the recommendation unit recommends linked discount stores that optimize the linked discount rate of the linked discount store using big data and artificial intelligence algorithms based on statistical information, a difference occurs in the customer visit rate depending on the store discount rate, and the discount rate is The higher the value, the lower the profit of the store, so a reference value for calculating the linked discount rate optimized to find an appropriate linked discount rate between linked discount stores is set, and the artificial intelligence algorithm is a recursive neural network (RNN), It is characterized in that it includes at least one of a Convolutional Neural Network (CNN) neural network and a GRU-CNN Hybrid neural network algorithm.

According to an embodiment of the present invention, a customer-customized store-linked discount rate and store recommendation method includes the steps of: inquiring a linked discount product or service in conjunction with a management server with a store recommendation app executed on a customer terminal possessed by the customer; providing, by the management server, in the form of a priority list according to the linked discount rate optimized for the linked discount product (store) inquired; requesting, by the customer terminal, an order for a linked discount product or service searched by a code registered in the store recommendation app; The management server receiving the order request from the store recommendation app includes the step of approving the payment by approving the code for the product or service, and applying a discount rate according to the set linked discount discount rate, wherein the store recommendation app is the management server Through this, when a customer uses a store linked to a specific store, a store-to-store discount service is provided to receive a discount on store products, and statistical information is collected from the customer terminal and statistical information using an artificial intelligence recommendation algorithm. Linked discount stores that provide optimized linked discount rates by optimizing the linked discount rates of linked discount stores by using big data based on statistical information Recommendation, and statistical information used when recommending the linked discount store includes at least one of customer information and linked discount provision history, movement information according to customer location movement, linked discount rate between stores, linked discount store information around the location, and customer information characterized in that it is included.

The management server can be set to provide an additional discount rate when the customer revisits within a certain period through the store terminal, and accumulates a certain percentage each time the store or a discount store linked to the store is revisited to provide additional overlapping discounts and, the additional discount rate is set to vary for each user as a result of analyzing previously provided customer information and statistical information using an artificial intelligence algorithm.

The customer-customized store-linked discount rate and store recommendation method may be performed by a computer program stored in a computer-readable storage medium.

The present invention can activate inter-store transactions through inter-store linkage discount setting, and induce continuous re-visit of the linked store by presenting an optimized discount rate to customers visiting the linked store using local currency. As it can increase sales, it has the effect of increasing sales for local stores and motivating customers to use the service.

In addition, after checking the list provided on the customer terminal, you can receive a recommendation for a store that you need to use or want to use to receive a linked discount, so you can easily find linked discount stores. Anyone can easily use the linked discount stores in the store, so customers and the store can establish a win-win strategy with each other and ultimately contribute greatly to the revitalization of the local economy.

In addition, by recommending linked discount stores that provide optimized linked discount rates based on big data and artificial intelligence algorithms for collected statistical data, both customers and stores can maximize profits through linked discounts. There are advantages that can be

In addition, when a customer touches on the screen, it is possible to list and display linked discount stores linked to the store, and by displaying the moving routes of the stores used by the customer, it is possible to guide the customer where to select the next destination, through this From the customer's point of view, it motivates them to use other discount stores, and the tourism factor of moving to various places (linked discount stores along the route) is added to generate interest, and from the store's point of view, it is possible to attract more customers. It can act as an incentive.

In addition, the discount rate can be increased by applying the linked discount between stores through the linked discount coupon, and there is an advantage that can contribute to the vitalization of the local community.

In addition, there is an advantage in that the transaction details for payment are stored based on the block chain, so that a safe transaction is made.

