JP2022186632A - Method and system for generating special sales event based on user intent of search queries - Google Patents

Abstract

To provide a method of generating a special sales event based on user intent of search queries.SOLUTION: A method of generating a special sales event is provided, comprising: receiving data associated with a plurality of product queries; extracting a trend associated with at least either of a specific population and a specific time period on the basis of the data associated with the plurality of product queries; and generating a special sales event involving featured products associated with the extracted trend.SELECTED DRAWING: Figure 1

Description

TECHNICAL FIELD The present disclosure relates to a method and system for generating special sale events based on user intent of search queries, and in particular, extracting trends based on data associated with multiple product queries, and The present invention relates to a method and system for generating a bargain sale event including planned products.

2. Description of the Related Art Recently, various online product information providing methods and product sales methods have been developed for the convenience of consumers, such as TV shopping, Internet shopping, and catalog shopping. Merchandise sellers, shopping mall operators, open market operators, and the like can plan bargain sales events in which merchandise can be purchased at lower prices or under better conditions as a means of publicizing or marketing online shopping services. For example, product sellers can target specific consumers, collect products expected to be needed by target consumers, etc., and provide events where the products can be purchased at discounted prices. Alternatively, product sellers, etc. can collect products expected to be needed by consumers, etc. at a specific time and provide information on the products, etc. to consumers, etc. at that time (or at an earlier time). Such a bargain sale event enables consumers, etc., who belong to a specific period or a specific group, to check and purchase necessary products at a glance. In addition, product sellers, shopping mall operators, and open market operators can improve sales by changing the stagnant product sales situation through bargain events, and potential customers can use related online shopping services. can be induced to use

According to the conventional technology, product sellers, shopping mall operators, open market operators, etc. passively analyze and grasp the shopping trends of consumers, etc., and generate bargain events corresponding to the shopping trends. Must. That is, an operator who intends to create a bargain sale event should directly grasp the shopping trend and directly prepare the subject and subject of the bargain sale event that match the trend. In addition, an operator who wants to create a bargain sale event must directly search for or input products that match the title of the created bargain event and/or the theme of the bargain sale event, that is, products included as planned products. Inconvenient.

Korean Patent Publication No. 10-1665980

The present disclosure provides a sale event generation method, computer program and apparatus (system) for solving the above problems.

The present disclosure can be embodied in various forms including methods, apparatus (systems), and computer programs.

According to one embodiment of the present disclosure, a method for generating a sale event based on user intent of a search query performed by at least one computing device comprises: receiving data associated with a plurality of product queries; Based on the data associated with the query, extracting trends associated with at least one of a specific population or a specific time period, and generating a sale event including featured products associated with the extracted trends.

According to another embodiment of the present disclosure, a method of training a proposed product extraction model performed by at least one computing device includes the steps of: receiving data regarding a learning sale event; receiving information regarding each of a plurality of products; and outputting one or more items associated with the sale event subject upon input of the sale event subject based on the learned sale event subject and information regarding each of the plurality of items. As above, it includes a step of learning a planned product extraction model.

According to yet another embodiment of the present disclosure, a computer program is provided for executing the above method on a computer.

According to yet another embodiment of the present disclosure, an information processing system includes a communication module, a memory and at least one computer coupled to the memory and configured to execute at least one computer readable program contained in the memory. at least one program for receiving data associated with a plurality of product queries; trends associated with at least one of a specific population or a specific time period based on the data associated with the plurality of product queries; and generate a special sale event containing the featured product associated with the extracted trend.

According to yet another embodiment of the present disclosure, an information processing system includes a communication module, a memory and at least one computer coupled to the memory and configured to execute at least one computer readable program contained in the memory. one processor, the at least one program receives data about the learning sale event, receives information about each of the plurality of products, and based on the data about the learning sale event and the information about each of the plurality of products, It includes an instruction word for learning a plan product extraction model so that when a special sale event topic is input, one or more products associated with the special sale event topic are output.

