Ai-based consumer activity data analysis and personalized reward and partnership benefit provision system
The AI-based consumer activity data analysis system addresses low refund rates and limited usability in existing reward systems by calculating consumer grades for differential payouts and integrating industry-wide partnerships, enhancing consumer convenience and loyalty through personalized recommendations and integrated service access.
Patent Information
- Application Number
- KR1020250113133
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2025-08-14
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2045-08-14
AI Technical Summary
Existing reward systems offer low refund rates, limited usability, and lack integration across industries, leading to restricted consumer participation and loyalty, with conventional services failing to provide intelligent content creation, personalized recommendations, and integrated user interfaces.
An AI-based consumer activity data analysis system that calculates consumer grades for differential reward payouts up to 90%, supports various benefit redemption methods, integrates industry-wide partnerships, and provides an integrated user interface for seamless partner search, consultation, and reward management.
Enhances consumer convenience and loyalty by offering high-refund rates, expands reward usage to multiple industries, and improves consumer engagement through personalized recommendations and integrated service access.
Smart Images

Figure 112025092892966-PAT00021_ABST
Abstract
Description
Technology Field
[0001] The present disclosure relates to an AI-based consumer activity data analysis and customized reward and partnership benefit provision system. Specifically, it relates to an AI-based consumer activity data analysis and customized reward and partnership benefit provision device that implements a high-reward refund structure linked to a wide range of industries based on consumer activity data and tier information, and supports partner search, consultation requests, community activities, and actual reward usage through an integrated user interface (UI / UX). Background Technology
[0002] Unless otherwise indicated in this specification, the contents described in this section are not prior art for the claims of this application, and are not to be recognized as prior art simply because they are included in this section.
[0003] Recently, consumers' real purchasing power has been continuously declining due to rising prices, economic recession, and the deepening of corporate-centric profit structures. Consequently, consumers are increasingly preferring benefits in the form of tangible refunds or returns over simple price discounts. In line with this trend, various reward platforms and point accumulation services have emerged; however, existing reward systems generally offer reward rates of only 5–10%, limiting the tangible benefits consumers can perceive. This makes it difficult to secure long-term consumer participation and loyalty. Furthermore, the use of rewards on existing platforms is often restricted to partner merchants or limited to specific products and services, making it difficult for consumers to freely utilize the benefits. Additionally, reward distribution is often limited to calculating a percentage of purchase amount or is heavily skewed toward advertising exposure and marketing activities, which limits the provision of customized benefits for consumers.
[0004] Furthermore, many existing reward apps are limited to B2C transaction environments and fail to provide refund structures linked to high-value B2B partnership industries such as finance, real estate, rentals, and legal and tax services. Consequently, the scalability of rewards and inter-industry interoperability are low. Moreover, existing consumer reward structures are designed around offline payments and simple point accumulation, lacking intelligent content creation and operation functions such as AI-based personalized recommendations, data analysis of consumer activities (mission completion, review writing, star ratings, etc.), and differential rewards based on consumer tiers. Additionally, conventional services lack integrated user interface (UI / UX) design, causing inconvenience for consumers who must switch between individual apps or websites to search for partners, request consultations, participate in communities, or check and withdraw accumulated points.
[0005] Therefore, a system and its method of operation are required that provide integrated UI / UX and comprehensive industry-wide partnership reward refunds utilizing consumer activity data, based on a consumer-centric, high-reward structure. Going beyond simple point accumulation, such a system can provide tangible economic benefits to consumers while offering partner companies effects such as reduced advertising costs, influx of new customers, and enhanced brand credibility. Prior art literature
[0006] 1. Korean Patent Publication No. 10-2025-0102899 (July 7, 2025) 2. Korean Patent Registration No. 10-2825381 (June 23, 2025) The problem to be solved
[0007] The problem that the technical concept of the present disclosure aims to solve is to provide a high-rate reward structure capable of refunding up to 90% depending on the consumer grade, based on the low refund rate of existing reward services.
[0008] Furthermore, the problem that the technical concept of the present disclosure aims to solve is to enhance the convenience of use for consumers by expanding rewards to various practical usage methods, such as cash withdrawal, gift certificates, coupons, and prize draws, rather than being limited to simple points or usable only at specific affiliates.
[0009] In addition, the problem that the technical concept of the present disclosure aims to solve is to enable consumers to enjoy benefits in various fields by extending the reward refund structure not only to B2C transactions but also to multiple B2B industry groups such as finance, real estate, rental, legal and tax services, travel, and telecommunications.
[0010] Furthermore, the problem that the technical concept of the present disclosure aims to solve is that, based on activity data such as consumer purchase history, mission completion, review writing, and star ratings, AI recommends affiliates, services, products, and events optimized for individual consumers, thereby improving reward utilization and re-engagement rates.
[0011] Furthermore, the problem that the technical concept of the present disclosure aims to solve is to resolve the inconvenience of having to separately handle affiliate search, consultation requests, reward verification and withdrawal, and community participation in existing distributed app and web environments, and to enable all functions to be performed within a single integrated mobile app (UI / UX).
[0012] Furthermore, the problem that the technical concept of the present disclosure aims to solve is to support the influx of new customers, enhancement of brand credibility, and securing of long-term customers for participating businesses by providing a reward-based promotional method with a high actual consumer conversion rate instead of indiscriminate advertising expenditure.
[0013] Meanwhile, the technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem
[0014] An AI-based consumer activity data analysis and customized reward / affiliation benefit provision device linked via a consumer terminal and a communication network according to the present disclosure for achieving the technical problem described above includes: a memory that stores at least one instruction; and a processor that performs an operation according to the instruction. The processor collects activity data such as a consumer's purchase history, service usage history, mission completion status, writing reviews, and star rating evaluation; calculates a consumer grade based on the activity data; applies a differential reward payout rate based on a certain percentage of the store's net profit; stores and manages information on a plurality of industry group affiliates including finance, real estate, rental, legal / tax, travel, and telecommunications in conjunction with the differentially applied reward payout rate; recommends services, products, and events for each consumer through an artificial intelligence model using the consumer grade, activity data, and affiliate information as input values; and can provide the output result of the recommendation, reward information, and affiliate location information in an integrated manner.
[0015] In addition, the processor provides functions for searching for affiliates, requesting consultation, participating in communities, checking rewards, and withdrawing through a consumer terminal, and classifies consumer grades into four or more levels, thereby allowing for differential application of reward payout rates by grade within a certain range.
[0016] In addition, the processor can grant additional rewards based on the results of mission completion, review writing, and star rating included in the consumer's activity data.
[0017] In addition, the processor determines whether missions have been achieved, whether review quality standards have been met, and whether star rating standards have been met among the consumer's activity data as binary values, calculates additional rewards by summing a pre-set fixed bonus whenever each item is satisfied, stores industry, location, reward rate, availability of consultation, and real-time event information for each affiliate, and can provide filtered and sorted data according to the consumer's terminal request.
[0018] In addition, the above processor provides additional rewards
[0019] Mathematical formula 5
[0020] Calculated through the above mathematical formula 5, where B0 is the basic additional bonus, Bm, Bq, and Bs are fixed bonus unit prices corresponding to each item of mission, review, and star rating, M is whether the mission is achieved, Q is whether the review quality standard is met, S is whether the star rating standard is met, and Rcap is the maximum payment limit for additional rewards per transaction, and whenever each condition is met, a fixed amount is added as an additional reward, and the final amount may be limited so as not to exceed a preset maximum.
[0021] Additionally, the processor calculates a Quality Index (QI) of a review written by a consumer, and if the calculated Quality Index (QI) exceeds a predefined threshold, the processor (130) provides an additional reward (Radd) in proportion to the base reward (Rbase).
