Method for detecting abusive transactions due to price fluctuation, information processing device, and computer-readable recording medium

Abstract

The present disclosure relates to a method for detecting an abusive transaction due to a price change, an information processing device, and a computer-readable recording medium. The method for detecting an abusive transaction includes acquiring price change information of a product and user behavior information associated with price change information of the product, determining, by an abusive transaction detection model based on price change information of the product and user behavior information, whether a first transaction corresponding to the user behavior information corresponds to an abusive transaction, and in response to determining that the first transaction corresponds to the abusive transaction, re-changing a changed price of the first transaction.

Description

Technical Field

[0001] This invention relates to a method for detecting unfair transactions based on price fluctuations, an information processing apparatus, and a computer-readable recording medium. Prior Technology

[0002] Recently, with the development of digital technology, online transactions have become increasingly active. Therefore, businesses are working on multiple fronts to ensure their market competitiveness in the e-commerce industry and maximize their profits.

[0003] As an example, price optimization techniques are needed to automatically adjust product prices based on market conditions. These techniques can utilize analytical data such as product cost, competitor pricing, consumer response, economic conditions, and supply and demand to formulate strategies and adjust prices.

[0004] In the application of this price optimization technique, data inconsistencies or human errors may lead to items being priced too low. This can be exploited for unfair transactions, such as purchasing large quantities of products at incorrectly set prices and reselling them, or buying high-priced products at extremely low prices and then accepting returns at the normal price to profit from the difference. Such unfair transactions can cause significant losses to businesses.

[0005] Therefore, in order to prevent unfair transactions that exploit price optimization techniques, technologies that can detect and respond to unfair transactions immediately are needed. Summary of the Invention

[0006] [The problem the invention aims to solve] To address the problems described above, the present invention provides a method for detecting unfair transactions based on price fluctuations, an information processing apparatus, and a computer-readable recording medium.

[0007] [Technical means to solve the problem] The present invention can be implemented in a variety of ways, including methods, apparatus (systems) and / or non-transitory computer-readable recording media containing computer-executable instructions.

[0008] According to an embodiment of the present invention, an unfair transaction detection method executed by at least one processor may include the following steps: acquiring product price change information and user behavior information related to the product price change information; determining, based on the product price change information and user behavior information, whether a first transaction corresponding to the user behavior information is equivalent to an unfair transaction using an unfair transaction detection model; and, in response to determining that the first transaction is an unfair transaction, changing the price of the first transaction again.

[0009] According to various embodiments of the present invention, the product price change information may include at least one of the following: the product price before the change, the current product price, the time of the price change, the current product cost price, or the current product price of competitors.

[0010] According to various embodiments of the present invention, the method for detecting improper transactions further includes the following steps: obtaining product-related information, which may include at least one of product identification information, product name, or product category.

[0011] According to various embodiments of the present invention, user behavior information may include at least one of product inquiry, order generation, payment, or return.

[0012] According to various embodiments of the present invention, the method for detecting improper transactions further includes the following steps: obtaining user information, which may include at least one of user identification information, user name, user email address, Internet Protocol address, device identification information, or user transaction history.

[0013] According to various embodiments of the present invention, the method for detecting improper transactions may further include the following steps: in response to determining that the first transaction is an improper transaction, indicating that the price change of the product corresponding to the first transaction is caused by an error.

[0014] According to various embodiments of the present invention, the method for detecting unfair transactions may further include the following steps: if the current price is lower than the product cost price, then the price is changed again based on the product's price change information.

[0015] According to various embodiments of the present invention, the step of detecting whether it is equivalent to unfair transaction may further include the following steps: if the difference between the current product price and the price of the current product from competing companies is within a specified range, it is determined that it is not an unfair transaction.

[0016] According to various embodiments of the present invention, the step of changing the price again may include the following steps: restoring the product price related to the first transaction to the existing price.

[0017] According to various embodiments of the present invention, the method for detecting improper transactions may further include the following step: generating an improper transaction detection model.

[0018] According to various embodiments of the present invention, the steps of generating an unfair transaction detection model may further include the following steps: obtaining pre-learning price change information of the product and pre-learning behavior information of the user; obtaining labeled data, wherein the labeled data serves as correct answer data and, based on the pre-learning price change information of the product, determines whether each transaction corresponding to the pre-learning behavior information of the user is equivalent to an unfair transaction; and pre-learning the model based on the pre-learning price change information of the product, the pre-learning behavior information of the user, and the labeled data.

[0019] According to various embodiments of the present invention, the step of obtaining the tagging data may further include the following steps: outputting the user's pre-learning behavior information and the product's pre-learning price change information; and receiving user input related to the abuse possibility score of each transaction.

[0020] According to various embodiments of the present invention, the step of obtaining tagging data may further include the following steps: outputting user pre-learning behavior information and product pre-learning price change information; and receiving user input related to whether each transaction is equivalent to an improper transaction.

[0021] An embodiment of the present invention provides a computer-readable recording medium that can store a computer program for executing an improper transaction detection method on a computer.

[0022] The information processing apparatus of various embodiments of the present invention includes: a communication module; a memory; and at least one processor connected to the memory for executing at least one computer-readable program included in the memory. The at least one program may include instructions to perform the following steps: acquiring product price change information and user behavior information related to the product price change information; determining, based on the product price change information and the user behavior information, whether a first transaction corresponding to the user behavior information is equivalent to an unfair transaction using an unfair transaction detection model; and, in response to determining that the first transaction is an unfair transaction, changing the price of the first transaction again.

[0023] [Comparison with the effectiveness of previous technologies] According to various embodiments of the present invention, in the process of determining whether an unfair transaction has occurred using an unfair transaction detection model, product price change information and user behavior information can be applied simultaneously. Therefore, unfair transactions caused by erroneous changes in product prices can be effectively detected.

[0024] According to various embodiments of the present invention, when determining whether a transaction is unfair, the cost price of the product and the prices of competing companies can be considered. Therefore, the occurrence of transactions that are conducted with normally permissible price fluctuations or competing companies' pricing strategies can be minimized.

[0025] According to various embodiments of the present invention, for transactions determined to be unfair transactions, the processor can change the product price again and provide a prompt that the price change is incorrect, thus effectively and immediately addressing unfair transactions.