In various embodiments of the present disclosure, product queries searched by consumers with similar characteristics and/or product queries searched by consumers at similar times are associated with the same shopping purpose or intent. As such, product queries or keywords with the same shopping purpose or intent can be clustered into multiple clusters. By extracting shopping trends from a plurality of clustered crowds, it is possible to grasp shopping trends that users (for example, managers of special sale events) do not discover, thereby generating various and many special sale events.

In various embodiments of the present disclosure, by entering keywords related to shopping trends (e.g., keywords indicating the product query corresponding to the shopping trend, the theme of the sale event, the target consumer, the time of year, etc.), the subject of the sale event can be automatically generated.

In various embodiments of the present disclosure, the featured merchandise included in the sale event can be automatically determined. As a result, a user (for example, an administrator of a bargain sale event) can search for products included as planned products of a bargain sale event, and can reduce operational resources required for the search. In addition, using the semantic similarity between the characteristics of the product extracted from information such as the title, category, introduction, review, tag, and image of the product and the topic of the sale event, it is matched with the theme of the sale event to be generated. products can be automatically extracted. As a result, it is possible to improve the variety and suitability of planned products included in the bargain sale event.

Embodiments of the present disclosure will be described based on the following accompanying drawings. Here, like reference numbers indicate like elements, but are not so limited.
FIG. 4 is a diagram illustrating an example of providing shopping content to a user through a screen of a user terminal according to an embodiment of the present disclosure; 1 is a schematic diagram illustrating a configuration in which an information processing system is communicatively coupled with a plurality of user terminals according to one embodiment of the present disclosure; FIG. 1 is a block diagram showing internal configurations of a user terminal and an information processing system according to an embodiment of the present disclosure; FIG. [0014] Fig. 4 is a flow chart illustrating a method for generating a sale event according to one embodiment of the present disclosure; [0014] Figure 4 illustrates an example of extracting trending keywords from data associated with multiple product queries according to an embodiment of the present disclosure; FIG. 4 illustrates an example of determining a topic for a sale event that indicates trending subject matter based on trending keywords, according to one embodiment of the present disclosure; FIG. 11 illustrates an example of extracting one or more products associated with a trend among a plurality of products based on the topic of a sale event according to one embodiment of the present disclosure; FIG. 4 is a diagram illustrating an example of learning a planned product extraction model according to an embodiment of the present disclosure; [0014] Figure 4 illustrates an example of a user (eg, operator) creating a sale event, according to one embodiment of the present disclosure; [0014] Figure 4 illustrates an example where one or more potential bargain events are provided to a user (e.g., operator) according to one embodiment of the present disclosure;

Hereinafter, specific contents for carrying out the present disclosure will be described in detail based on the accompanying drawings. However, in the following description, a detailed description of known functions and configurations will be omitted if it may unnecessarily obscure the gist of the present disclosure.

In the accompanying drawings, the same or corresponding components are provided with the same reference numerals. Also, in the description of the embodiments below, duplicate descriptions of the same or corresponding components may be omitted. However, the omission of a description of an element should not be construed as omitting such element from being included in an embodiment.

Advantages and features of the disclosed embodiments, and the manner in which they are achieved, will become apparent with reference to the embodiments described below with reference to the accompanying drawings. The present disclosure may, however, be embodied in various different forms and should not be construed as limited to the embodiments set forth below. However, the examples are provided only so that this disclosure will be complete and will enable those of ordinary skill in the art to accurately recognize the categories of the invention.

A brief description of terms used in the present disclosure will be provided and a specific description of the disclosed embodiments will be provided. The terms used in this specification have been selected as common terms that are currently widely used as much as possible while considering the function in this disclosure, but this is not the intention or precedent of an engineer engaged in the related field. , may change due to the emergence of new technologies. Also, certain terms may be arbitrarily chosen by the applicant, the meaning of which will be detailed in the description of the invention. Accordingly, terms used in this disclosure should be defined based on their meanings and overall context of this disclosure, rather than simply terminology.