[0022] Mathematical formula 7
[0023]
[0024] Calculated according to, and in the above mathematical formula 7 is a quality index threshold, Rbase is a basic reward amount calculated from a transaction, γ is a proportional weighting factor, and Rcap is a maximum additional reward per transaction. By analyzing the consumer's past purchase patterns, reward usage patterns, frequency of use of affiliates, and preferred industries, affiliates and products with a high probability of repurchase within a certain period can be recommended first, and based on the location information of the consumer's terminal, a list of nearby affiliates and reward rates can be visualized and provided on a map-based screen.
[0025] In addition, the processor provides a repurchase probability score (Pi) for each affiliate or product based on the consumer's past usage behavior data.
[0026] Mathematical formula 8
[0027]
[0028] Calculated through the above mathematical formula 8, where Pi is the repurchase probability score of affiliate or product i, R^i is the normalized value of the recency indicator, F^i is the normalized value of the frequency indicator, M^i is the normalized value of the monetary indicator, L^i is the normalized value of the indicator based on the reward usage pattern, and wR, wF, wM, and wL are weights for each indicator. When a consultation request is received from a consumer terminal, it is automatically assigned to an affiliate representative or sales personnel, and the assignment result can be provided to the consumer terminal as a real-time notification.
[0029] In addition, the above processor, in order to quantitatively evaluate the service quality of affiliated merchants, uses a merchant comprehensive evaluation index (Sstore).
[0030] Mathematical formula 9
[0031]
[0032] Calculated according to the above mathematical formula 9, Sstore may be a comprehensive evaluation index of franchise stores, Ravg may be a value obtained by converting the average of consumer star ratings into a certain range, Freview may be a value obtained by standardizing the monthly review writing frequency into a certain range, and Qservice may be a service quality index obtained by converting into a certain range. Effects of the invention
[0033] According to the aforementioned means for solving the problem of the present disclosure, economic benefits that consumers can directly experience are provided through a differential reward payment structure based on consumer grades and a high refund rate of up to 90%, thereby improving long-term service usage and re-participation rates.
[0034] In addition, according to the aforementioned means for solving the problem of the present disclosure, accumulated rewards can be used in various ways, such as cash withdrawal, gift certificates, coupons, and prize draws, thereby increasing consumer convenience and reducing the expiration rate of rewards.
[0035] In addition, according to the aforementioned means for solving the problem of the present disclosure, by implementing a reward refund structure linked to various industry groups such as finance, real estate, rental, legal and tax services, travel, and telecommunications, consumers can enjoy benefits in various lifestyle and business fields, and the market scalability of the platform is greatly improved.
[0036] In addition, according to the aforementioned means for solving the problem of the present disclosure, by analyzing activity data such as a consumer's purchase history, writing reviews, performing missions, and star ratings, and having the AI recommend personalized services and products, the utilization of rewards can be increased and consumer repurchase and service loyalty can be strengthened.
[0037] In addition, according to the aforementioned means for solving the problem of the present disclosure, affiliates can utilize reward-based marketing with a high actual purchase conversion rate without indiscriminate advertising expenditure, thereby maximizing advertising efficiency and reducing unnecessary costs.
[0038] In addition, according to the aforementioned means for solving the problem of the present disclosure, by providing various functions such as location-based partner search, consultation request, reward verification and withdrawal, and community participation in an integrated manner within a single app, users can intuitively utilize the entire service and minimize the churn rate.
[0039] Furthermore, according to the aforementioned means for solving the problem of the present disclosure, consumer reward refunds lead to purchase promotion, which forms a virtuous cycle structure that leads to increased sales for affiliate companies and expanded platform revenue, thereby contributing to the revitalization of the domestic economy and the enhancement of competitiveness in local commercial districts in the long term.
[0040] The effects obtainable from the exemplary embodiments of the present disclosure are not limited to those mentioned above, and other unmentioned effects can be clearly derived and understood by those skilled in the art to which the exemplary embodiments of the present disclosure belong from the description below. That is, unintended effects resulting from the implementation of the exemplary embodiments of the present disclosure can also be derived by those skilled in the art from the exemplary embodiments of the present disclosure. Brief explanation of the drawing
[0041] FIG. 1 is a drawing showing an AI-based consumer activity data analysis and customized reward / affiliation benefit provision system according to an embodiment. FIG. 2 is a drawing showing an AI-based consumer activity data analysis and customized reward / partnership benefit provision system according to an embodiment. FIG. 3 is a diagram illustrating the process of AI-based consumer activity data analysis and customized reward and partnership benefit provision according to an embodiment. Specific details for implementing the invention
[0042] Hereinafter, various embodiments of the present disclosure are described in conjunction with the accompanying drawings. As various embodiments of the present disclosure may be subject to various modifications and may have various forms, specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the various embodiments of the present disclosure to specific forms, and it should be understood that they include all modifications and / or equivalents and substitutions that fall within the spirit and scope of the various embodiments of the present disclosure. In relation to the description of the drawings, similar reference numerals have been used for similar components.
[0043] In various embodiments of the present disclosure, terms such as “comprising” or “having” are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.
[0044] In various embodiments of the present disclosure, expressions such as “or” include any and all combinations of the words listed together. For example, “A or B” may include A, may include B, or may include both A and B.
[0045] Expressions such as "first," "second," "first," or "second" used in various embodiments of the present disclosure may modify various components of the various embodiments, but do not limit such components. For example, such expressions do not limit the order and / or importance of such components and may be used to distinguish one component from another.
[0046] When it is mentioned that a component is "connected" or "joined" to another component, it should be understood that the component may be directly connected or joined to the other component, but that a new component may also exist between the component and the other component.
[0047] In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware and software. Additionally, a plurality of "modules," "units," "parts," etc. may be integrated into at least one module or chip and implemented as at least one processor, except where each needs to be implemented in specific individual hardware.
[0048] Terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the various embodiments of the present disclosure.
[0049] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.
[0050] FIG. 1 is a diagram showing an AI-based consumer activity data analysis and customized reward and partnership benefit provision system according to an embodiment.
[0051] Referring to FIG. 1, the AI-based consumer activity data analysis and customized reward / partnership benefit provision system according to the embodiment may be configured to include a user-customized AI-based consumer activity data analysis and customized reward / partnership benefit provision device (100) and a user terminal (200). In the embodiment, the AI-based consumer activity data analysis and customized reward / partnership benefit provision device (100) collects activity data such as the consumer's purchase history, service usage history, mission completion status, review writing, and star rating evaluation. Subsequently, based on the collected activity data, the consumer grade is calculated, a reward payout rate is applied differentially according to a certain percentage of the store's net profit, and information on multiple industry group partners, such as finance, real estate, rental, legal / tax, travel, and telecommunications, is stored and managed. Using the consumer grade, activity data, and partner information as input values, customized services, products, and events for each consumer are recommended through an artificial intelligence model, and the AI recommendation results, reward information, partner location information, and availability of various consultations are provided in an integrated manner.
[0052] In particular, the aforementioned review writing information may include a function that automatically posts review text and attached images to linked external review platforms (e.g., blogs, social media, etc.) based on affiliate information contained in the receipt when a consumer photographs and submits the receipt. In this case, the user can simultaneously earn rewards and write a review on an external platform through a single activity (photographing and submitting the receipt).
[0053] In addition, the AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) is linked with a user terminal (200) via a communication network, and can process requests for searching for affiliates, requests for consultation, community participation, reward verification, and withdrawal requests entered from the user terminal (200) in real time.
[0054] In the embodiment, the user terminal (200) may be implemented in the form of a smartphone, tablet PC, laptop computer, etc., and receives customized recommendation information, reward accumulation status, partner location and event information transmitted from the device (100) and displays it on the screen.
[0055] In addition, the user terminal (200) is equipped with a location information collection unit, so that a list of nearby affiliates and reward rates can be visualized and provided on a map screen based on the current location. The user-customized AI service recommendation and linked productivity content generation system according to the embodiment performs consumer data analysis, customized AI recommendation, high-rate reward management, affiliate information linkage, and integrated UI / UX provision on a single platform, thereby providing consumers with highly tangible economic benefits and convenience, and simultaneously providing affiliates with the effects of reduced advertising costs, influx of new customers, and improved brand credibility.