1 is a block diagram showing the overall configuration of a customer-customized store-linked discount rate and store recommendation system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a detailed configuration of the management server of FIG. 1 .
3 is a flowchart illustrating a process in which a customer is provided with an optimized linked discount service in a customer-customized store-linked discount rate and store recommendation method according to an embodiment of the present invention.
4 is a view for explaining a process in which a customer uses a store that can be linked to a discount.
5 is a view showing an exemplary customer search screen of the store recommendation app (110).
6 is a view showing the store management screen of the store recommendation app 110 by way of example.
7 is a diagram illustrating an event setting screen of the store recommendation app 110 by way of example.
8 is a diagram for explaining the concept of a discount link between stores.
9 is a diagram illustrating an example of setting a discount link between stores using geofencing.
10 is a diagram illustrating an example of a citizenship card and a discount coupon issued to a customer to use a linked discount store.
11 is a diagram for explaining a hybrid convolutional neural network (CNN)-gate recurrent units (GRU) neural network algorithm, which is a hybrid recommendation algorithm.
12 is a diagram illustrating a process for an optimized linked discount rate and advertisement recommendation using a complex recommendation algorithm.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings. However, the spirit of the present invention is not limited to the presented embodiment, and those skilled in the art who understand the spirit of the present invention may add, change, delete, etc. other elements within the scope of the same spirit, through addition, change, deletion, etc. Other embodiments included within the scope of the invention may be easily suggested, but this will also be included within the scope of the invention. In addition, components having the same function within the scope of the same idea shown in the drawings of each embodiment will be described using the same reference numerals.

1 is a block diagram showing the overall configuration of a customer-customized store-linked discount rate and store recommendation system according to an embodiment of the present invention.

The customer-customized store-linked discount rate and store recommendation system of the present invention is connected to the store terminal 200 or the customer terminal 100 through the communication network 400 as shown in FIG. 1 to provide a store-linked discount and store recommendation service It includes a management server 300 to provide.

Here, the communication network 400 may include a mobile communication network 400 such as LTE, 3G, etc., as well as an Internet network, a private network, etc., and the known detailed configuration required to build the communication network 400 or link between the communication networks 400 is provided. more may be included.

The management server 300 is connected from a store terminal 200 located in a store to register a linked discount store and a linked discount target product, and a customer terminal possessed by a customer to provide a shop-to-shop link discount and link discount store recommendation service. Linked with the store recommendation app 110 executed in (100), and provided so that the linked discount store registered through the shop recommendation app 110 executed on the customer terminal 100 can be inquired, and linked discount product order or service When making a reservation, payment for a product order or service is performed, and payment approval information is transmitted to the store terminal 200 in which the product or service is registered.

In addition, the management server 300 may provide a revenue sharing process capable of distributing revenue between the linked discount stores when linking between stores.

Through this, for example, store A shares (distributes) the advertising revenue for store B, and store A uses a store terminal (smartphone, PC, POS, etc.) , period, region, and customer can also be classified to set advertisements.

In addition, the management server 300 may maximize the mutual revenue by receiving the income obtained from the fee dimension or advertising costs obtained from the linked discount store and reflecting a certain ratio in the application of the linked discount rate.

In addition, the store recommendation app 110 of the present invention is an application that is a kind of computer program installed and executed on the store terminal 200 or the customer terminal 100 possessed by the customer for receiving a discount service linked between stores, and a management server It may be provided through 300 , and various linked discount services may be provided in connection with the management server 300 .

In particular, the store recommendation app 110, through the management server 300, if the customer uses a store linked to a specific store, but provides a discount service between stores to receive a discount on the store product, the customer terminal (100) It collects statistical information from customers and uses an artificial intelligence recommendation algorithm to provide a recommendation for a discount store that suits the customer based on the statistical information, and links the discount store by using big data based on the statistical information. By optimizing the discount rate, it is possible to recommend a linked discount store that provides an optimized linked discount rate.

Furthermore, statistical information used when recommending linked discount stores may include at least one of customer information and linked discount offer details, movement information according to customer location movement, link discount rate between stores, information on linked discount stores near the location, and customer information. have.

The store recommendation app 110 is 4 and 10, when the customer with the customer terminal 100 visits the store, the clerk completes customer confirmation by inquiring a preset QR code or phone number for the store product. The store recommendation app installed in the customer terminal can provide a link discount store discount coupon according to the use of the link discount store product in connection with the management server so that the discount coupon is automatically issued.

In addition, when issuing a discount coupon, a list of at least one linked discount shop recommended by an artificial intelligence algorithm may be provided to the linked discount store discount coupon.

For example, the list of linked discount stores can be provided in the form of priority based on the order of movement information collected by statistical information and customer preference (included in customer information). In order to receive a linked discount, you can receive a recommendation for a store that you need to use or want to use, so you can easily find a linked discount store. The Lee Rang store can be a win-win for each other and can ultimately contribute greatly to the revitalization of the local economy.