[0026] The effects of this invention are not limited to those mentioned above. Those skilled in the art to which this invention pertains (hereinafter referred to as "skilled persons") can clearly understand other effects not mentioned through the description of the claims in the invention application. Simple Explanation of the Diagram

[0027] Hereinafter, several embodiments of the present invention will be described with reference to the accompanying drawings. In this process, similar reference numerals indicate similar structures, but are not limited thereto.

[0028] Figure 1 is an illustration of an improper transaction detection method according to an embodiment of the present invention.

[0029] Figure 2 is a schematic diagram of the connection structure of an information processing system according to an embodiment of the present invention for communicating with multiple user terminals in order to execute an improper transaction detection method.

[0030] Figure 3 is a block diagram illustrating the internal structure of a user terminal and information processing system according to an embodiment of the present invention.

[0031] Figure 4 is an example diagram of a tag user interface used to obtain tagging data during the pre-learning process of an improper transaction detection method according to an embodiment of the present invention.

[0032] Figure 5 is an illustrative flowchart illustrating the process of pre-learning an improper transaction detection model according to an embodiment of the present invention.

[0033] Figure 6 is an example diagram of the data used in the reasoning process of an improper transaction detection model according to an embodiment of the present invention.

[0034] Figure 7 is an example diagram illustrating a method for measuring product price change information according to an embodiment of the present invention.

[0035] Figure 8 is a block diagram of an improper transaction detection method and an information processing device for implementing it according to an embodiment of the present invention.

[0036] Figure 9 is a diagram illustrating the data types of an embodiment of the method for detecting improper transactions according to the present invention.

[0037] Figure 10 is an exemplary flowchart illustrating an embodiment of the improper transaction detection method of the present invention. Implementation

[0038] The following is a detailed description of the specific contents used to implement the present invention with reference to the accompanying drawings. However, in the following description, specific descriptions of well-known functions or structures will be omitted when there is a possibility that they may unnecessarily obscure the spirit of the present invention.

[0039] In the accompanying drawings, the same or corresponding structural elements are given the same reference numerals. Furthermore, in describing the following embodiments, repeated descriptions related to the same or corresponding structural elements will be omitted. However, even if the description of structural elements is omitted, this does not mean that such structural elements are not included in any embodiment.

[0040] The advantages, features, and implementation methods of the disclosed embodiments become clear with reference to the embodiments described in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and can be implemented through different methods. Furthermore, these embodiments are provided to ensure the completeness of this disclosure, and are only provided to enable those skilled in the art to fully understand the scope of the present invention.

[0041] In this specification, the terminology used will be briefly described and specific embodiments will be illustrated. While widely used terms have been selected as much as possible in consideration of the functionality of the invention, the terminology used herein may vary depending on the intent or convention of those skilled in the art, the emergence of new technologies, etc. Furthermore, in certain cases, terms arbitrarily chosen by the applicant may exist; in such cases, their meanings are detailed in the corresponding description of the invention. Therefore, the terminology used in this specification should be defined based on its meaning and the full text of this specification, and not limited to the name of the term.

[0042] In this specification, unless explicitly specified in the context as singular, singular expressions include plural expressions. Furthermore, unless explicitly specified in the context as plural, plural expressions include singular expressions. Throughout this specification, when indicating that a structural element includes any structural element, it means, unless specifically stated otherwise, that other structural elements may also be included, and not excluded.

[0043] Furthermore, in this specification, the terms "module" or "unit" as used refer to software structural elements or hardware structural elements that perform a certain function. However, "module" or "unit" is not limited to software or hardware. "Module" or "unit" may reside in accessible storage media, or may be reproduced on one or more processors. Thus, as an example, "module" or "unit" may include structural elements such as software structural elements, object-oriented software structural elements, class structural elements, and task structural elements, processes, functions, attributes, programs, subroutines, program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, lists, arrays, or variables. The functionality provided internally by structural elements and "modules" or "units" may be formed by combining fewer structural elements and "modules" or "units" or further separated into additional structural elements and "modules" or "units".

[0044] According to one embodiment of the present invention, a "module" or "unit" may be implemented by a processor and memory. The term "processor" is broadly interpreted to include general-purpose processors, central processing units (CPUs), microprocessors, digital signal processors (DSPs), controllers, microcontrollers, state machines, etc. In many contexts, "processor" may also refer to application-specific integrated circuits (ASICs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), etc. For example, "processor" may also refer to a combination of a digital signal processor and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a digital signal processing chip, or any other combination of the aforementioned structures. Furthermore, "memory" is broadly interpreted to include any electronic component capable of storing electronic information. "Memory" may also refer to various types of processor-readable media such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic or optical data storage devices, and registers. When the processor reads information from or writes information to memory, the memory and the processor are in electronic communication. The memory integrated into the processor is in a state of electronic communication with the processor.

[0045] In this specification, "system" may include at least one of server devices and cloud devices, but is not limited thereto. For example, a system may also consist of more than one server device. As another example, a system may consist of more than one cloud device. As yet another example, a system may be configured and operated by combining server devices and cloud devices.

[0046] In this specification, "user" refers to a user using an e-commerce platform or a user account of an application. A user account may refer to an account or related information generated and used by a user through an application.

[0047] In this specification, "display" means any display device associated with a computing device, for example, any display device controlled by or capable of displaying any information / data provided by the computing device.

[0048] In this specification, a "machine learning module" can include any module used to reason about an answer to a given input. According to one embodiment, a machine learning module can include an artificial neural network module, which includes an input layer, multiple hidden layers, and an output layer. Each layer can include multiple nodes. Although multiple machine learning modules are referred to as additional machine learning modules in this specification, this is not a limitation; a portion or all of multiple machine learning modules can be implemented by a single machine learning module. Furthermore, a single machine learning module can include multiple machine learning modules. In this specification, the terms "machine learning module" and "artificial neural network module" are used interchangeably to refer to the same or similar modules.

[0049] Hereinafter, several embodiments of the present invention will be described in detail with reference to the accompanying drawings. The process and related descriptions shown in the drawings are merely examples, and in some embodiments, they can be implemented in different ways. For example, one or more steps may be omitted, or the order of the steps may be changed, or one or more steps may be executed simultaneously, or one or more steps may be executed repeatedly.

[0050] In this specification, "e-commerce system" refers to a system that uses the internet or digital technology to trade goods or services online. Examples include public markets or independent shopping malls.