In this disclosure, singular expressions may include pluralities and plural expressions may include the singular unless the context clearly dictates otherwise. In this disclosure, when a part "includes" a component, it does not exclude other components, but it can also include other components, unless specifically stated to the contrary. do.

Also, the term "module" or "unit" as used herein means a software or hardware component that performs a certain role. However, "module" or "unit" is not meant to be limited to software or hardware. A "module" or "unit" may be configured to reside on an addressable storage medium and configured to run on one or more processors. Thus, by way of example, a "module" or "part" may refer to components such as software components, object-oriented software components, class components, task components, as well as processes, functions, attributes, procedures, subroutines, It may include at least one of program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays or variables. Components and "modules" or "sections" may be combined with a smaller number of components and "modules" or "sections" or may be combined with additional components and "modules" or "sections". ” can be further separated into

According to one embodiment of the present disclosure, a "module" or "unit" may be embodied with a processor and memory. "Processor" should be interpreted broadly to include general purpose processors, central processing units (CPUs), microprocessors, digital signal processors (DSPs), controllers, microcontrollers, state machines, and the like. In some circumstances, a "processor" can also refer to an application specific semiconductor (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), or the like. A "processor" may be, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a DSP core, or any other such configuration. Any combination of processing devices may also be indicated. Also, "memory" should be interpreted broadly to include any electronic component capable of storing electronic information. "Memory" includes RAM (Random Access Memory), ROM (Read Only Memory), NVRAM (Non-Volatile Random Access Memory), PROM (Programmable Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable and Programmable Read-Only Memory) Various types of processor-readable media such as Erasable Programmable Read-Only Memory)), flash memory, magnetic or optical data storage devices, registers, etc. can also be indicated. Memory is said to be in electronic communication with a processor when the processor can read information from, or record information in, the memory. Memory integrated with the processor is in electronic communication with the processor.

In this disclosure, “shopping content” refers to broadcasts about shopping products provided by online shopping services via the Internet, social media, etc., information on special sale events, shopping product information, shopping payment information, shopping seller and manufacturer information can indicate information or content associated with online shopping such as; For example, shopping content may include, but is not limited to, content including text, images, or videos, live broadcast content, and recorded broadcast content for descriptions, advertisements, and public relations of products for online shopping. not something. As another example, the shopping content may include the title of the sale event, tags associated with the sale event, information regarding the products featured in the sale event, and the like.

For purposes of this disclosure, "special sales events" may generally refer to events in which one or more products associated with a particular theme are advertised, exhibited, and sold during a specified period of time. For example, a sale event can be provided by an online shopping service, social media, etc., and can be targeted to a specific user population at a specific time (eg, season, month of the year, day or time). In another example, a sale event can be created or provided to advertise, display and sell one or more products associated with a particular product subject matter, such as summer vacation clothing, baby toys and camping equipment. In one embodiment, the bargain event includes a bargain event title indicating the subject of the bargain event, the product to be sold, etc., planned product information (eg, product image, product price, product description, etc.), effective period of the bargain event, and/or The data can include, but is not limited to, sales conditions.

FIG. 1 is a diagram illustrating an example of providing shopping content to a user 100 through a screen 120 of a user terminal 110 according to an embodiment of the present disclosure. On the screen 120 shown in FIG. 1, the user (eg, consumer) 100 executes an application or the like that provides shopping content via the user terminal (eg, smartphone) 110, thereby providing information on bargain events as shopping content. Here is an example of what is For example, as shown in the figure, the user 100 can, via the user terminal 110, receive the title of a specific bargain sale event ("How about building a house with lighting?"), a hashtag associated with the bargain sale event, as shopping content. ("# table stand", "# floor stand", "# furniture", "# lighting") and information on the planned products included in the special sale event (for example, product image, product name, product price, etc.) can be provided.