[0056] In addition, in the embodiment, the device (100) for providing user-customized AI-based consumer activity data analysis and customized rewards / affiliation benefits differentiates the reward rate provided when making payments at affiliated stores and performs a store entry fee structure, a deposit deduction method, granting regional exclusivity rights, incentives for excellent merchants, and AI-based consultation linkage functions to increase convenience for store operations and consumer use. In the embodiment, store subscription fees and deposit charging and deduction methods can be performed. For example, in the case of general store owners, a monthly subscription fee (e.g., 85,000 won) is charged, deposit ranges by industry (50 / 100 / 150 / 200 / 500 million won) are set, and the deposit can be deducted according to the return rate when consumers use it. In the embodiment, the balance can be returned upon contract termination, and a deposit management logic is included that requires recharging if the deposit balance is below a set threshold (e.g., 300,000 won). The deposit deduction structure provided in the embodiment enables flexible budget management based on consumer usage and increases the stability of reward payments on the platform.
[0057] In addition, the device (100) for AI-based consumer activity data analysis and providing customized rewards and partnership benefits can implement a regional exclusive store system. For example, by restricting additional entry of the same type of store within a radius of 1 km, exclusive business rights are granted to the initial store. In the embodiment, for the regional exclusive store, a monthly subscription fee (150,000 won) and a deposit (3 million won or more) are applied, and for excellent store owners, additional support for the return rate is provided based on the results of the consumer satisfaction evaluation. This acts as a strong motivator when attracting initial franchisees and functions as an incentive structure that encourages service quality. In addition, in the embodiment, promotional materials in the form of banners and stickers are provided upon entry, and the cost of the provided promotional materials can be deducted upon cancellation. The banner includes consumer-friendly phrases such as “This store has an owner who only thinks of the customers,” thereby strengthening credibility and favorability.
[0058] In addition, by introducing an AI-based question-and-answer service linkage and a consultation request function, consumers can quickly process inquiries related to affiliated industries. Furthermore, by identifying the consumer's inquiry intent, the consultation request is automatically assigned to the relevant industry manager, thereby strengthening connectivity between consumers and merchants and improving the consultation conversion rate. Moreover, the AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) according to the embodiment specifies conditions such as the contract period, entry fee fluctuation conditions, exclusive rights for initial entrants, and excellent entry company evaluation methods into quantitative contractual conditions, and in particular, improves the precision of system operation through a deposit deduction structure, recharge conditions, and a return rate addition structure based on consumer satisfaction.
[0059] In addition, in the embodiment, the device (100) for providing AI-based consumer activity data analysis and customized rewards and partnership benefits moves away from simple reward rate competition and combines deposit-based deduction operation, granting regional exclusivity, linking with AI consultation, and consumer satisfaction additional rewards, thereby ensuring reliability and sustainability among the three parties: the partner, the consumer, and the platform. This is differentiated from the simple discount and accumulation structure of conventional reward apps and induces platform stability and long-term participation of partner companies.
[0060] FIG. 2 is a block diagram of an AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) according to an embodiment.
[0061] In the embodiment, the AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) may be implemented as a server. In the embodiment, a server is a computing system that provides services to other computers or devices in a computer network or stores and manages data. The AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) accepts requests from other computers or devices called clients and provides responses or data to those requests. The configuration of the AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) illustrated in FIG. 2 is merely a simplified example.
[0062] The communication module (110) can be configured regardless of the mode of communication, such as wired or wireless, and can be configured with various communication networks, such as a Personal Area Network (PAN) or a Wide Area Network (WAN). Additionally, the communication module (110) can operate based on the known World Wide Web (WWW) and may utilize wireless transmission technologies used for short-range communication, such as Infrared Data Association (IrDA) or Bluetooth. For example, the communication module (110) may be responsible for transmitting and receiving data necessary to perform a technique according to one embodiment of the present disclosure.
[0063] Memory (120) may refer to any type of storage medium. For example, memory (120) may include at least one type of storage medium among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk, and optical disk. Such memory (120) may also constitute the database shown in FIG. 1.
[0064] The memory (120) can store at least one instruction that can be executed by the processor (130). Additionally, the memory (120) can store any form of information generated or determined by the processor (130) and any form of information received by the server (200). Additionally, the memory (120) stores various types of modules, instruction sets, or models.
[0065] The processor (130) can perform technical features according to embodiments of the present disclosure to be described below by executing at least one instruction stored in memory (120). In one embodiment, the processor (130) may be composed of at least one core and may include a processor for data analysis and / or processing, such as a central processing unit (CPU) of a computer device, a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU).
[0066] This processor (130) can train a neural network or model designed in a machine learning or deep learning manner. To this end, the processor (130) can perform calculations for training the neural network, such as processing input data for training, extracting features from input data, calculating errors, and updating the weights of the neural network using backpropagation. Additionally, the processor (130) can perform inference for a specific purpose using a model implemented in an artificial neural network manner.
[0067] In an embodiment, the processor (130) collects activity data such as the consumer's purchase history, service usage history, mission completion status, review writing, and star rating evaluation. The processor (130) obtains the consumer's activity information from a database stored on a consumer terminal or server, and the activity information may include the consumer's purchase history, service usage history, mission completion status, review writing content, and star rating evaluation results. Specifically, the purchase history may include payment date and time, payment amount, payment method, and identification information of the payment target partner, and the service usage history may include the type of partner service accessed or used by the consumer, frequency of use, duration of use, etc. Additionally, the mission completion status indicates the result of performing a predefined consumer participation event or task, and the review writing information may include evaluation text and related metadata entered by the consumer regarding the partner or service. In the embodiment, the star rating represents a quantified satisfaction score given by a consumer to an affiliate or service, and the processor (130) can normalize and pre-process this activity data and use it for subsequent calculations such as consumer rating calculation, reward payout rate calculation, and personalized service recommendation.
[0068] Subsequently, the processor (130) calculates a consumer grade based on the activity data and applies a differential reward payout rate according to a certain percentage of the store's net profit. In the embodiment, the processor (130) analyzes the collected consumer activity data to calculate a consumer grade and can apply a differential reward payout rate according to a certain percentage of the store's net profit based on the calculated consumer grade. To this end, the processor (130) calculates a consumer activity score by weighting indicators such as purchase frequency, payment amount, service usage consistency, mission completion rate, review writing frequency and quality, and average star rating value. Subsequently, the consumer grade is determined by comparing the consumer activity score with a predefined grade standard table, and the grade may be divided into multiple stages, for example, from Grade 1 to Grade 4. Based on the determined consumer grade, a pre-contracted rate is applied within the range of 10% to 50% of the store's net profit, and a differential reward payout rate is calculated for each grade within the range of 60% to 90%.
[0069] For example, a reward calculated by multiplying the contract ratio of the store's net profit by a coefficient of 0.9 may be paid to consumers of the highest tier, while a relatively lower coefficient may be applied to consumers of lower tiers. The processor (130) determines the final reward amount by dynamically reflecting the fluctuation in net profit by affiliate, the event period, and additional compensation conditions during the process of calculating the reward payout rate, and transmits the result to the consumer terminal so that it can be checked in real time.
[0070] Additionally, the processor (130) can calculate a consumer activity score (S) based on consumer activity data, determine a consumer grade (G) by comparing the calculated score with a predefined grade standard, and calculate a final reward amount (R_t) by reflecting the store net profit, pre-contract ratio, and grade-specific payout rate according to the determined grade.
[0071] Specifically, the consumer activity score (S) can be calculated according to the following mathematical formula 1.