[0051] In this specification, "abusive transaction" refers to transactions occurring on e-commerce platforms, online platforms, or similar entities due to abnormal price fluctuations. For example, an abusive transaction could include a surge in product sales when prices are set abnormally low.

[0052] Figure 1 is an illustration of an improper transaction detection method according to an embodiment of the present invention.

[0053] Referring to FIG1, in response to inputting data 110 including product price change information 111 and user behavior information 112 into the unfair transaction detection model 120, an output value for determining whether a specific transaction is equivalent to an unfair transaction 130 is obtained.

[0054] An embodiment of the improper transaction detection method of the present invention can be executed by at least one processor (e.g., at least one processor of a cloud-based server, information processing system and / or information processing device).

[0055] Specifically, firstly, the processor can obtain product price change information 111 and user behavior information 112 related to the product price change information 111. For example, the product price change information 111 and user behavior information 112 can be obtained by the processor through monitoring. As another example, the processor can obtain product price change information 111 and user behavior information 112 from an external device. However, the present invention is not limited thereto.

[0056] Next, the processor inputs product price change information 111 and user behavior information 112 into the unfair transaction detection model 120 to determine a specific transaction. For example, it can output whether the first transaction is equivalent to an unfair transaction. The unfair transaction detection model 120 can represent a pre-trained machine learning module. The unfair transaction detection model 120 can be learned / trained using artificial neural network modules such as Large Language Models (LLMs) or open-source distributed modules; however, the unfair transaction detection model 120 is not limited to these.

[0057] In one embodiment, the product price change information 111 may include at least one of the following: the product price before the change, the current product price, the time of the price change, the current product cost price, and the current product competitor's price. For example, the unfair transaction detection model 120 may consider the difference between the product price before the change and the current product price, the comparison between the current product cost price and the price, the comparison between the current product price and competitor's price information, the time of the price change, etc., to determine whether it is equivalent to an unfair transaction 130.

[0058] Furthermore, the product price change information 111 may also include product-related information. For example, product-related information may include at least one of product identification information, product name, and product category. Since the appropriate range of price fluctuations for a product or the extent of losses to a company caused by unfair transactions can vary depending on the product attributes, product-related information is one of the factors that need to be considered in the process of detecting unfair transactions.

[0059] In one embodiment, user behavior information 112 may include at least one of product inquiry, order generation, payment, or return. For example, an unfair transaction detection model 120 may learn to infer whether an unfair transaction 130 is constituted by matching user behavior information 112 with product price change information 111.

[0060] In one embodiment, in addition to product price change information 111 and user behavior information 112, user-related information (i.e., user information) can be obtained. The obtained user information can be input into the unfair transaction detection model 120. For example, user information may include at least one of user identification information, user name, user email address, Internet Protocol address, device identification information, and user transaction history. The unfair transaction detection model can learn to apply user information to specify users who have engaged in unfair transaction behavior and identify users with a high probability of engaging in unfair transactions through user transaction history.

[0061] In one embodiment, in response to the determination that a first transaction is an unfair transaction, the processor may again change the price of the first transaction. For example, changing the price again means changing the product price again after implementing the first price change related to the first transaction determined to be an unfair transaction. The processor's second price change may include, but is not limited to, restoring the product price to the product price before the first price change. For example, changing the price again may include at least one of the following: taking into account that the competitor's price has changed to an appropriate price within a specified range of the competitor's price; changing the product price to the average of the product prices over a specific period before the first price change; or calculating the price using an unfair transaction detection model.

[0062] In one embodiment, when the first transaction is determined to be an improper transaction, the processor may respond by indicating that the first price change related to the first transaction was caused by an error. Specifically, the indication of the first price change being caused by an error may include notifying the system administrator of the price change error via a pop-up window or by sounding an alert, but is not limited to these methods.

[0063] Figure 2 is a schematic diagram of the connection structure of an information processing system according to an embodiment of the present invention for communicating with multiple user terminals in order to execute an improper transaction detection method.

[0064] Referring to Figure 2, in one embodiment, the information processing system 230 may include a computer-executable program (e.g., a downloadable application or an accessible unfair transaction detection model) related to the unfair transaction detection method, and one or more server devices and / or databases or one or more distributed computing devices and / or distributed databases based on cloud computing services capable of storing, providing, and executing data. For example, the information processing system 230 may include additional systems (e.g., servers) for executing the unfair transaction detection method.

[0065] Multiple user terminals 210_1, 210_2, and 210_3 can communicate with the information processing system 230 via network 220. Network 220 enables communication between multiple user terminals 210_1, 210_2, and 210_3 and the information processing system 230. For example, network 220 can be implemented using wired networks such as Ethernet, Power Line Communication, telephone line communication devices, and RS-serial communication, as well as mobile communication networks, wireless local area networks (WLANs), Wi-Fi, Bluetooth, and ZigBee, or combinations thereof, depending on the setup environment. The communication method is not limited; in addition to communication networks that network 220 can include (e.g., mobile communication networks, wired internet, wireless internet, broadcast networks, satellite networks, etc.), it can also include short-range wireless communication between user terminals 210_1, 210_2, and 210_3.

[0066] In Figure 2, mobile phone terminal 210_1, tablet terminal 210_2, and personal computer (PC) terminal 210_3 are shown as examples of user terminals, but the model is not limited to these. User terminals 210_1, 210_2, and 210_3 can be any computing device capable of wired and / or wireless communication and capable of executing or installing applications. For example, user terminals may include smartphones, mobile phones, navigators, computers, laptops, digital broadcasting terminals, personal digital assistants (PDAs), portable multimedia players (PMPs), tablet computers, game consoles, wearable devices, Internet of Things (IoT) devices, virtual reality (VR) devices, augmented reality (AR) devices, etc. Furthermore, although Figure 2 shows three user terminals 210_1, 210_2, and 210_3 communicating with the information processing system 230 via network 220, it is not limited to this, and the number of user terminals communicating with the information processing system 230 via network 220 can also be different.

[0067] In one embodiment, user terminals 210_1, 210_2, and 210_3 can receive information or data from other user terminals via network 220, or send information or data to other user terminals.

[0068] In Figure 2, although the information processing system 230 is a separate device from the user terminals 210_1, 210_2, and 210_3, it is not limited to this. The information processing system 230 can also be integrated with the user terminals 210_1, 210_2, and 210_3.