To provide the user 100 with a sale event, an information processing system in communication with the user terminal 110 can generate and transmit one or more sale events to the user terminal 110 . In one embodiment, the information handling system can automatically generate one or more sale events based on various search queries (eg, product search queries). At this time, the information processing system provides information on one or more automatically generated special sale events to the operator of the special sale event, and according to the operator's selection of at least one of the one or more special sale events, Can publish selected sale events. Alternatively, the information handling system can immediately publish one or more automatically generated sale events. In another embodiment, the information processing system generates candidate sales event subjects, candidate target consumers for the sales event, candidate trend keywords for the sales event, candidate sales event topics, and/or candidate planned products included in the sales event. and can be provided to managers of bargain events. At this time, the operator's selection of at least one of candidate sales event subjects, target consumer candidates for the sales event, trend keyword candidates for the sales event, candidate topics for the sales event, and/or candidate planned products included in the sales event In response, the information handling system can generate one or more sale events.

An information handling system can receive data associated with multiple product queries to generate a sale event. Here, the 'product query' refers to a search query that a user inputs through an online shopping application or web page, a search application or web page, etc. executed by a user terminal in order to search for or purchase a specific product. can be done. In addition, the data associated with the product query includes the product query itself (for example, search query, text indicating search keywords, image, etc.), the date and time when the product query was input, and the user terminal where the product query was input. Information (eg, user terminal location, IP information, etc.), information of the user who entered the product query (eg, user account or profile information, user's age, gender, occupation, etc.) can be included.

The information processing system extracts a trend associated with at least one of a specific group or a specific period based on data associated with a plurality of product queries, and generates a bargain sale event including planned products associated with the extracted trend. can be generated. Here, the specific group may refer to a group generated by classifying a plurality of users based on one or more criteria (eg, age, gender, interests, occupation, etc.). For example, the specific group is a group classified by age group such as teens, 20s and 30s, a group classified by occupation such as office worker, self-employed person and university student, and a group of men and women. and a group classified by marital status, such as married and unmarried. Also, the specific period may include seasons, anniversaries, weekdays or weekends, public holidays, consecutive holidays, specific years, specific months of the year, specific weeks of the month, sections between specific points in the day, and the like.

The information processing system clusters the data associated with the plurality of product queries received into a plurality of crowds, and among the plurality of crowds, one or more clusters associated with at least one of a specific group or a specific period of time. One or more trend keywords can be extracted. For example, the information processing system, the data associated with a plurality of product queries, based on the query intent of the product query, the subject, the information of the user who searched for the product query, the date and time when the product query was input, etc. Can be clustered into multiple crowds. For example, an information handling system can cluster data associated with multiple product queries into multiple crowds using a statistical or probability-based clustering model, such as a statistical theme embedding model.

The information processing system can then determine a sale event title that indicates the trending subject matter based on one or more trending keywords. For example, the information processing system can combine one or more trend keywords to generate a sales event title consisting of phrases or sentences indicative of the subject of the trend. Additionally or alternatively, the information processing system uses a language model, such as a statistically-based language model or an artificial neural network-based language model, to identify one or more trending keywords as a sales event indicative of a trending subject. can be converted (or generated) to

Thereafter, the information processing system can determine one or more products associated with one or more trend keywords among the plurality of products as plan products, and generate a bargain sale event including the plan products. In one embodiment, the information processing system can determine one or more products associated with one or more trend keywords among a plurality of products as planned products based on the title of the bargain sale event. For example, the information processing system uses the planned product extraction model to extract one or more products associated with one or more trend keywords that match the title of the bargain sale event from among the plurality of products. , one or more extracted commodities can be determined as planned commodities. Additionally or alternatively, the information processing system extracts at least one product that matches the title of the bargain sale event from among the plurality of products using the planned product extraction model, and stores the extracted at least one product. can be output via the user terminal. After that, the information processing system can determine one or more products as planned products according to the user's selection of one or more products from the output list.