[0072] Mathematical formula 1
[0073]
[0074] In mathematical equation 1, here, z F is purchase frequency (normalized value), z A is the payment amount (normalized value), z U is service usage continuity, z M is the mission completion rate, z R is the frequency of writing reviews, z Q represents the average value of the review quality or star rating. wk represents the weight of each indicator, and the sum of the weights is 1. The consumer activity score (S) calculated in the example can be mapped to a consumer grade (G) by comparing it with a preset grade boundary value (θ) according to the following mathematical formula 2.
[0075] Mathematical formula 2
[0076]
[0077] In mathematical formula 2, θ0 to θ4 are boundary values defining the interval from grade 1 to grade 4, satisfying the condition θ0 < θ1 < θ2 < θ3 < θ4. 1{·} is an indicator function that returns 1 if the condition is true. Additionally, in the embodiment, according to the determined consumer grade (G), the processor (130) can calculate the final reward amount (Rt) according to the following mathematical formula 3 by reflecting the store net profit (NPt), the pre-contract ratio (α), and the grade-specific payout rate (βG).
[0078] Mathematical formula 3
[0079]
[0080] In mathematical formula 3, α is a pre-contract ratio included in the range of 0.10 to 0.50, and βG is a grade-specific payout rate included in the range of 0.60 to 0.90. The clip(x, a, b) function performs the function of limiting the calculated value to Rmin if it is smaller than the minimum limit (Rmin), and to Rmax if it is larger than the maximum limit Rmax. In the embodiment, the processor (130) can provide differentiated economic benefits to consumers by calculating the consumer grade through the weighted sum of activity indicators and calculating the reward amount by combining the store net profit, the contract ratio, and the grade-specific payout rate.
[0081] Subsequently, the processor (130) stores and manages information on multiple industry group affiliates, including finance, real estate, rental, legal and tax, travel, and telecommunications, in conjunction with the differentially applied reward payout rate. In an embodiment, the processor (130) can store and manage information on multiple industry group affiliates, including finance, real estate, rental, legal and tax, travel, and telecommunications, in conjunction with the differentially applied reward payout rate. To this end, the processor (130) stores affiliate metadata in a database, including industry classification, trade name, branch location, scope of services provided, contracted reward rate, net profit ratio, event information, and availability of consultation for each affiliate. Additionally, the processor (130) maps the reward payout rate by consumer grade to the affiliate information, thereby managing the calculation of differentiated rewards according to the contract conditions of each affiliate even within the same industry. For example, for partners in the financial industry, reward rates by specific services such as loan consultation, card issuance, and insurance subscription can be stored, and for partners in the real estate industry, reward rates based on property type, contract size, and brokerage commission rate can be stored.
[0082] Additionally, the processor (130) can update the database when changes in partner information (changes in contract conditions, registration of new branches, initiation of events, etc.) occur, and can update recommendation information and benefit information to be provided to consumer terminals in real time based on the updated partner information and differentially applied reward payout rates. As a result, consumers can receive optimized reward benefits linked with partners in various industries, and partners can perform efficient marketing through differentiated reward policies tailored to industry characteristics.
[0083] Subsequently, the processor (130) uses consumer grades, activity data, and affiliate information as input values to recommend customized services, products, and events to consumers through an artificial intelligence model, and provides integrated output results of the recommendations, reward information, and affiliate location information. In an embodiment, the processor (130) can recommend customized services, products, and events to consumers through an artificial intelligence model using the calculated consumer grade, collected activity data, and stored affiliate information as input values. To this end, the processor (130) converts reward payout rates, purchase frequency, preferred industries, location-based usage patterns, past event participation history, and review writing trends according to consumer grades into feature vectors and provides them as input values to a trained recommendation model (e.g., a collaborative filtering-based model, a deep learning-based ranking model, or a hybrid recommendation model). In an embodiment, the artificial intelligence model matches the input feature vectors with service attributes, product catalogs, and event metadata for each affiliate to output items with high consumer preference in the form of probabilities or scores. The output results are sorted by priority, and each recommendation item may include additional information such as the applicable reward rate for the affiliate or service, the estimated refund amount, and event participation conditions.
[0084] In addition, the processor (130) integrates and provides map data including partner location coordinates, travel paths, and information on the density of nearby partners based on the current location of the consumer terminal along with the recommended items. As a result, the consumer can intuitively check recommended service, product, and event information, the location of the corresponding partner, applicable reward information, and the availability of real-time consultation on a single screen, and can immediately make a reservation, purchase, or request a consultation. Furthermore, in the embodiment, the processor (130) can continuously improve recommendation accuracy by collecting feedback data such as the click-through rate, purchase conversion rate, and event participation rate of the recommended items, and periodically retraining the parameters of the artificial intelligence model.
[0085] Additionally, the processor (130) provides functions for searching for affiliates, requesting consultation, participating in a community, checking rewards, and withdrawing funds through the consumer terminal. In an embodiment, the processor (130) may provide functions for searching for affiliates, requesting consultation, participating in a community, checking rewards, and withdrawing funds through the consumer terminal. In an embodiment, the affiliate search function searches for a list of nearby affiliates based on the location information of the consumer terminal and sorts and displays the results according to filter conditions such as industry, reward rate, event status, and distance. At this time, the search results may be provided in a map-based screen or list format and may include detailed information for each affiliate (address, contact information, operating hours, applicable reward rate, review rating, etc.). Additionally, the consultation request function provides a consultation request form for the affiliate or service item selected on the consumer terminal, receives data such as consultation content, contact information, and desired time entered by the user, and transmits it in real time to the person in charge of the affiliate or designated sales personnel. The processor (130) may display the consultation request processing status (waiting, in progress, completed) on the consumer terminal.
[0086] Furthermore, the aforementioned community participation function enables consumers to exchange experiences through dedicated talk rooms, small group bulletin boards, and review sharing spaces within the consumer's device, and supports activities such as writing posts, entering comments, and uploading images and videos. Additionally, community activity history can be linked to reward accumulation conditions. The aforementioned reward checking function displays the currently accumulated reward (X-Pay) amount, payout rates by tier, usage history, and amounts scheduled to expire via the My Page screen of the consumer's device.
[0087] In the embodiment, the withdrawal function links with a withdrawal account or simple payment service set by the consumer to process the transfer of a withdrawable amount among the accumulated rewards in real time, and provides the withdrawal request and processing result in the form of a notification on the consumer terminal screen. In this way, the processor (130) can ensure the continuity of the consumer experience and minimize the service dropout rate by providing the entire process from searching for partner information to consultation, participation, and reward utilization in an integrated manner within the consumer terminal.
[0088] Additionally, the processor (130) classifies consumer grades into four or more levels and applies a differential reward payout rate for each grade within the range of 60% to 90%. In an embodiment, the processor (130) collects activity data from the databases of consumer terminals and servers, including purchase frequency, total payment amount, usage persistence (activity over the last 30 / 90 days), mission completion rate, review writing frequency and quality, average star rating, refund / cancellation rate, and indicators of fraudulent use. Each collected indicator is normalized and outlier-corrected based on a period (e.g., the last 90 days), and missing values can be replaced by a predefined rule (recency-weighted interpolation or conservative default values). Additionally, the processor (130) calculates a consumer activity score by applying weights to the normalized indicators. The weights may be periodically updated by an operational policy or a learning model and may include a recency-weighting coefficient to give higher importance to recent activities. The calculated activity score is provided as input to the grade calculation step.
[0089] Subsequently, the processor (130) determines the consumer grade by comparing the activity score with the grade criteria table. The grades are defined as at least four levels, and VIP or promotional grades may be added as needed. For example, Grade G1 corresponds to cases where the activity score is 0 or higher and less than u^, and is set as a new subscriber or basic grade. Grade G2 corresponds to cases where the activity score is u^ or higher and less than u**, and is classified as an 'Active' grade with active service usage. Grade G3 corresponds to cases where the activity score is u** or higher and less than y', and is classified as an 'Excellent' grade with high usage performance. Grade G4 corresponds to cases where the activity score is y' or higher, and is classified as the highest grade, 'Superior' grade. The optionally operated Grade G5 is granted when the activity score is θ₄ or higher and may be applied to VIP customers or special event participants. The grades according to the embodiment are set stepwise according to the activity score range, and each grade can be used as a standard for differential application of reward payout rates and service benefits.