[0069] Figure 3 is a block diagram illustrating the internal structure of a user terminal and information processing system according to an embodiment of the present invention.

[0070] Referring to Figure 3, the user terminal 210 can also be referred to as any computing device capable of executing applications, web browsers, or wired / wireless communication. For example, it may include the mobile phone terminal 210_1, tablet terminal 210_2, PC terminal 210_3, etc., as shown in Figure 2. As shown, the user terminal 210 may include memory 312, processor 314, communication module 316, and input / output interface 318. Similarly, the information processing system 230 may include memory 332, processor 334, communication module 336, and input / output interface 338. As shown in Figure 3, the user terminal 210 and the information processing system 230 can send and receive information and / or data via network 220 using the respective communication modules 316 and 336. Furthermore, the input / output device 320 can input information and / or data to the user terminal 210 through the input / output interface 318, or output information and / or data generated by the user terminal 210.

[0071] Memory modules 312 and 332 may include any non-transitory computer-readable recording medium. According to one embodiment, memory modules 312 and 332 may include non-volatile mass storage devices such as read-only memory (ROM), disk drives, solid-state drives (SSDs), or flash memory. As another example, non-volatile mass storage devices such as read-only memory, solid-state drives, flash memory, and disk drives, as additional permanent storage devices different from memory modules, may be included within the user terminal 210 or the information processing system 230. Furthermore, memory modules 312 and 332 may store an operating system and at least one program code.

[0072] Such software structural elements can be loaded from an additional computer-readable recording medium other than the memories 312 and 332. This additional computer-readable recording medium may include a recording medium that can be directly connected to the user terminal 210 and the information processing system 230, such as a floppy disk drive, magnetic disk, magnetic tape, DVD / CD-ROM drive, and / or memory card. As another example, the software structural elements may also be loaded into the memories 312 and 332 via communication modules 316 and 336, instead of a computer-readable recording medium. For example, at least one program may be loaded into the memories 312 and 332 based on a file-installed computer program provided by a developer or a file distribution system that distributes application installation files via network 220.

[0073] In one embodiment, the memory 332 of the information processing system 230 may store an improper transaction detection model and a computer program for executing the improper transaction detection method. The data stored in the memory 332 can be accessed by the processor 334 of the information processing system 230.

[0074] Processors 314 and 334 can process computer program instructions by performing basic arithmetic, logic, and input / output operations. Instructions can be provided to processors 314 and 334 via memory 312 and 332 or communication modules 316 and 336. For example, processors 314 and 334 can execute instructions received based on program code stored in recording devices such as memory 312 and 332. As a specific example, processor 334 can access an improper transaction detection model and a computer program for executing improper transaction detection methods stored in memory 332.

[0075] Communication modules 316 and 336 can provide a structure or function that enables user terminal 210 and information processing system 230 to communicate with each other via network 220, and can provide a structure or function that enables user terminal 210 and / or information processing system 230 to communicate with other user terminals or other systems (e.g., additional cloud systems, etc.). As an example, under the control of communication module 316, requests or data generated by processor 314 of user terminal 210 based on program code stored in a recording device such as memory 312 can be transmitted to information processing system 230 via network 220. Conversely, user terminal 210 can receive control information or instructions provided by processor 334 of information processing system 230 via communication module 316 through communication module 336 and network 220.

[0076] Input / output interface 318 refers to a device used to connect input / output device 320. As an example, input devices may include devices with audio sensors and / or image sensors, such as cameras, keyboards, microphones, and mice; and output devices may include displays, speakers, haptic feedback devices, etc. As another example, input / output interface 318 can be an interface device for a device that integrates input and output functions, such as a touchscreen. For example, as the processor 314 of user terminal 210 executes computer program instructions loaded in memory 312, service screens and other data based on information and / or data provided by information processing system 230 or other user terminals can be displayed on the monitor through input / output interface 318. In Figure 3, although input / output device 320 is not included in user terminal 210, it is not limited to this and can also be integrated with user terminal 210 as a single device. Furthermore, the input / output interface 338 of the information processing system 230 can be connected to the information processing system 230, or it can be an interface device of any input / output device (not shown) that the information processing system 230 may include. In FIG3, although the input / output interfaces 318 and 338 are separated from the processors 314 and 334 as separate structural elements, this is not a limitation, and the input / output interfaces 318 and 338 may also be included within the processors 314 and 334.

[0077] Compared to the structural elements in Figure 3, the user terminal 210 and information processing system 230 may include more structural elements. However, it is not necessary to explicitly show most of the existing technology structures. In one embodiment, the user terminal 210 may include at least a portion of the input / output devices 320. Furthermore, the user terminal 210 may also include other structural elements such as a transceiver, a Global Positioning System (GPS) module, a camera, various sensors, and a database. For example, if the user terminal 210 is a smartphone, it may include structural elements typically found in smartphones, such as an accelerometer, a gyroscope, a microphone module, a camera module, various physical buttons, buttons using a touchpad, output / output ports, and a vibrator for vibration.

[0078] During the operation of a program used for applications, processor 314 can input or receive selected text, images, videos, voice and / or work through input devices such as touch screens, keyboards, cameras with audio sensors and / or image sensors, and microphones connected to input / output interface 318. It can store the received text, images, videos, voice and / or work in memory 312 or provide it to information processing system 230 through communication module 316 and network 220.

[0079] The processor 314 of user terminal 210 can manage, process, and / or store information and / or data received from input / output device 320, other user terminals, information processing system 230, and / or multiple external systems. The information and / or data processed by processor 314 can be provided to information processing system 230 via communication module 316 and network 220. The processor 314 of user terminal 210 can transmit and output information and / or data to input / output device 320 via input / output interface 318. For example, processor 314 can output or display the received information and / or data to the screen of user terminal 210.

[0080] The processor 334 of the information processing system 230 can manage, process, and / or store information and / or data received from multiple user terminals 210 and / or multiple external systems. The information and / or data processed by the processor 334 can be provided to the user terminals 210 through the communication module 336 and the network 220.

[0081] Figure 4 is an example diagram of a tag user interface used to obtain tagging data during the pre-learning process of an improper transaction detection method according to an embodiment of the present invention.