In one embodiment, the planned product extraction model includes information about each of a plurality of products (for example, product name, product description, category, related product, tag, introduction, review, image, price, brand, manufacturer, etc.). Based on this, calculate the matching score with the topic of the special sale event for each of the multiple products, extract a predetermined number of products from the multiple products based on the calculated matching score, and extract the predetermined number of products It can be a model trained to output a product list containing products. Additionally or alternatively, the featured product extraction model may include an embedding generative model trained to generate embedding data for each of the plurality of products and embedding data for the topic of the sale event. At this time, the planned product extraction model calculates a matching score for each of the plurality of products with the topic of the special sale event based on the embedded data regarding each of the plurality of products and the embedded data regarding the topic of the special sale event. It may be a model trained to extract a predetermined number of products among the plurality of products based on the matching score and output a product list containing the extracted predetermined number of products.

According to the above-described embodiments and the like, a trend associated with a specific group and/or a specific period can be extracted by reflecting information on a plurality of product queries actually input by consumers, etc., and matching the extracted trend You can generate a special sale event that As a result, a bargain event can be generated by grasping trends that are not discovered by the manager of the bargain sale event in a timely manner, and the bargain event can include various planned products that match the newly grasped trends. In addition, automatically generating a bargain sale event having a bargain sale event theme, bargain sale event title, and/or planned product matching the extracted trend can reduce the operational burden on the operator. In addition, the manager can efficiently generate/publish a bargain sale event by automatically recommending a bargain sale event theme, bargain sale event title, and/or planned products that match the current product sales trend.

FIG. 2 is a schematic diagram illustrating a configuration in which an information processing system 230 is communicatively coupled with a plurality of user terminals 210_1, 210_2, 210_3 according to one embodiment of the present disclosure. As shown, a plurality of user terminals 210_1, 210_2, and 210_3 are connected via a network 220 to an information processing system 230 that provides a bargain event providing service and/or a bargain event generating service. Here, the plurality of user terminals 210_1, 210_2, and 210_3 are terminals of consumers to whom the bargain event is provided and/or operators (for example, bargain event planners and managers) to whom the bargain event generation service is provided. can include In one embodiment, information processing system 230 is capable of storing, providing, and executing computer-executable programs (e.g., downloadable applications) and data associated with sale event provisioning services and/or sale event generation services, and the like. One or more server devices and/or databases, or one or more distributed computing devices and/or distributed databases of a cloud computing service infrastructure.

The bargain sale event providing service and/or bargain sale event generation service provided by the information processing system 230 includes a bargain sale event generation application, a shopping application, a search application, and a mobile browser application installed in each of the plurality of user terminals 210_1, 210_2, and 210_3. Alternatively, it can be provided to the user via a web browser or the like. For example, the information processing system 230 may provide information corresponding to a request for generating a sale event received from the user terminals 210_1, 210_2, and 210_3 through a sale event creation application or the like, or perform corresponding processing. .

A plurality of user terminals 210_1, 210_2, 210_3 can communicate with an information processing system 230 via a network 220. FIG. Network 220 may be configured to allow communication between multiple user terminals 210_1, 210_2, 210_3 and information processing system 230. FIG. Depending on the installation environment, the network 220 may be a wired network such as Ethernet (registered trademark), PLC (Power Line Communication), telephone line communication device and RS-serial communication, mobile communication network, WLAN ( Wireless LAN), Wi-Fi, Bluetooth® and ZigBee, etc. or a combination thereof. The communication method is not limited, and the user terminals 210_1 and 210_2 as well as the communication method using a communication network (eg, mobile communication network, wired Internet, wireless Internet, broadcasting network, satellite network, etc.) that can include the network 220. , 210_3 is also included.