[0090] In the embodiment, the payout rate may be increased or decreased based on affiliate contract conditions, industry-specific limits, and event period weighting, but the system enforces upper and lower limits so that the final payout rate is always between 60% and 90%. Additionally, when an affiliate transaction occurs, a base amount for reward calculation based on the affiliate contract (e.g., Net Profit × Contract Rate) is calculated. Subsequently, the processor (130) checks the current grade of the consumer and calculates the scheduled reward by multiplying the base amount by the payout rate corresponding to the grade. In the embodiment, if there are activity-based bonus points such as mission rewards or review quality rewards, they may be additionally reflected within a predefined limit (maintaining the final payout rate upper limit of 90%). Additionally, if a refund / cancellation or anomaly occurs, the return or withholding of rewards may be applied during the post-settlement stage.
[0091] In addition, dynamic adjustments can be performed through policy linkage in the embodiments. For instance, in the case of the event period weighting policy, the payout rate by tier is multiplied by a promotion coefficient during specific periods while maintaining a cap (90%), and in the case of the industry / affiliate limit policy, a cap is applied to ensure that the industry-specific reward budget or per-case / daily / monthly limits are not exceeded. Furthermore, in the case of the risk management policy, if an excessive refund rate or repeated reviews from the same device or multiple accounts are detected, the payout rate is temporarily reduced or payments are suspended, and in the case of the scale discount / loyalty bonus policy, the tier maintenance period is extended or early promotion to the next tier is possible based on cumulative usage or duration.
[0092] Subsequently, the processor (130) stores the basis indicators for the grade calculation, the applied weights / thresholds, the calculated payout rate, the final reward amount, and the reason for the correction in a log. Additionally, the grade, the applied payout rate, the calculation reference date, the next re-evaluation scheduled date, and the reason for the payment restriction (if any) are transparently notified to the consumer terminal. The operator can check the grade change history and the results of each step of the payment formula in the form of an audit trail through the administrator console.
[0093] For example, if consumer A's activity score is evaluated as u** or higher and less than y' and is calculated as G3, the processor (130) applies an 80% payout rate to the reward calculation standard amount of the transaction. Even if a +5% bonus is set during the event period, the final payout rate is clamped at 90% so as not to exceed the upper limit of 90%.
[0094] Subsequently, the processor (130) grants additional rewards based on the results of mission performance, review writing, and star rating evaluation included in the consumer's activity data. In an embodiment, the processor (130) may analyze the results of mission performance, review writing, and star rating evaluation included in the consumer's activity data and grant additional rewards separately from the basic rewards.
[0095] In the embodiment, the processor (130) can calculate an additional reward (Radd) by converting the mission completion rate, review quality score, and star rating included in the consumer's activity data into normalized values in the range of 0 to 1, calculating a weighted sum by assigning a preset weight to each item, and then multiplying this by the base reward (Rbase). In the embodiment, the calculation of the additional reward can be performed according to the following mathematical formula 4.
[0096] Mathematical formula 4
[0097]
[0098] In mathematical formula 4, m is the mission completion rate (normalized value from 0 to 1), q is the review quality score (e.g., a value normalized to a range of 0 to 1 by comprehensively evaluating text length, number of valid keywords, and whether images or videos are attached), s is the star rating normalized on a 5-point scale (s = star rating / 5), a, b, and c are the weights of each item, and the sum of the three weights does not exceed the additional reward upper limit rate δ. The Radd calculated in this way increases in proportion to the consumer's basic reward, but can be flexibly adjusted to reflect the importance of each item.
[0099] Additionally, the processor (130) can determine whether the mission is achieved, whether the review quality standard is met, and whether the star rating standard is met among the consumer's activity data as binary values (1 or 0), and calculate an additional reward by summing a pre-set fixed bonus whenever each item is met. Additionally, the calculation of the additional reward can be performed according to the following mathematical formula 2.
[0100] Mathematical formula 5
[0101]
[0102] In mathematical formula 5, B0 is the basic additional bonus (when always paid), and Bm, Bq, and Bs are fixed bonus unit prices corresponding to each item (mission, review, star rating). M is whether the mission is achieved (1 if achieved, 0 if not met), and Q is whether the review quality standard is met (e.g., 1 if text of 300 characters or more + photo attached, 0 if not met). S is whether the star rating standard is met (e.g., 1 if ★4.5 or higher, 0 if lower), and Rcap is the maximum additional reward payout limit per transaction. According to the present embodiment, whenever each condition is met, a fixed amount is added as an additional reward, and the final amount is limited so as not to exceed Rcap.
[0103] In the embodiment, the mission execution information indicates whether a predefined consumer participation task (e.g., using a specific affiliate, making a payment of more than a certain amount within a set period, participating in a promotional event, etc.) has been completed, and the processor (130) calculates an additional reward amount by assigning a weight to each mission based on the difficulty of the mission, the degree of satisfaction of completion conditions, the time required for execution, etc.
[0104] In the embodiment, the review writing information may include evaluation text written by a consumer after using an affiliate or provided service, attached images / videos, the time of writing, the number of characters, whether keywords are included, etc., and the processor (130) calculates a quality index of the review and provides additional rewards if it exceeds a predefined quality standard. For example, the quality index may be calculated by combining text length, positive / negative sentiment analysis results, the validity of attached media, etc.
[0105] In an embodiment, the processor (130) may limit the text length, sentiment analysis results, and attached media validity to a range of 0 to 1, respectively, in order to calculate the Quality Index (QI) of a review written by a consumer.
[0106] The quality index can be calculated according to the following mathematical formula 6.
[0107] Mathematical formula 6
[0108]
[0109] In mathematical formula 6, L^ is the value normalized by dividing the text length by the reference length, S^ is the value converted from the sentiment analysis result to a range of 0 to 1, M^ is the value normalized by dividing the number of attached media M by the reference number, wL, wS, and wM are weights representing the importance of each item, and their sum is maintained at 1. clip(x,0,1) is an operation that limits the calculation result to between 0 and 1.
[0110] Subsequently, if the quality index (QI) exceeds a predefined threshold, the processor (130) may calculate an additional reward (Radd) in proportion to the base reward (Rbase). In this case, a proportional weighting factor (γ) is applied, but limited so as not to exceed the additional reward upper limit (Rcap). The calculation of the proportional additional reward can be performed according to the following mathematical formula 7.
[0111] Mathematical formula 7
[0112]
[0113] In mathematical formula 7 is the quality index threshold (e.g., 0.6), Rbase is the base reward amount calculated from the transaction, γ is the proportional weighting factor, and Rcap is the maximum additional reward per transaction.
[0114] Additionally, a star rating refers to a quantified satisfaction score (e.g., a 1- to 5-point scale) given by a consumer to an affiliate or service, and the processor (130) may provide additional rewards differentially depending on whether the star rating is submitted and the level of the score. For example, a basic bonus point reward may be provided for simple submission, and an additional bonus point reward may be provided for giving a score above a certain level.
[0115] In addition, the processor (130) may provide additional rewards by calculating a composite participation reward when multiple items are performed simultaneously, in addition to individually evaluating the three items of mission execution, review writing, and star rating. At this time, the reward calculation may be performed by summing the basic weights for each item, or by applying a weighted sum method that applies interaction coefficients between items. In the embodiment, the additional reward calculated is added to the basic transaction reward and reflected in real-time on the reward accumulation screen of the consumer terminal, and the withdrawal availability time, validity period, usage conditions, etc., may be the same as or separately set from the basic reward.