[0082] Referring to Figure 4, in one embodiment, the processor can pre-learn an unfair transaction detection model 120. During the pre-learning process, the processor can acquire labeled data 430. Labeled data 430 refers to data that labels user interface users (e.g., administrators) assign unfair transaction probability scores or determine whether a transaction constitutes an unfair transaction based on price change identification information, transaction identification information, etc. Labeled data 430 also refers to the correct answer data based on the product's pre-learned price change information to determine whether each transaction corresponding to the user's pre-learned behavioral information constitutes an unfair transaction.

[0083] The processor may provide a tag user interface 420 to the tag user interface user 410 in order to obtain tag data 430. The tag user interface user 410 may use the tag user interface 420 output by the processor to input the tag data 430 into the first input field 425, such as an unfair transaction probability score. However, the tag data 430 may have other forms besides an unfair transaction probability score. For example, the tag data 430 may include the tag user interface user 410's judgment value, which is related to whether it is equivalent to an unfair transaction.

[0084] More specifically, the processor displays pre-learning data, such as user pre-learning behavior information and product pre-learning price change information, in the tag user interface 420. According to the example in Figure 4, the tag user interface 420 may include a first area 423, a second area 424, and a first input field 425. The first area 423 of the tag user interface 420 may include user behavior information. For example, user behavior information may include at least one of the following related information: the number of product queries 423a, whether an order was generated 423b, whether payment was made 423c, and whether a return was made 423d, but is not limited to this.

[0085] The second area 424 of the tag user interface 420 may include product price change information. This product price change information may be related to user behavior information. For example, the product price change information may include at least one of the following: the price before the price change 424a, the price after the price change 424b, the time of the price change 424c, the cost price at the time of the price change 424d, and competitor pricing information 424e at the time of the price change, but is not limited to this. On the other hand, the type and configuration of information displayed in the tag user interface 420 may be implemented in various ways, but is not limited to this.

[0086] In the first input field 425 of the tag user interface 420, an unfair transaction probability score, which quantifies the likelihood that a specific transaction (e.g., the first transaction) related to the price change information of the currently displayed product and user behavior information is equivalent to an unfair transaction, can be input by the tag user interface user 410. The tag user interface user 410 can receive input related to the unfair transaction probability score of each transaction. For example, the higher the unfair transaction probability score, the higher the probability that a specific transaction is equivalent to an unfair transaction. If the unfair transaction probability score is higher than a threshold, the corresponding transaction can be judged as an unfair transaction. Regarding the first transaction, Figure 4 shows that the unfair transaction probability score input by the tag user interface user 410 into the first input field 425 is 80 points. As an example, if the threshold is 60 points, then the unfair transaction probability score of the first transaction is 80 points, which is higher than the threshold, and therefore, it can be judged as an unfair transaction.

[0087] In another embodiment, the processor may receive a judgment value used to determine whether a specific transaction is equivalent to an improper transaction as input to the tag user interface user 410. The processor may receive input from the tag user interface user 410 used to determine whether each transaction is equivalent to an improper transaction. For example, if the tag user interface user 410 determines that a specific transaction is an improper transaction, it inputs "○"; if it determines that a specific transaction is not an improper transaction, it can input "×" as the judgment value.

[0088] In one embodiment, the tab user interface 420 includes buttons for switching screens to provide tab user interfaces related to multiple transactions. For example, it may include a first screen switching button 421 and a second screen switching button 422. In response to receiving a user's selection of the first screen switching button 421 or the second screen switching button 422, a screen for the currently displayed transaction and other transactions can be output.

[0089] In one embodiment, the processor may store the generated tagging data 430 in a storage medium including internal memory, external storage devices, etc.

[0090] Figure 5 is an illustrative flowchart illustrating the process of pre-learning an improper transaction detection model according to an embodiment of the present invention.

[0091] Referring to Figure 5, in one embodiment, the processor can pre-learn an unfair transaction detection model 120. During the pre-learning process of the unfair transaction detection model 120, the processor can acquire pre-learning price change information 510 for products related to each transaction and pre-learning behavioral information 520 for users. Furthermore, the processor can acquire tagging data (e.g., tagging data 430 in Figure 4) 430 corresponding to each transaction. That is, the unfair transaction detection model can be pre-learned based on the pre-learning price change information 510 for products, the pre-learning behavioral information 520 for users, and the tagging data 430.

[0092] In one embodiment, in relation to each transaction, if the processor inputs pre-learning price change information 510 for products and pre-learning behavior information 520 for users into the unfair transaction detection model 120, the processor can calculate a value (i.e., a predicted value 530) for determining whether each transaction is equivalent to an unfair transaction. Next, the processor can calculate an error value 540 representing the error between the predicted value 530 and the label data 430 of the corresponding transaction. For example, the error value 540 may include mean squared error, mean absolute error, etc., but is not limited to these. The calculated error value 540 can be backpropagated to the unfair transaction detection model 120. The processor can generate a pre-learned unfair transaction detection model 120 by repeatedly performing this learning process in the direction of minimizing the error value 540.

[0093] However, as mentioned above, the pre-learning method for the improper transaction detection model 120 is merely an example, and various pre-learning methods can be performed. Furthermore, reinforcement learning can also be applied in addition to pre-learning.

[0094] Figure 6 is an example diagram of the data used in the reasoning process of an improper transaction detection model according to an embodiment of the present invention.

[0095] Referring to Figure 6, in one embodiment, the processor receives product price change information (e.g., 111 in Figure 1), user behavior information (e.g., 112 in Figure 1), and user information, which can be stored in memory in a table format as shown in Figure 6. The left side 610 of the table includes user pre-learning behavior information, and the right side 620 may include product price change information. However, the table in Figure 6 is only an example, and its result and arrangement are not limited thereto.

[0096] In one embodiment, user behavior information may include platform data 611 and order identification information 612. Platform data 611 refers to the operating system or execution environment of the terminal device used when purchasing a product or utilizing a service. For example, platform data 611 may include one of Android, iOS, or Windows. Order identification information 612 refers to a fixed identification value assigned to each order. Its form is not limited to numbers and can be a unique identification character composed of numbers, letters, or a combination thereof. Order identification information 612 can be used as standard information required to specify order details within the system.

[0097] In one embodiment, the processor may receive product identification information 621 as part of the product pre-learning price change information 510. Specifically, product identification information 621 refers to a fixed identification value assigned to each product. Its form is not limited to numbers and can be an inherent identification character composed of numbers, letters, or combinations thereof. Product identification information 621 can be used as standard information required to specify products within the system.