Furthermore, the store recommendation app 110 implements and displays an augmented reality/virtual reality-based 3D/2D map map or a linked discount store in the form of a GUI (Graphical User Interface), and the customer terminal 100 is the location where the linked discount store is located. (Coordinates) When close to the vicinity, a store discount coupon may be issued along with a pop-up notification to the customer terminal 100 . The proximity of the location can be determined by receiving the location coordinates using GPS location information or GIS-based geographic information, and determining whether it is a preset radius (eg, within 1km, etc.) It can also be used for understanding.

In addition, at this time, the augmented reality/virtual reality-based 3D/2D map map may be provided through an agreement with an existing map map provided from portal sites or apps such as Daum and Naver, and modified or edited, or provided as it is.

In addition, the graphical user interface (GUI) of the linked discount store is displayed on the map map in the form of a shop-shaped icon with reference to FIGS. 8 and 9, and when the customer touches a specific shop on the screen, the linked discount shop is linked with the store can be listed and displayed, and the moving routes of the stores used by the customer can be displayed on the map map to guide the customer where to select the next destination (linked discount store to use).

Through this, it will motivate customers to use other discount stores, add a tourism element to move to various places (linked discount stores along the route), and generate interest, and will attract more customers from the store's point of view. It can act as an incentive to make it happen.

In addition, the management server 300 applies and pays according to a discount rate linked between stores when paying for a product order or service, and uses local currency as a payment method.

By downloading, installing, and executing the store recommendation app 110 from the management server 300 in the customer terminal 100 possessed by the customer, the store-linked discount service provided by the management server 300 can be provided.

In addition, in order to receive the discount service between stores in the customer terminal 100, customer information must be provided and registered, and the local currency for payment of the discount product or service in the store recommendation app 110 is registered to provide a payment service. can receive

In addition, at this time, each customer collects statistical information customized to each customer and receives citizenship, a type of membership, so that he/she can use the linked discount service through citizenship. In addition, as an example of citizenship, it is provided in the form shown in FIG. 10, and may be issued in the form of a card, a mobile card, or the like.

In addition, the type of local currency to be used for payment service can be divided into card-type, paper-type, and mobile-type local currency issued within the region. It is possible to register in the store recommendation app 110 for payment service for a linked discount using a barcode or QR code in the form of a code, or by using a card number in the case of a card type.

In addition, the management server 300 may issue a linked discount coupon that can be reused in the form of a discount coupon after payment approval through the store recommendation app 110 .

FIG. 2 is a block diagram showing a detailed configuration of the management server 300 of FIG. 1 .

As shown in FIG. 2, the management server 300 includes a customer registration unit 310, a linked discount unit 320, a payment approval unit 330, a recommendation unit 340, a security management unit 350, and a service DB ( 360), a model DB 361, and a statistics DB 362 are further included.

The customer registration unit 310 receives and registers customer information in order to receive the inter-store discount service from the customer terminal 100 or provides the inter-store discount service to the customer in order to register the store product at the store terminal 200 . Register the store to

The linked discount unit 320 applies the inter-store discount rate set for the corresponding product or service when the payment approval unit 330 requests a payment.

In addition, the linked discount unit 320 applies the linked discount rate in the case of a linked discount store according to the basic discount rate and the linked discount rate provided from the store terminal 200, and applies the basic discount rate when it is not a linked discount store to make payment for the product We provide a linked discount service so that you can do this.

Here, the linked discount is an inter-store discount, which means that when a product of a pre-registered linked discount store is used, a linked discount service is provided according to a set linked discount rate. Here, the pre-registered linked discount store may be a affiliated store registered in the operation server without a separate linked discount store registration process.

For example, as shown in FIG. 8 , if a customer (user) visits an optician after visiting an aquarium and then orders a menu at a pub, the linked discount is sequentially applied according to the pre-registered linkage discount registration between stores. In this case, even if only two stores are used, the linked discount is made by the linked discount between the two stores, and additional linked discount can be applied to additional visits after that. And for the store, it can attract many customers, so a win-win strategy can be achieved.

In addition, the linked discount rate may be applied singly or in duplicate depending on at least one of the number of times of use between linked discount stores, the distance between stores, franchises, affiliate stores, trade name designation between stores and public facilities linkage settings, and family store settings.

The distance link between stores may be set to automatically link when located within a specific distance from a specific store, and the store designation link may be set to link by mutual designation between affiliate stores regardless of distance.

In addition, the public facility linkage setting may be set to be mutually linked if it is a public facility regardless of the distance.