[0116] Additionally, the processor (130) stores industry, location, reward rate, availability of consultation, and real-time event information for each affiliate, and provides filtered and sorted information according to a request from a consumer terminal. The processor (130) can store affiliate metadata including industry, location information, reward rate, availability of consultation, and real-time event information for each affiliate in a database, and provide said metadata by filtering and sorting it according to a request from a consumer terminal.
[0117] In the embodiment, industry information may be classified into multiple industry groups, such as finance, real estate, rental, legal / tax, travel, and telecommunications, and each industry group may include detailed service codes or keywords. Location information includes the partner's address, GPS coordinates, and identification information at the administrative district level, and supports distance-based search by comparing it with the location coordinates of the consumer terminal. The reward rate is a payout rate pre-contracted with the partner and may be stored by distinguishing between a basic payout rate and an event payout rate. Consultation availability includes the real-time consultation availability status, estimated response time, and consultation channels (telephone, chat, visit, etc.) for each partner, and can be periodically updated through the partner's administrator terminal or API integration. Additionally, real-time event information includes time-limited information such as discount events, reward rate increase promotions, and new service launches.
[0118] In an embodiment, the processor (130) analyzes search conditions entered from a consumer terminal (e.g., industry, minimum reward rate, maximum distance, availability of consultation, availability of event) and filters the stored affiliate metadata according to the conditions. The filtered results are sorted according to predefined sorting criteria (e.g., reward rate descending order, distance ascending order, event priority, overall score), and multiple sorting criteria may be applied if necessary.
[0119] The sorted list of affiliates is transmitted to a consumer terminal and displayed in a map-based UI or list format, and each affiliate item can be visually provided with industry, location, reward rate, availability of consultation, and event details. In addition, the processor (130) can analyze the consumer's past usage history, preferred industry, and reward usage patterns to generate a customized ranking with personalized weights applied even within the same filter and sorting request.
[0120] Additionally, the processor (130) analyzes the consumer's past purchasing patterns, reward usage patterns, frequency of use of affiliates, and preferred industries, and prioritizes recommending affiliates and products that are highly likely to be repurchased within a certain period in the future. In the embodiment, the processor (130) can prioritize recommending affiliates and products that are highly likely to be repurchased within a certain period in the future by analyzing the consumer's past purchasing patterns, reward usage patterns, frequency of use of affiliates, and preferred industries.
[0121] Specifically, the above purchase pattern includes a list of affiliates paid to by the consumer, payment amount, payment interval, payment time zone, and seasonal purchase trends, and the processor (130) converts this into time-series data to model consumption behavior. The above reward usage pattern includes the timing of use, amount used, place of use, withdrawal frequency, and gifticon / coupon conversion rate of accumulated rewards, through which the consumer's reward consumption preference is derived. The frequency of affiliate usage includes the number of times the same affiliate or affiliates within the same industry have been revisited, the time of the last visit, and the visit cycle. In addition, the preferred industry can be determined by combining the purchase frequency, reward usage rate, review writing frequency, and positive evaluation rate.
[0122] Based on the analysis results, the processor (130) applies a machine learning or statistical prediction model (e.g., logistic regression, time series forecasting, collaborative filtering, graph-based recommendation algorithm) to calculate the probability value or score of the likelihood that the consumer will repurchase within a predefined period (e.g., 7 days, 30 days, 90 days). At this time, the input vector of the model may include not only RFM indicators such as purchase cycle, recency, frequency, and monetary value, but also reward-specific indicators such as the reward balance and the amount scheduled to expire.
[0123] The calculated repurchase probability score is mapped to each affiliate and product, and the processor (130) generates a recommendation list by assigning priority in order of highest score. In the embodiment, the processor (130) can calculate a repurchase probability score (Pi) for each affiliate or product based on the consumer's past usage behavior data. In this embodiment, key indicators such as purchase cycle, recency, usage frequency, and reward usage pattern are normalized and then combined using a weighted sum method to calculate the repurchase probability score.
[0124] Mathematical formula 8
[0125]
[0126] In Equation 8, Pi is the repurchase probability score (range 0–1) for the affiliate or product iii, and R^i is the normalized value of the Recency indicator, calculated by dividing the number of days elapsed since the last purchase date by the reference range. F^i is the normalized value of the Frequency indicator, calculated based on the number of purchases within a certain period, and M^i is the normalized value of the Monetary indicator, calculated based on the payment amount within a certain period. L^i is the normalized value of the Reward Usage Pattern indicator, calculated based on the ratio of accumulated rewards used at the relevant affiliate or product. wR, wF, wM, and wL are the weights for each indicator (the sum is 1). Each weight can be pre-set. The closer the value of Pi calculated by Equation 8 above is to 1, the higher the probability of repurchasing the relevant affiliate or product within a short period. In the embodiment, the processor (130) maps the calculated Pi values to affiliates and products, then generates a recommendation list by assigning priorities in order of highest score, and displays it on a consumer terminal.
[0127] The recommendation list may include additional information such as the expected repurchase time, applicable reward rates, ongoing events, and expected refund amount. Additionally, the processor (130) can place immediately available affiliates and products with a high likelihood of repurchase at the top by considering the consumer's location information and the real-time business status of affiliates.
[0128] Recommendation results are provided via personalized recommendation screens on consumer devices, notification messages, or email and push notifications, and consumer clicks, reservations, and purchases are stored as feedback data for subsequent analysis.
[0129] Additionally, the processor (130) visualizes and provides a list of nearby affiliates and reward rates on a map-based screen based on the location information of the consumer terminal. The processor (130) can visualize and provide a list of nearby affiliates and the reward rates of the affiliates on a map-based screen based on the location information of the consumer terminal. In an embodiment, the processor (130) obtains current location coordinates (latitude, longitude) through a GPS receiver of the consumer terminal or a network base station-based location estimation module. The obtained location coordinates are compared with the location coordinates of each affiliate included in a pre-stored affiliate database (DB) and are used to search for affiliates located within a predefined radius (e.g., 500m, 1km, 5km). The searched list of affiliates may include information such as industry, business name, address, business hours, availability of real-time consultation, applicable reward rates, and ongoing events. The processor (130) also extracts the reward rates for each affiliate and visually displays them on a map-based UI.
[0130] Additionally, on map-based screens, markers can be placed at the locations of each partner, and their color, size, or label can be differentiated based on the reward rate. For example, markers can be displayed in darker colors or larger sizes as the reward rate increases, and clicking a marker can provide detailed partner information and benefits in a pop-up format.
[0131] Additionally, the processor (130) supports various search and sorting options, such as map zooming, industry-specific filters, minimum reward rate conditions, and event status, allowing the map display results to be updated in real time according to the conditions desired by the consumer. In the embodiment, the map-based screen is accessible through the main UI of the consumer terminal or a dedicated partner search menu, and can dynamically update the list of partners and reward rate information within the display range by processing interaction events such as user touch, drag, and zooming in / out. Through this, the consumer can intuitively identify nearby partners and reward rates based on their current location, allowing them to quickly select and use a partner that can maximize reward benefits.
[0132] Additionally, when a consultation request is received from a consumer terminal, the processor (130) automatically assigns it to a partner company representative or sales personnel and provides the assignment result to the consumer terminal as a real-time notification. In an embodiment, when a consultation request is received from a consumer terminal, the processor (130) may automatically assign it to a partner company representative or designated sales personnel to process the request and provide the assignment result to the consumer terminal as a real-time notification. Specifically, the processor (130) receives consultation request data entered from the consumer terminal, and the data may include partner company identification information, consultation request type (e.g., product inquiry, contract consultation, after-sales service), consumer contact information, desired consultation time, and additional requests.
[0133] In an embodiment, the processor (130) queries the list of representatives or sales personnel of the affiliate in the affiliate database (DB) and automatically assigns a consultation personnel by considering conditions such as working status, current number of consultations, industries that can be handled, geographical proximity, and response history. The assignment algorithm may apply one or more of priority rules, round-robin, and weight-based load balancing methods, and may dynamically adjust the assignment weight according to the specific personnel's area of expertise or recent response speed.