[0098] User information may include user identification information 613. User identification information 613 refers to the identification value fixedly assigned to each user. Its form is not limited to numbers; it can be a unique identifier composed of numbers, letters, or a combination thereof. User identification information 613 can be used as standard information required to specify users within the system. In one example, user identification information 613 can be linked to a separate database storing user transaction history. Therefore, an improper transaction detection model can query user transaction history through user identification information 613.

[0099] Product price change information may include at least one of the following: product price 622, product category 623, order quantity 624, specific event occurrence time 625, price before price change 626, or price change time 627. Product price 622 refers to the current price (the price after the price change), and product category 623 refers to the group or category to which the product belongs. Product category 623 is represented by text, but is not limited to this; it may be represented by numbers, text, or a combination thereof. Order quantity 624 refers to the order quantity of a specific product ordered by a user based on product price 622. Specific event occurrence time 625 refers to the time when a specific event (e.g., a product order) occurs in the processor. Price before price change 626 refers to the price before the current product price (product price 622) changed, and price change time 627 refers to the time when the product price changed. As shown in Figure 6, event occurrence time 625 and price change time 627 may be displayed as "year / month / day / time:minute:second," but are not limited to this.

[0100] Figure 7 is an example diagram illustrating a method for measuring product price change information according to an embodiment of the present invention.

[0101] Referring to FIG7, in one embodiment, the processor may receive product price change information (e.g., 111 in FIG1) and user behavior information (e.g., 112 in FIG1). The processor may receive the product price change information 111 and the user behavior information 112 immediately at 701, or execute them within a specific period 702, or execute them only when a specific event occurs at 703, or execute them at predetermined time intervals, or execute them in combination thereof.

[0102] As described above, by inputting product price change information 111 and user behavior information 112 into the unfair transaction detection model, the processor can monitor whether unfair transactions have occurred. For example, if there is a surge in product order volume, a surge in product return volume, a surge in product inquiry volume, or a product price lower than the cost price, the unfair transaction detection model can determine whether a specific transaction is equivalent to an unfair transaction.

[0103] Therefore, according to embodiments of the present invention, an improper transaction detection model can be used to automatically determine whether an improper transaction has occurred. Compared with the situation where managers manually identify improper transactions, improper transactions can be dealt with immediately and effectively.

[0104] Figure 8 is a block diagram of an improper transaction detection method and an information processing device for implementing it according to an embodiment of the present invention.

[0105] Referring to Figure 8, in one embodiment, the transaction system 827 provided by the e-commerce platform can receive input related to user transactions, enabling user 810 to execute transactions for products, etc. If user 810 uses the transaction system 827 to conduct a transaction, the processor can collect 824 related data. For example, the collected data may include product price change information and user behavior information related to the product price change information.

[0106] The collected data 824 can be input into the unfair transaction detection model 120 in the form of real-time feature data (e.g., a specific vector). That is, the unfair transaction detection model 120 receives product price change information and user behavior information converted into real-time feature data, and can determine whether the first transaction corresponding to the user behavior information is equivalent to an unfair transaction based on this. Moreover, in response to the determination that the first transaction is an abuse, the product price, such as the price of the first transaction, can be changed again, as determined 826. This product price determination 826 is performed by the unfair transaction detection model 120, or by an additional module, or can be implemented manually by the administrator; the method is not limited to these.

[0107] On the other hand, the collected data 824 can also be used for pre-learning. In this case, if user 810 executes a transaction using transaction system 827, the processor collects data 824 related to the corresponding transaction (e.g., product price change information and user behavior information), and can store the collected data 824 in database 822. Moreover, by receiving tag data from tag user interface user 410 through tag user interface 821, the processor can store the received tag data in database 822.

[0108] In one embodiment, the processor performs model learning 823 of the unfair transaction detection model 120 based on collected data 824 and labeled data stored in the database 822, for example, pre-learning. This pre-learning corresponds to the pre-learning illustrated with reference to FIG5. However, model learning 823 is not limited to pre-learning; for example, it can be performed through various methods such as reinforcement learning. Subsequently, the processor can generate a pre-learned unfair transaction detection model 120a through model learning 823. The pre-learned unfair transaction detection model 120a can be distributed in the transaction system 827 provided by the e-commerce platform and used as the unfair transaction detection model 120.

[0109] Figure 9 is a diagram illustrating the data types of an embodiment of the method for detecting improper transactions according to the present invention.

[0110] Referring to Figure 9, in one embodiment, the processor may acquire product price change information 111. In this case, the product price change information 111 may include at least one of the following: the product's price before the price change 911, the current product price 912, the time of the price change 913, the current product cost price 914, and the current product competitor's price 915.

[0111] As an example, the processor can use information related to the price change deviation to compare the time of the price change 913 with the time of the specific transaction by using the price before the price change 911 of the product. It can also use the correlation between the price change and the user's transaction behavior.

[0112] Furthermore, by comparing the current product cost price 914 with the current product price 912, if the current product price 912 is lower, the processor can directly determine it as an unfair transaction and take emergency measures. The processor can also consider the current product competitor's price 915 to determine whether the current product price 912 is correctly determined. Based on this structure, the possibility of mistakenly identifying transactions based on normally permissible price changes or competitor pricing strategies as unfair transactions can be minimized.

[0113] In one embodiment, the product price change information 111 may further include product-related information (i.e., product information 916). For example, product-related information may include at least one of product identification information 916a, product name 916b, and product category 916c. Since the appropriate price range for a product or the extent of loss to a company due to unfair transactions can vary depending on product attributes, product information 916 is one of the factors that need to be considered during the detection of unfair transactions. As an example, compared to expensive home appliances, inexpensive stationery may cause relatively less loss to a company due to incorrectly determined prices. When detecting whether an unfair transaction has occurred, the processor may consider the characteristics of such products.

[0114] In one embodiment, user behavior information 112 may include at least one of product inquiry 921, order generation 922, payment 923, and return 924.

[0115] In one embodiment, user behavior information 112 may include user-related information (i.e., user information 925). User information 925 may include at least one of user identification information 925a, user name 925b, user email address 925c, Internet Protocol address 925d, device identification information 925e, and user transaction history 925f. As an example, user information is used to specify users who have engaged in improper transaction behavior, and can be used to identify users with a high probability of engaging in improper transactions through user transaction history.