In addition, referring to Figures 8 and 9 using geofencing technology, automatic linkage setting is automatically established between stores within a set specific distance, and designated linkage in which a store selects and links a desired store regardless of the distance. It is also possible to provide an allow linkage setting function to set whether to allow the linkage, which allows other stores to select and link their own stores.

In addition, in the case of a store registered as a family, when the store sets a family discount, the customer can receive a family discount in addition to the linked discount for the product. That is, in the case of a family store, a total of three discount benefits (basic discount, linked discount, family discount) can be received depending on whether the store provides a discount. can receive various discount benefits, and stores can maximize profit creation, so there is a win-win advantage.

The family store may be a store affiliate (franchise) or a store registered as a family store in advance for mutual sponsorship between stores. In addition, when a group preferred by regions, age groups, students, housewives, office workers, families, etc. is created, and related stores are registered in the group, the stores may be referred to as family stores. For example, in the case of a "business associate's dinner party" group, if stores such as restaurants, cafes, karaoke rooms, and billiard rooms belong to the group, it becomes a family store.

In this case, the linked discount unit 320 provides a function to guide the stores to be searched for the linked discount, and referring to FIGS. 6 and 7 , when requested by the customer terminal 100, the linked discount store search and A linked discount product search service is provided, and recommended shops and product services are provided upon request from the customer terminal 100 , or various discount rates may be directly set and provided in the store terminal 200 according to event settings.

As shown in FIG. 5 , the payment approval unit 330 receives the barcode or QR code of local currency registered in the customer terminal 100 and approves the payment according to whether the corresponding QR code is approved.

In addition, payment may be approved using the card number of the card-type local currency of the customer terminal 100 or a registered mobile phone number.

In the case of using a mobile phone number, it is the case of a phone delivery order, and a mobile phone number registered with the customer terminal 100 is requested to be input, and compared with the mobile phone number included in the customer information stored in the DB of the management server 300 . By performing authentication, this can be the way the order is placed.

The recommendation unit 340 may use an artificial intelligence recommendation algorithm provided after collecting statistical information on customer information and linkage discount provision history, etc. Based on big data, it is possible to provide to optimize the linked discount rate of linked discount stores, or to recommend linked discount stores to use next according to the optimized linked discount rate.

Here, the recommendation algorithm may be, for example, a Hybrid GRU-CNN artificial intelligence algorithm, an artificial intelligence algorithm using similarity analysis, and the like.

Specifically, in the case of recommending linked discount stores that optimize the linked discount rate of linked discount stores by using big data based on statistical information, for example, when customer discount rates are prioritized, between stores with the highest linked discount rate. However, since there is a difference in the customer visit rate depending on the store discount rate, the higher the discount rate, the more customers, but if the discount rate is higher, the profit of the store will decrease. It may be a recommendation, and a reference value for calculating an optimized linked discount rate may be set.

For example, set the discount rate determining parameters, and the parameters include commercial district characteristics (school, work area, etc.), customers (gender, age group, recommendation/participation/visit rate, tendency, interest, hobbies, shops used, hours of use, etc.), With the industry (restaurant, cafe, etc.) as a variable, the discount rate should vary according to these variables, and the optimal linkage that maximizes store profits or customer profits using artificial intelligence algorithms based on big data collected statistical information By calculating the discount rate, it is possible to recommend linked discount stores.

In addition, the discount of the store service in the conventional system is fixed to one service, and if it is a 10% discount per store (product), a 10% discount is provided to all customers, respectively, but the linked discount service provided by the present invention is It is not fixed with one service, but uses artificial intelligence so that the discount rate varies depending on the customer.

Specifically, the artificial intelligence algorithm uses customer information and statistical information to analyze customer usage rate, number of store visits, type of store visited, industry type, similarity between industries, weather, commercial area, day of the week, season, gender, age, etc. A discount rate can be recommended.

In particular, there is an advantage of receiving an additional discount if the linked stores are visited within a certain period (eg, 7 days) by the setting of the store terminal. For example, a discount of 0% on the first visit, a discount of 2% on the second visit, and a 3% discount on the third visit are provided so that customers can receive a double discount. That is, it can be calculated as 5% (5+0) -> 7% (5+2) -> 10% (5+2+3). As the result of analysis using an intelligent algorithm, it is different for each user.