[0134] In the embodiment, when the assignment is completed, the processor (130) transmits the consultation request details in real time to the assigned representative or salesperson and simultaneously provides the assignment result to the consumer terminal in the form of a notification. The notification may include the representative's name, job title, contact information, estimated response time, and consultation channel (telephone, chat, visit, etc.).
[0135] Additionally, the processor (130) can track the consultation progress status (waiting, in progress, completed) in real time after assignment and provide an immediately updated notification to the consumer terminal when the status changes. If a reason arises that prevents consultation, it can automatically reassign another person in charge or substitute sales personnel and notify the consumer terminal of this fact again. This ensures that consumer consultation requests are immediately connected to the appropriate personnel without a manual approval process, thereby minimizing consultation delays and improving consumer satisfaction.
[0136] Additionally, the processor (130) analyzes the consumer's reward usage habits and the scheduled expiration date of the balance, automatically recommends affiliates, products, and services that can utilize a certain percentage or more of the rewards scheduled to expire, and provides a one-click payment or reservation function along with a notification before expiration to the consumer's terminal. Specifically, the processor (130) collects reward accumulation and usage history from the consumer's terminal or server, and calculates the scheduled amount and expiration date based on the validity period of each reward.
[0137] The calculated amount scheduled to expire is analyzed along with indicators such as the consumer's past purchase industry, reward usage patterns, preferred product categories, and usage frequency, thereby automatically generating a list of partner companies and products that can efficiently utilize the amount scheduled to expire. In the embodiment, the generated list is sorted based on reward usage efficiency, estimated refund amount, partner location, and real-time stock and reservation availability, and push notifications are sent to the consumer's device according to a pre-set period (e.g., 7, 3, or 1 day) prior to the scheduled expiration date. The push notification includes a one-click payment and reservation button, allowing the consumer to immediately use the reward and complete the payment within the app. Through this, consumers can minimize economic losses due to the expiration of rewards, partner companies can expect increased consumer revisits and sales, and the platform can improve the reward utilization rate and activate service usage.
[0138] Additionally, the processor (130) can combine the consumer's real-time location information and weather data to automatically calculate a list of affiliates or services whose reward rates fluctuate according to weather conditions, and provide this list along with the fluctuating reward rates to the consumer terminal. In an embodiment, the processor (130) obtains the current location through the consumer terminal's GPS module and receives the real-time temperature, precipitation, humidity, wind speed, fine dust index, etc., of the location from an external weather data API. Subsequently, by matching this with the weather-linked reward policies for each industry and service type stored in the affiliate database (DB), it calculates a list of affiliates where the reward rate increases or special benefits are applied under the current weather conditions. For example, if the reward rate for taxi and designated driver affiliates is increased by 10% during heavy rain, or if the reward rate for cafes and ice cream shops is adjusted upward during a heatwave, the processor (130) displays this information along with the estimated additional refund amount to the consumer terminal. Furthermore, the processor (130) can analyze the consumer's past usage patterns by weather and provide industries and products with high preference sorted by priority. According to this embodiment, consumers can recognize and utilize immediate benefits resulting from weather changes, and affiliates can expect increased demand tailored to weather conditions.
[0139] Additionally, the processor (130) can apply a reward rate provided when making a payment at an affiliated store differently according to the consumer grade or affiliate conditions, and can perform functions such as managing the store entry fee structure to increase the operational stability of the merchant and the convenience of consumer use, a deposit deduction method, granting regional exclusivity, providing incentives for excellent merchants, and linking with AI-based consultation. Specifically, the processor (130) executes the logic for managing the merchant's subscription fee and deposit by referring to the merchant's grade, industry, and contract conditions. For example, in the case of a general merchant, a monthly subscription fee (e.g., 85,000 won) is charged, and different deposit ranges (e.g., 50 / 100 / 150 / 200 / 500 million won) are set according to the industry. Then, when a consumer transaction occurs, the deposit is deducted according to a pre-set return rate, and control is made to request a recharge if the deposit balance falls below a threshold (e.g., 300,000 won). When the contract is terminated, the procedure for returning the balance shall be carried out, but the costs for producing and providing promotional materials supported at the time of termination shall be deducted.
[0140] Additionally, the processor (130) can also implement a regional exclusive store system. For example, by restricting additional entry of the same type of store within a radius of 1 km, it grants exclusive business rights to the initial store, and applies a monthly subscription fee (e.g., 150,000 won) and a deposit (e.g., 3 million won or more) to the regional exclusive store. At this time, additional support is provided to the return rate based on the results of the consumer satisfaction evaluation, thereby providing incentives to franchisees with excellent service quality.
[0141] In an embodiment, the processor (130) calculates a store comprehensive evaluation index (Sstore) to quantitatively evaluate the service quality of an affiliated merchant, and classifies the merchant as an excellent merchant if the evaluation index is greater than or equal to a predefined excellent judgment criterion value. Specifically, the store comprehensive evaluation index can be calculated according to the following mathematical formula 9.
[0142] Mathematical formula 9
[0143]
[0144] In mathematical formula 9, Sstore is the comprehensive evaluation index of franchise stores, Ravg is the average consumer star rating converted into a range of 0 to 100 points, Freview is the standardized monthly review writing frequency converted into a range of 0 to 100 points, and Qservice is the service quality index (range of 0 to 100 points) calculated by combining response speed, repurchase rate, claim processing rate, etc.
[0145] In mathematical formula 9, the weights of each item are set, for example, to an average star rating of 0.4, a review writing frequency of 0.3, and a service quality index of 0.3, which reflect the contribution to service quality. In the embodiment, the processor (130) can determine a merchant as an excellent merchant if the overall evaluation index exceeds a preset standard value (e.g., 85 points).
[0146] By configuring it in this way, the processor (130) can perform quantitative and objective selection of excellent merchants by integrating consumer evaluation data and service operation indicators, and based on this, can automate follow-up measures such as adding reward rates, providing incentives, and raising recommendation rankings within the platform.
[0147] Additionally, the processor (130) can perform logic for providing promotional materials and deducting costs. Upon entry, promotional materials in the form of banners and stickers are provided to the franchisee, and the banners include consumer-friendly phrases such as “This store has a boss who only cares about customers” to enhance credibility and favorability. Upon contract termination, the costs of producing and supporting the promotional materials are deducted from the deposit. Furthermore, the processor (130) provides a function for linking with an AI-based question-and-answer service and automatically assigning consultation requests. When a consumer inputs an inquiry related to a partner industry, the intent of the question is identified through natural language processing-based analysis, and the inquiry is automatically assigned to the relevant industry manager or consultation personnel. This strengthens the connection between the consumer and the franchisee and improves the consultation conversion rate. Moreover, the processor (130) manages the contract period, entry fee fluctuation conditions, exclusivity rights for initial entrants, and evaluation methods for excellent tenants as quantitative contract conditions, and improves the precision and stability of system operation by including a deposit deduction structure, recharge conditions, and logic for adding return rates based on consumer satisfaction.
[0148] Through this, the processor (130) according to the embodiment moves away from simple reward rate competition and combines deposit-based deduction operation, granting of regional exclusivity, linkage with AI consultation, and consumer satisfaction-based incentives, thereby ensuring reliability and sustainability among the three parties—affiliates, consumers, and platforms—and providing technical effects that are differentiated from the simple discount and accumulation structure of conventional reward apps.
[0149] Hereinafter, we will look at FIG. 3. The method for creating user-customized AI service recommendations and linked productivity content illustrated in FIG. 3 can be performed by an AI-based consumer activity data analysis and customized reward / affiliation benefit provision device (100) including a processor (130).