[0116] In one embodiment, the processor may receive labeled data 430 during the learning process of the unfair transaction detection model 120. In this case, the labeled data 430 may include at least one of price change identification information 931, an unfair transaction probability score 932, and whether a transaction is considered unfair 933. The price change identification information 931 can be used to identify the specific transaction that was labeled. The unfair transaction probability score 932 can be used as an indicator of the probability that a specific transaction is considered unfair. The whether a transaction is considered unfair 933 can be used as data indicating whether a specific transaction is considered unfair or whether it is ambiguous. The whether a transaction is considered unfair 933 can be expressed in words, but is not limited to this, and may also be expressed in numbers or a combination thereof.

[0117] In one embodiment, when acquiring product price change information 111 and user behavior information 112, the processor can receive this information in various ways, and is not limited to a specific method. Specifically, the product price change information 111 and user behavior information 112 may have at least one format selected from Excel format, ASCII file format, CSV format, and XML format.

[0118] Figure 10 is an illustrative flowchart illustrating an embodiment of the improper transaction detection method of the present invention. Figure 10 illustratively illustrates a series of processes in which the improper transaction detection method of the present invention detects improper transactions and takes subsequent measures through pre-learning of an improper transaction detection model.

[0119] Referring to Figure 10, firstly, the processor can obtain the product's pre-learning price change information and the user's pre-learning behavior information and store them in the database (step 1010).

[0120] Next, the processor can obtain the probability of each transaction being an improper transaction or the marking data indicating whether it is an improper transaction through a pre-learning step (step 1020). For example, similar to the description with reference to FIG4, the user can obtain the marking data by receiving input values ​​through the tag user interface.

[0121] Furthermore, the processor can pre-learn an unfair transaction detection model based on the acquired pre-learning data (e.g., pre-learning price change information and user pre-learning behavior information) and labeled data 430 (step 1030). This allows the generation of a pre-learned unfair transaction detection model.

[0122] Subsequently, in the reasoning step, the processor can obtain price change information of the product for a specific transaction (e.g., the first transaction) and user behavior information related to the price change information of the product (step 1040).

[0123] Then, the processor inputs the acquired product price change information and user behavior information into the pre-learned unfair transaction detection model. The unfair transaction detection model can determine whether a specific transaction (e.g., the first transaction) corresponding to the user behavior information is equivalent to an unfair transaction based on the input product price change information and user behavior information.

[0124] In one embodiment, the unfair transaction detection model can calculate and output an unfair transaction probability score based on product price change information and user behavior information (step 1050). Furthermore, it can determine whether a specific transaction (e.g., the first transaction) is equivalent to an unfair transaction based on the output unfair transaction probability score (step 1050a). Specifically, if the unfair transaction probability score output by the unfair transaction detection model 120 is above a threshold, the processor can determine that the corresponding transaction is an unfair transaction. The benchmark value can be a predetermined value or can be adjusted and set by the operating system administrator according to operational objectives and strategies.

[0125] In one embodiment, after determining whether a specific transaction is equivalent to an unfair transaction, the processor may perform follow-up measures for the specific transaction determined to be an unfair transaction. For example, the follow-up measures may include at least one of marking the price change of the product corresponding to the transaction determined to be an unfair transaction (e.g., the first transaction) as an error 1060a, changing the price of the product again 1060b, and indicating the product price change error 1060c.

[0126] On the other hand, additionally or alternatively, in the process of the processor determining whether a particular transaction is equivalent to an unfair transaction, before applying the unfair transaction detection model, a step 1040a of comparing the product cost price with the current product price and a step 1040b of comparing the product competitor's price with the current product price may also be included.

[0127] Specifically, step 1040a, which compares the product cost price with the current product price, may include the following steps: if the current price is lower than the product cost price, the corresponding transaction is determined to be an unfair transaction, without needing to separately execute the unfair transaction detection model's judgment process. Furthermore, step 1040b, which compares the product competitor's price with the current product price, may include the following steps: only if the difference between the current product competitor's price 915 and the current product price 912 is above a predetermined range, is the judgment 1050 applying the unfair transaction detection model executed. That is, by comparing the current product competitor's price 915 with the current product price 912, if the difference is within a predetermined range, the corresponding transaction can be determined not to be an unfair transaction. As described above, by considering the product cost price and competitor's price and judging whether a transaction is unfair based on predetermined rules, the resource consumption caused by executing the unfair transaction detection model can be reduced and the judgment speed improved.

[0128] To enable a computer to execute the method, it can be provided by a computer program stored on a computer-readable recording medium. The medium can also be temporary storage for continuously storing, executing, or downloading a computer executable program. Furthermore, the medium can be a combination of various recording or storage units composed of one or more hardware components, and is not limited to media that directly access any computer system; it can also be distributed across a network. As examples, the medium includes magnetic media such as hard drives, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floppy disks; and structures storing program instructions such as ROM, RAM, and flash memory. Furthermore, as another example, the medium can be a recording or storage medium managed by an application store that sells applications, or a website or server that provides and sells various other software.

[0129] The methods, operations, or techniques of this invention can be implemented by various units. For example, the techniques can also be implemented by hardware, firmware, software, or a combination thereof. In relation to the disclosure of this invention, those skilled in the art should understand that the various exemplary logic blocks, modules, circuits, and algorithm steps described can also be implemented by electronic hardware, computer software, or a combination thereof. To clearly illustrate this interchangeability between hardware and software, various exemplary structural elements, blocks, modules, circuits, and steps have been simply described above from a functional perspective. However, whether such functionality is implemented by hardware or software depends on the design requirements of the specific application and the overall system. Those skilled in the art can also implement the described functions in various ways for each specific application; therefore, such implementations should not be construed as departing from the scope of this invention.

[0130] In hardware instances, the processing unit for executing the technology may also be implemented within one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), graphics processing units (GPUs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions of the present invention, computers, or combinations thereof.

[0131] Therefore, the various exemplary logic blocks, modules, and circuits described in conjunction with this invention can also be implemented or executed by a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete logic gates or transistor logic, discrete hardware components, or any combination designed to perform the functions of this invention. The general-purpose processor can be a microprocessor; alternatively, the processor can also be any processor, controller, microcontroller, or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor, a microprocessor, multiple microprocessors, one or more microprocessors combined with a digital signal processing chip, or any other combination of structures.