In addition, an additional discount rate is applied not only when visiting the same store but also when visiting a linked store. As a result, the artificial intelligence algorithm optimizes the linked discount rate to provide a higher discount rate so that customers continue to revisit the linked store. The store terminal may set the maximum discount rate provided to the customer so that the optimized linked discount rate by artificial intelligence is calculated within the range that does not exceed the maximum discount rate.

The Hybrid GRU-CNN artificial intelligence algorithm used for this purpose is a Hybrid GRU-CNN neural network model for maximizing the above-mentioned discount rate determination parameter data. can be used to

Sequence data loading for better machine learning learning potential You can use spatiotemporal matrix and spatiotemporal matrix mapping feature vectors using time-ordered data (statistics provided by statistics DB) and CNN module.

In addition, the Hybrid GRU-CNN neural network can be an ideal Hybrid GRU-CNN neural network for processing sequence data, which is the advantage of the GRU module that can handle time well, and high-dimensional data processing, which is the advantage of the CNN module, to combine the two neural networks. The Hybrid GRU-CNN neural network is shown in FIG. 11 .

The framework neural network of Hybrid GRU-CNN consists of a GRU module and a CNN module, and the input is temporal sequence information, and the collected discount rate determining parameter or advertisement recommendation decision parameter data and the output of the spatiotemporal matrix system are predictions for future loads. .

Looking in detail with reference to FIGS. 11 and 12 , the CNN module is good at processing in terms of the CNN module, and the CNN module of two-dimensional data images, such as a spatio-temporal matrix, is a matrix for weighting data to determine a linkage discount rate to extract local features. and effective decision values can be obtained, and the composite weight data may include customer preference weights, recommendation suitable weights, etc. shown in FIG. 12 .

Here, the structure of the CNN module includes two convolutions through a convolutional layer and a pooling layer. In this case, it may include a layer and merge operation, a convolution operation of each convolution layer, and a pooling operation.

In addition, after the second pooling operation, the high-dimensional data is flattened into one-dimensional data, and the output of the CNN module is a fully connected layer.

The input of the GRU module is time-ordered data, and the GRU module contains many gate repeat units, all of which form a fully connected layer. Finally, the prediction result of the neural network for calculating the optimized linkage discount rate can be obtained by calculating the average value of all neurons.

In addition, such a Hybrid GRU-CNN neural network can be the hybrid recommendation algorithm of FIG. 12, and it collects store information, customer information, commercial area information, and environmental information (commerce area, geographic information, etc.) -Use as an input variable for the CNN neural network, calculate the optimized linked discount rate, which is a decision value obtained by calculating the average value according to customer preference weights and recommendation suitable weights (business district characteristics, industry characteristics, etc.), and link based on the calculated linked discount rate It allows us to recommend the best suitable store.

Furthermore, if there is no initial statistical information, which is an input variable of the algorithm when recommending advertisements in the optimized linked discount store or store, the link provided to the customer with the most similar condition by using a cold start to search for conditions similar to the customer A service may be provided to recommend discount stores.

Recommendation conditions are performed based on customer information, and the customer information collected for this purpose includes commercial district characteristics (school, near work, etc.), customer characteristics (age, disposition, interest, hobbies, shops used, hours of use, etc.), industry characteristics (restaurants, cafes, etc.) Here, the virtual similar condition may be a condition in which the number of matching or similar information of the above-described collection information is the largest.

In addition, the recommendation algorithm method applied to similar conditions can utilize an artificial intelligence algorithm using similarity analysis. In addition to the similarity calculation method, the dimension of the interaction matrix is reduced and approximated by two or more small matrices with k potential components You can also use the matrix factorization method that can predict the rank of each user item by multiplying the corresponding row and column.

In addition, it explores the state space of a set of possible frequent items by using an association rule or a mining rule in which frequently consumed items are connected to the edge of the graph, and removes branching in the infrequent search space, A set of frequently used items is used to create rules, and these rules can also generate recommendations (recommendations).

For example, the management server 300 initially collects customer information in order to provide a recommended linked discount store recommendation service to the customer terminal 100, and based on the collected customer information, selects the most suitable linked discount store for the age group. It is also possible to recommend or create a related rule in consideration of age, gender, hobbies, interests, etc. included in customer information, and recommend linked discount stores that match it.