In FIG. 2, mobile phone terminal 210_1, tablet terminal 210_2, and PC terminal 210_3 are shown as examples of user terminals, but user terminals 210_1, 210_2, and 210_3 are capable of wired and/or wireless communication. Yes, and can be any computing device on which a sale event generation application, shopping application, search application, mobile browser application or web browser, etc. can be installed and executed. For example, user terminals include AI speakers, smart phones, mobile phones, navigation systems, computers, notebooks, digital broadcasting terminals, PDA (Personal Digital Assistants), PMP (Portable Multimedia Player), tablet PCs, game consoles, It can include wearable devices, Internet of things (IoT) devices, VR (virtual reality) devices and AR (augmented reality) devices, set-top boxes, and the like. Also, although FIG. 2 shows three user terminals 210_1, 210_2, and 210_3 communicating with the information processing system 230 via the network 220, this is not a limitation, and different numbers of user terminals may communicate via the network 220. can also be configured to communicate with information processing system 230 via

According to one embodiment, the information processing system 230 can receive data associated with multiple product queries from multiple user terminals 210_1, 210_2, 210_3. Thereafter, the information processing system 230 generates a sale event based on the data associated with the plurality of product queries thus received, and uses the plurality of user terminals 210_1, 210_2, 210_3 (e.g., sale event operators) to generate a sale event. and/or the consumer's user terminal). Additionally or alternatively, the information processing system 230 may generate, based on the data associated with the plurality of product queries, potential sales event subjects, potential target consumers for the Candidate titles and/or proposed products to be included in the sale event can be generated and provided to a plurality of user terminals 210_1, 210_2, 210_3 (eg, user terminals of the sale event operator). At this time, the information processing system 230 selects at least one of a candidate sale event subject, a sale event target consumer candidate, a sale event trend keyword candidate, a sale event topic candidate, and/or a plan product candidate included in the sale event. User selections for can be received from a plurality of user terminals 210_1, 210_2, 210_3 to generate one or more sale events.

FIG. 3 is a block diagram showing internal configurations of the user terminal 210 and the information processing system 230 according to one embodiment of the present disclosure. The user terminal 210 can represent any computing device capable of wire/wireless communication, capable of running a sale event generation application, a shopping application, a search application, a mobile browser application or a web browser, etc. For example, FIG. 2, such as a mobile phone terminal 210_1, a tablet terminal 210_2 and a PC terminal 210_3. As shown, user terminal 210 may include memory 312 , processor 314 , communication module 316 and input/output interface 318 . Similarly, information handling system 230 may include memory 332 , processor 334 , communication module 336 and input/output interface 338 . As shown in FIG. 3, user terminal 210 and information handling system 230 can be configured to communicate information and/or data over network 220 using respective communication modules 316 , 336 . The input/output device 320 is also configured to input information and/or data to the user terminal 210 and output information and/or data generated from the user terminal 210 via the input/output interface 318. can.

Memory 312, 332 may include any non-transitory computer-readable medium. According to one embodiment, the memories 312, 332 may include random access memory (RAM), read only memory (ROM), disk drives, solid state drives (SSDs), and flash memory. may include a permanent mass storage device such as a As another example, permanent mass storage devices such as ROM, SSD, flash memory, and disk drives are included in the user terminal 210 or the information processing system 230 as separate permanent storage devices separated from memory. be able to. In addition, the memories 312 and 332 may store operating systems and at least one program code (eg, code for a sale event generation application installed and operated in the user terminal 210).

Such software components can be loaded from a computer-readable medium separate from the memories 312,332. Such separate computer-readable recording media can include recording media directly connectable to such user terminal 210 and information processing system 230, such as floppy drives, disks, tapes, etc. , DVD/CD-ROM drives and memory cards. As another example, software components and the like can be loaded into memory 312, 332 via communication modules that are not computer readable media. For example, at least one program can be loaded into memory 312 , 332 based on a computer program installed by files provided over network 220 by a developer or a file distribution system that distributes application installation files.