[0132] In firmware and / or software instances, the technology may also be implemented by instructions stored on computer-readable media such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, compact disc (CD), magnetic or optical data storage devices. The instructions may be executed by more than one processor, or may cause the processor to perform the functions described in this invention according to a specific implementation.

[0133] When implemented in software, the technology can be used to store or transmit one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, including any medium that facilitates the transfer of computer programs from one location to another. Storage media can also be any available medium accessible to a computer. As a non-limiting example, such computer-readable media may include random access memory, read-only memory, electrically erasable programmable read-only memory, CD-ROM or other optical disc storage, magnetic disk storage or other magnetic storage devices, or any other computer-accessible medium used to transfer or store prescribed program code in the form of instructions or data structures. Furthermore, any access may be appropriately referred to as a computer-readable medium.

[0134] For example, if the software uses coaxial cable, fiber optic cable, stranded wire, digital subscriber line (DSL), or wireless technologies such as infrared, wireless, and microwave to transmit the software from a web browser, server, or other remote source, then coaxial cable, fiber optic cable, stranded wire, digital subscriber line (DSL), or wireless technologies such as infrared, wireless, and microwave are included within the definition of media. In this application, the terms "disk" and "disc" include CDs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically reproduce magnetic data, and discs, conversely, utilize lasers to reproduce optical data. The combinations described are also included within the scope of computer-readable media.

[0135] The software module can reside within random access memory, flash memory, read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, cache, hard disk, removable disk, CD-ROM, or any known form of storage medium. As an example, the storage medium is connected to a processor so that the processor reads information from or records information to the storage medium. Alternatively, the storage medium can be integrated into the processor. The processor and storage medium can also reside within a dedicated integrated circuit. The dedicated integrated circuit can also reside within a user terminal. Alternatively, the processor and storage medium can also exist as separate structural elements within the user terminal.

[0136] In the embodiments described above, although the currently disclosed main implementation method can be applied through one or more independent computer systems, the present invention is not limited thereto, and can also be implemented in any computing environment such as a network or a distributed computing environment. Furthermore, the main implementation method of the present invention can also be implemented by multiple processing chips or devices, and storage can be similarly affected by multiple devices. Such devices may also include personal computers, network servers, and portable devices.

[0137] While this specification describes content related to certain embodiments, those skilled in the art can make various modifications and alterations without departing from the scope of this invention. Furthermore, such modifications and alterations also fall within the scope of the appended patent applications.

[0138] 110: Enter data 111: Product Price Change Information 112: User Behavior Information 120: Improper Transaction Detection Model 130: Does this constitute an unfair transaction?

Claims

1. A method for detecting abusive transactions, executed by at least one processor, for detecting abusive transactions on an e-commerce system, wherein, The method includes the following steps: obtaining product price change information and user behavior information related to the product price change information; generating an unfair transaction detection model, which includes the following steps: obtaining labeled data, which serves as correct answer data and is related to determining whether each transaction corresponding to the user's pre-learning behavior information is equivalent to an unfair transaction based on the product's pre-learning price change information, wherein the step of obtaining the labeled data further includes the following steps: outputting the user's pre-learning behavior information and the product's pre-learning price change information; and receiving user input related to the unfair transaction possibility score of each transaction, or user input related to whether each transaction is equivalent to an unfair transaction; using the unfair transaction detection model, determining whether a first transaction corresponding to the user behavior information is equivalent to an unfair transaction based on the product price change information and the user behavior information; and responding to the determination that the first transaction is an unfair transaction, changing the price of the first transaction again.

2. As in Request 1, the method for detecting improper transactions, wherein, The product price change information includes at least one of the following: the product price before the change, the current product price, the time of the price change, the current product cost price, or the current product price from competitors.

3. As in Request 2, the method for detecting improper transactions, wherein, The method further includes the following steps: obtaining product-related information, which includes at least one of product identification information, product name, or product category.

4. As in Request 1, the method for detecting improper transactions, wherein, The user behavior information includes at least one of product inquiry, order generation, payment, or return.

5. As in request item 4, the method for detecting improper transactions, wherein, The method further includes the following steps: obtaining user information, wherein the user information includes at least one of user identification information, user name, user email address, Internet Protocol address, device identification information, or user transaction history.

6. As in Request 1, the method for detecting improper transactions, wherein, The method further includes the following steps: in response to the determination that the first transaction is an improper transaction, indicating that the price change of the product corresponding to the first transaction is caused by an error.

7. The method for detecting improper transactions as described in Request 1, wherein, The method further includes the following steps: if the current price is lower than the product cost price, then the price is changed again based on the product's price change information.

8. The method for detecting improper transactions as described in Request 1, wherein, The step of detecting whether it is equivalent to the unfair transaction further includes the following steps: if the difference between the current product price and the price of the current product from the competing company is within a specified range, it is determined that it is not an unfair transaction.

9. The method for detecting improper transactions as described in Request 1, wherein, The step of changing the price again includes the following steps: restoring the product price related to the first transaction to the existing price.

10. The method for detecting improper transactions as described in Request 1, wherein, The steps for generating the unfair transaction detection model further include the following steps: obtaining pre-learning price change information of the product and pre-learning behavior information of the user; and pre-learning model based on the pre-learning price change information of the product, the pre-learning behavior information of the user, and the labeled data.

11. A computer-readable recording medium, wherein, The computer program contains a method for performing any of the requests 1 to 10 on the computer.

12. An information processing apparatus, wherein, include: Communication module; Memory; The system also includes at least one processor connected to the memory for executing at least one computer-readable program included in the memory, the at least one program including instructions to perform the following steps: acquiring price change information of a product and user behavior information related to the price change information of the product; generating an unfair transaction detection model, including instructions to perform the following steps: acquiring marker data, the marker data serving as correct answer data and related to determining whether each transaction corresponding to the user's pre-learning behavior information is equivalent to an unfair transaction based on the product's pre-learning price change information, wherein the instructions to acquire the marker data further include instructions to perform the following steps: outputting the user's pre-learning behavior information and the product's pre-learning price change information; and receiving user input related to the unfair transaction possibility score of each transaction, or user input related to whether each transaction is equivalent to an unfair transaction; using the unfair transaction detection model, determining whether a first transaction corresponding to the user behavior information is equivalent to an unfair transaction based on the product's price change information and the user behavior information; and in response to determining that the first transaction is an unfair transaction, changing the price change of the first transaction again.

Images