Furthermore, the recommendation unit 340 provides a link discount service between stores so that a customer can receive a discount on a store product when a customer uses a store linked to a specific store, and a store recommendation app executed on the customer's smartphone when inquiring about a linked discount store (110) provides a service for the affiliated store to display its advertisement in a specific store so that the customer visit rate increases when exposed at the top of the screen, and after providing the advertisement, the advertisement revenue for the affiliated store is distributed to the specific store , The specific store and the affiliated store may further include a function of providing advertisement settings by categorizing advertisement content, advertisement amount, period, region, and customer through the store recommendation app 110 executed on the store terminal. have.

In addition, like the optimized linked discount rate recommendation, it will recommend advertisements for linked discount stores related to the store based on statistical information on the store usage history of customers and linked discount stores collected as big data using artificial intelligence algorithms. Similarly to the above-described discount rate determining parameter, an advertisement recommendation parameter may also be set.

The security management unit 350 can store and manage information about customer information, store products, payment, linked discount rate, etc. from the management server 300 by using the block chain server 500 to distribute and store it in several systems based on a block chain. .

The block chain server 500 is composed of a plurality of nodes, and among them, the payment information of the customer, such as local currency transaction details, is distributed to a plurality of nodes and stored in the form of blocks. Whenever the corresponding details are generated by payment, they are added in the form of blocks, and the blocks generated later contain the hash value of the previously generated blocks.

The user's local currency generation and transaction details stored in the block chain server 500 and the user's local currency payment transaction history and advertising revenue distribution stored in the DBMS are periodically or in some event (user's local currency generation history, transaction If there is a history or a new distribution of advertising revenue between stores), they are synchronized with each other.

In addition, information on customer information, stores, products, and linked discount rates can also be stored and managed in a distributed manner through the blockchain network, and security is strengthened to ensure safe payment and information protection.

Furthermore, information distributed and managed through the blockchain server 500 may be stored by a merkle tree structure.

For example, the transaction details for each payment are stored including the hash value in the lowest child node, and the hash value is applied to the intermediate node on the path leading to the lowest child node in the Merkle root (parent node), which is the highest level of the Merkle tree. It is hashed and stored for sharing.

Through this, when determining the authenticity of the stored transaction details, the transaction details included in the payment information copied to the customer's customer terminal 100 and the transaction details stored in the block chain server 500 are compared, and the path of the Merkle tree This is done by comparing only the hashed hash values.

At this time, since it is not necessary to perform comparison operations on blocks of all nodes as comparison operations are performed on the path of the Merkle tree, authenticity can be determined with a relatively easy amount of computation, and forgery of transactions can be found easily and quickly. , it is possible to easily verify the transaction even in the customer terminal 100 with a small capacity.

More specifically, the management server 300 performs a lightweight encryption algorithm that provides identification information, each capable of identification, to each terminal, and utilizes the identification information of each terminal as a private key. do.

The lightweight encryption algorithm is an encryption technology designed to be implemented in a limited environment such as a smart device such as a customer terminal 100 and a store terminal 200 utilized in the present invention, and a symmetric key encryption algorithm HIGHT (HIGh security and light weigHT) , LEA (Lightweight Encryption) and LSH (Lightweight Secure Hash), a hash function, can be used.

By encrypting/decrypting the transmitted and received information using such a lightweight encryption algorithm, it is possible to prevent illegal control due to external leakage or external hacking of the corresponding data.

In addition, the management server 300 performs the above-described artificial intelligence algorithm with the service DB 360 for storing customer information, store product information, discount information, etc. for providing services such as linked discount store recommendation and payment. It further includes a model DB 361 for storing and managing a plurality of predictive models, revenue models, and recommended models, and a statistical DB 362 for providing customer usage details (store usage details, payment details, etc.).

As the model DB 361 stores and manages a plurality of prediction models, revenue models, and recommendation models for performing artificial intelligence algorithms, the algorithms used according to the learning contents such as linked discount store recommendation, advertisement recommendation, and customer demand prediction may vary, and multiple algorithms may be used.

In addition to the aforementioned Hybrid GRU-CNN, the machine learning-based artificial intelligence neural network learning algorithm used here may further include a Recursive Neural Network (RNN), a support vector machine (SVM), or a Convolutional Neural Network (CNN) neural network if necessary. can

3 is a flowchart illustrating a process in which a customer is provided with an optimized linked discount service in a customer-customized store-linked discount rate and store recommendation method according to an embodiment of the present invention.