[0150] Meanwhile, FIG. 3 is merely illustrative, and the concept of the present invention is not to be interpreted as being limited to that depicted in FIG. 3. For example, each step may be configured in a different order than that depicted in FIG. 3, at least one of the steps depicted in FIG. 3 may not be performed, or one or more steps not depicted in FIG. 3 may be additionally performed. Below, a method for user-customized AI service recommendation and linked productivity content generation will be described in order. Since the operation (function) of the user-customized AI service recommendation and linked productivity content generation method according to the embodiment is essentially the same as the function of the user-customized AI service recommendation and linked productivity content generation system, descriptions that overlap with FIG. 1 and FIG. 2 will be omitted.
[0151] FIG. 3 is a diagram illustrating the process of user-customized AI service recommendation and linked productivity content creation according to an embodiment.
[0152] In step S110, activity data such as consumer purchase history, service usage history, mission completion status, review writing, and star ratings are collected; based on the said activity data, a consumer grade is calculated, and a differential reward payout rate is applied according to a certain percentage of the store's net profit. In step S120, linked to the differentially applied reward payout rate, information on affiliates in multiple industry sectors, including finance, real estate, rental, legal and tax services, travel, and telecommunications, is stored and managed. In step S130, using the consumer grade, activity data, and affiliate information as input values, services, products, and events are recommended to each consumer through an artificial intelligence model; and in step S140, the output results of the recommendations, reward information, and affiliate location information are provided in an integrated manner.
[0153] Meanwhile, the methods according to the various embodiments of the present invention described above can be implemented in the form of an application or software program that can be installed on an existing electronic device.
[0154] In addition, the whole or part of the method may be composed of multiple software function modules and implemented on an operating system (OS). Alternatively, each step may be composed of a single software function module, or each step may be combined to form a single software function module and implemented on an operating system. Therefore, even if all of the embodiments of the present disclosure are not implemented as a single software function module, if multiple software function modules implement each step of the present disclosure and multiple software function modules are implemented on a single operating system, it can be understood that the method of the present disclosure has been implemented.
[0155] In addition, the methods according to the various embodiments of the present invention described above can be implemented solely through software upgrades or hardware upgrades of existing electronic devices. Furthermore, the various embodiments of the present invention described above can also be performed through an embedded server equipped in an electronic device or an external server of the electronic device.
[0156] Meanwhile, according to one embodiment of the present invention, the various embodiments described above may be implemented as software comprising instructions stored on a computer-readable recording medium using software, hardware, or a combination thereof. In some cases, the embodiments described herein may be implemented as the processor itself. According to the software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein.
[0157] Meanwhile, a computer or a similar device may include a device according to the disclosed embodiments, which is capable of calling instructions stored from a storage medium and operating according to the called instructions. When said instructions are executed by a processor, the processor may perform a function corresponding to said instructions directly or by using other components under the control of said processor. The instructions may include code generated or executed by a compiler or an interpreter.
[0158] A computer-readable recording medium may be provided in the form of a non-transitory computer-readable recording medium. Here, "non-transitory" simply means that the storage medium does not contain a signal and is tangible, without distinguishing whether data is stored semi-permanently or temporarily on the storage medium. In this context, a non-transitory computer-readable medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short moment, such as registers, caches, or memory. Specific examples of non-transitory computer-readable media may include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, and ROMs.
[0159] As described above, exemplary embodiments have been disclosed in the drawings and specification. Although specific terms have been used to describe the embodiments in this specification, they are used only for the purpose of explaining the technical concept of this disclosure and are not intended to limit the meaning or the scope of this disclosure as defined in the claims. Therefore, those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of this disclosure should be determined by the technical concept of the appended claims.
Claims
Claim 1 A memory storing at least one instruction for AI-based consumer activity data analysis and provision of customized rewards and partnership benefits linked via a consumer terminal and a communication network; and a processor that performs operations according to said instruction, wherein the processor collects activity data such as a consumer's purchase history, service usage history, mission completion status, review writing, and star rating evaluation, calculates a consumer grade based on said activity data, applies a differential reward payout rate according to a certain percentage of the store's net profit, stores and manages information on multiple industry group partners including finance, real estate, rental, legal / tax, travel, and telecommunications in conjunction with said differentially applied reward payout rates, recommends services, products, and events for each consumer through an artificial intelligence model using the consumer grade, activity data, and partner information as input values, and provides integrated output results of said recommendations, reward information, and partner location information, and the processor [represents] the consumer grade G using mathematical formula 2 Calculated through, where in Equation 2, S is the consumer activity score, θ0 to θ4 are boundary values defining the interval from Grade 1 to Grade 4, satisfying the condition θ0<θ1<θ2<θ3<θ4, .1{·} is an indicator function that returns 1 if the corresponding condition is true, and the above processor is Equation 3 The final reward amount Rt is calculated through [this], and in mathematical formula 3, NPt is the store's net profit, α is a pre-contract ratio within the range of 0.10 to 0.50, and βG is a grade-specific payout rate within the range of 0.60 to 0.90; the clip(x, a, b) function performs the function of limiting the calculated value to Rmin if it is smaller than the minimum limit Rmin, and to Rmax if it is larger than the maximum limit Rmax; the processor provides functions for searching for affiliates, requesting consultations, participating in communities, checking rewards, and withdrawing through the consumer terminal; the consumer grade is classified into four or more levels, and grade-specific reward payout rates are applied differentially within a certain range; the processor grants additional rewards based on the results of mission completion, review writing, and star rating evaluation included in the consumer's activity data; the processor determines whether the mission has been achieved, whether review quality standards have been met, and whether star rating standards have been met among the consumer's activity data as binary values, and whenever each item is satisfied, a pre-set fixed bonus is summed to [add] additional Calculates rewards, stores industry, location, reward rate, consultation availability, and real-time event information for each partner, filters and sorts and provides them according to a request from a consumer terminal, and the above processor provides additional rewards mathematical formula 5 Calculated through, and in the above mathematical formula 5, B0 is the basic additional bonus, Bm, Bq, and Bs are fixed bonus unit prices corresponding to each item of mission, review, and star rating, M is whether the mission has been completed, Q is whether the review quality standard has been met, S is whether the star rating standard has been met, and Rcap is the maximum payout limit for additional rewards per transaction, a fixed amount is added as an additional reward whenever each condition is met, and the final amount is limited so as not to exceed a preset maximum, and the above processor calculates a repurchase probability score for each affiliate or product based on the consumer's past usage behavior data mathematical formula 8 Calculated through, and in mathematical formula 8 is an affiliate or product It is the repurchase likelihood score, and is the normalized value of the recency indicator, calculated by dividing the number of days elapsed since the last purchase date by the reference range, and is a normalized value of the frequency indicator, calculated based on the number of purchases within a certain period, and is a normalized value of the monetary indicator, calculated based on the payment amount within a certain period, and is a normalized value of the reward usage pattern indicator, calculated based on the ratio of accumulated rewards used at the relevant affiliate or product, and , , and is the weight for each indicator, the sum of each weight is 1, each weight is pre-set, and calculated by the above mathematical formula 8 The closer the value is to 1, the higher the likelihood of repurchasing the relevant affiliate or product within a predefined period, and the above processor uses a comprehensive merchant evaluation index to quantitatively evaluate the service quality of affiliate merchants. mathematical formula 9 Calculated through, and in mathematical formula 9 is a comprehensive franchise evaluation index, and is a value obtained by converting the average consumer star rating into a range of 0 to 100 points, and is a value converted into a range of 0 to 100 points by standardizing the monthly review writing frequency, and is a service quality index calculated by combining response speed, repurchase rate, and claim processing rate, and in Equation 9, the weight for the average star rating is 0.4, the weight for the frequency of writing reviews is 0.3, the weight for the service quality index is 0.3, and the above processor is the calculated franchisee comprehensive evaluation index AI-based consumer activity data analysis and customized reward / partnership benefit provision device that classifies a merchant as an excellent merchant when the value is greater than or equal to a preset excellent judgment threshold. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete
Citation Information
Patent Citations
Blockchain-based community operating methods and systems for shopping mall and therefore
KR1020200137818A