First, a linked discount product (store) or service provided by the management server 300 is inquired with the store recommendation app 110 executed on the customer terminal 100 possessed by the customer (S102).

You can request an order for a linked discount product or service searched by the QR code of the local currency registered in the store recommendation app 110. The linked discount product (store) inquired is in the form of a priority list according to the optimized linked discount rate. may be provided (S104, S106).

Here, the optimized linkage discount rate can be calculated based on the artificial intelligence algorithm as described above, and the linkage discount rate is the linkage network (number of use between stores, distance between stores, franchise, affiliate stores, and store name designation between stores) and public facility linkage setting and family store setting), so it is possible to derive an optimized linkage discount rate by considering these specific factors between stores, and the more data, the more optimized linkage The amount of computation required to calculate the discount rate may increase.

In addition, at this time, in the case of a phone delivery order, by requesting to input a mobile phone number registered to the customer terminal 100, and performing authentication by comparing it with the mobile phone number included in the customer information stored in the DB of the management server 300, Orders may be placed.

After that, the management server 300 receiving the order request from the store recommendation app 110 approves the QR code for the product or service, and applies the discount rate according to the set linked discount rate, so that payment is approved (S108).

A payment approval completion notification is transmitted from the payment approval management server 300 to the store terminal 200 and the order-requested customer terminal 100 .

Also, when payment is approved, a coupon-type linked discount coupon set for the linked discount product or service is issued, and is provided through the store recommendation app 110 (S110).

Through the linked discount coupon, the discount rate can be increased by applying the linked discount between stores repeatedly, and it has the advantage of contributing to the revitalization of the local community.

In addition, the payment history and store usage history as statistical information are stored in the statistics DB 362 of the management server and can be utilized when applying a big data-based artificial intelligence algorithm (S112).

Furthermore, before providing the payment service, the customer terminal 100 receives and registers customer information in order to receive the inter-store discount service, and the store recommendation app 110 uses local currency for payment of discount products or services. The process of registering may be preceded.

In addition, the method of operating a customer-customized store-linked discount rate and store recommendation method according to an embodiment of the present invention may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, Flash memory, Solid State Drive (SSD), and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

In addition, the configuration such as a computer or computer program used in the present invention is transformed like a smartphone in the form of a mobile communication terminal, and as computing power increases dramatically, it is also used in the same meaning as a smartphone or an application executed on a smartphone. can be used

100 ; customer terminal
110 ; Store recommendation app
200 ; store terminal
300 ; management server
310; customer register
320 ; Linked discount department
330; Payment Approval Department
340; recommendation
350 ; Security Management Department
360 ; service DB
361 ; model DB
362 ; Statistics DB
400 ; communications network
500 ; blockchain server

Claims (10)

In order to provide a discount service linked between stores, it is executed on the customer terminal possessed by the customer, and includes a store recommendation app that is provided to search for linked discount stores registered in advance through a management server that is linked through a communication network,
The store recommendation app is
Through the management server, when a customer uses a store linked to a specific store, a discount service linked between stores is provided so that a discount for store products is provided,
It collects statistical information from the customer terminal, uses an artificial intelligence recommendation algorithm, provides a link discount store suitable for the customer based on the statistical information, and provides a link discount using big data based on statistical information. By optimizing the store's linked discount rate, we recommend linked discount stores that provide an optimized linked discount rate.
Statistical information used when recommending the linked discount store includes at least one of customer information, linked discount offer details, movement information according to the movement of customer location, linked discount rate between stores, and linked discount store information around the customer location. A customer-customized store-linked discount rate and store recommendation system. According to claim 1,
The store recommendation app is
When a customer visits a store, when the clerk completes customer verification by inquiring a preset QR code or phone number for the store product, a store discount coupon is automatically issued according to the use of the linked discount store product with the store recommendation app installed on the customer terminal. ,
A customer-customized store-linked discount rate and store recommendation system, characterized in that the linked discount store discount coupon provides a list of at least one linked discount store recommended by an artificial intelligence algorithm. 3. The method of claim 2,
The store recommendation app is
When the customer terminal comes close to the location (coordinate) where the pre-registered and set linked discount store is located, the linked discount store discount coupon is automatically issued,
In the store recommendation app, a map map based on augmented reality/virtual reality or a GUI of a linked discount store is implemented and displayed,
When the customer terminal is close to the location (coordinate) of the location (coordinates) where the discount store is located, the customer terminal is a customer-customized store-linked discount rate and store recommendation system, characterized in that issuing the store discount coupon together with a pop-up notification to the customer terminal. According to claim 1,
The management server
a customer register for receiving and registering customer information in order to receive the inter-store discount service from the customer terminal, or for registering the store to provide the inter-store discount service to the customer in order to register store products in the store terminal located in the store;
a linked discount unit that applies the inter-store discount rate set for the product or service when the payment approval unit requests a payment;
a payment approval unit that receives a barcode or QR code of local currency registered in the customer terminal and performs payment approval according to whether the code is approved;
After collecting the above statistical information, it can be provided to recommend linked discount stores to the customer using an artificial intelligence recommendation algorithm, and linkage of linked discount stores using big data and artificial intelligence algorithms based on statistical information a recommendation unit that optimizes the discount rate, recommends a linked discount store that provides an optimized linked discount rate, and recommends providing an advertisement suitable for the linked discount store;
A security management department that distributes, stores and manages customer information, store products, payment, and linked discount rate information in multiple systems based on blockchain using a blockchain server.
Customer-tailored store-linked discount rate and store recommendation system that further includes. 5. The method of claim 4,
The recommendation is
In the case of recommending linked discount stores that optimize the linked discount rate of linked discount stores by using big data and artificial intelligence algorithms based on statistical information, the customer visit rate differs depending on the store discount rate. Since the rate of return of the discount rate is lowered, a reference value is set for calculating the optimized linked discount rate to find an appropriate linked discount rate between linked discount stores.
The artificial intelligence algorithm includes at least one of a Recursive Neural Network (RNN), a support vector machine (SVM), a Convolutional Neural Network (CNN) neural network, and a GRU-CNN Hybrid neural network algorithm. Store recommendation system. Inquiring a linked discount product or service in conjunction with a management server with the store recommendation app executed on the customer terminal possessed by the customer;
providing, by the management server, in the form of a priority list according to the linked discount rate optimized for the linked discount product (store) inquired;
requesting, by the customer terminal, an order for a linked discount product or service searched by a code registered in the store recommendation app;
The management server receiving the order request from the store recommendation app, approving the code for the product or service, and applying a discount rate according to the set linked discount discount rate to approve payment,
The store recommendation app is
Through the management server, when a customer uses a store linked to a specific store, a discount service is provided between stores to receive a discount on store products,
It collects statistical information from the customer terminal, uses an artificial intelligence recommendation algorithm, provides a link discount store suitable for the customer based on the statistical information, and provides a link discount using big data based on statistical information. By optimizing the linked discount rate of the store, it recommends a linked discount store that provides an optimized linked discount rate,
Statistical information used when recommending the linked discount store includes at least one of customer information, linked discount offer details, movement information according to the movement of customer location, linked discount rate between stores, and linked discount store information around the customer location. Customer-tailored store-linked discount rate and store recommendation method. 7. The method of claim 6,
The store recommendation app is
When a customer visits a store, when the clerk completes customer verification by inquiring a preset QR code or phone number for the store product, a store discount coupon is automatically issued according to the use of the linked discount store product with the store recommendation app installed on the customer terminal. ,
A customer-customized store-linked discount rate and store recommendation method, characterized in that the linked discount store discount coupon provides a list of at least one linked discount store recommended by an artificial intelligence algorithm. 8. The method of claim 7,
The store recommendation app is
When the customer terminal comes close to the location (coordinate) where the pre-registered and set linked discount store is located, the linked discount store discount coupon is automatically issued,
Implementing and displaying an augmented reality/virtual reality-based map map or linked discount store GUI in the store recommendation app,
When the customer terminal is close to the location (coordinate) where the discount store is located, the store discount rate and store recommendation method according to claim 1, wherein the store discount coupon is issued to the customer terminal with a pop-up notification. 7. The method of claim 6,
The management server can be set to provide an additional discount rate when the customer revisits within a certain period through the store terminal,
Each time you revisit the store or the discount store linked to the store, a certain percentage is accumulated and additional discount is provided.
The additional discount rate is a result of analyzing previously provided customer information and statistical information using an artificial intelligence algorithm, and is set to be different for each user. A computer program stored in a computer-readable storage medium for performing the customer-customized store-linked discount rate and store recommendation method of any one of claims 6 to 9.

Images