The processors 314, 334 can be configured to process computer program instructions by performing basic arithmetic, logic and input/output operations. Instructions may be provided to processors 314 , 334 by memory 312 , 332 or communication modules 316 , 336 . For example, the processors 314,334 can be configured to execute instructions received by program code stored in a storage device, such as the memory 312,332.

The communication modules 316 and 336 can provide configuration and functions for the user terminal 210 and the information processing system 230 to communicate with each other via the network 220 so that the user terminal 210 and/or the information processing system 230 can communicate with other users. Configurations and functions for communicating with terminals or other systems (eg, separate cloud systems, etc.) can be provided. As an example, the processor 314 of the user terminal 210 may generate a request or data (e.g., a request to generate a sale event, a request to issue a sale event, a plurality of products, etc.) generated by program code stored in a storage device such as the memory 312 . data associated with the query, etc.) may be communicated to information processing system 230 over network 220 under the control of communications module 316 . Conversely, control signals or instructions provided by control of processor 334 of information handling system 230 may be received by user terminal 210 via communication module 316 of user terminal 210 via communication module 336 and network 220 . For example, the user terminal 210 may receive, from the information processing system 230 via the communication module 316, a candidate sale event subject, a sale event target consumer candidate, a sale event trend keyword candidate, a sale event topic candidate and/or a sale event candidate. You can receive planned product candidates included in .

Input/output interface 318 may be a means for interfacing with input/output devices 320 . By way of example, input devices include devices such as cameras, keyboards, microphones, mice, etc. that include audio sensors and/or image sensors, and output devices include devices such as displays, speakers, haptic feedback devices, etc. can include As another example, the input/output interface 318 may be a means for interfacing with a device such as a touch screen or the like that has a single integrated structure or function for performing input and output. For example, a service screen constructed using information and/or data provided by the information processing system 230 or other user terminals when the processor 314 of the user terminal 210 processes instructions of a computer program loaded in the memory 312. etc. may be displayed on the display via the input/output interface 318 . Although the input/output device 320 is not included in the user terminal 210 in FIG. Input/output interface 338 of information handling system 230 also couples to information handling system 230 or for interfacing with devices (not shown) for input or output that information handling system 230 may include. can be a means of Although FIG. 3 shows the input/output interfaces 318 and 338 as elements configured separately from the processors 314 and 334, the present invention is not limited to this and the input/output interfaces 318 and 338 are configured to be included in the processors 314 and 334. You can also

User terminal 210 and information processing system 230 may include more components than those shown in FIG. However, it is not necessary to explicitly show most prior art components. According to one embodiment, the user terminal 210 can be implemented to include at least a portion of the input/output device 320 described above. In addition, the user terminal 210 may further include other components such as a transceiver, a GPS (Global Positioning System) module, a camera, various sensors and a database. For example, if the user terminal 210 is a smart phone, it can generally include components that a smart phone has, such as an acceleration sensor, a gyro sensor, a camera module, various physical buttons, buttons using a touch panel, input/output ports, and Various components such as a vibrator for vibration may be further included in the user terminal 210 . According to one embodiment, the processor 314 of the user terminal 210 can be configured to run an application such as that provides a sale event generation service. Code associated with the application and/or program can then be loaded into the memory 312 of the user terminal 210 .

While a program for an application that provides a sale event generation service, etc., is running, the processor 314 controls a touch screen, a keyboard, a camera including an audio sensor and/or an image sensor, a microphone, etc. coupled with the input/output interface 318. It can receive text, image, video, voice and/or action input or selected through an input device, and store the received text, image, video, voice and/or action in the memory 312 or communicate. It may be provided to information processing system 230 via module 316 and network 220 . For example, the processor 314 indicates user selections for potential sales event topics, potential target consumers for the sales event, potential trend keywords for the sales event, potential sales event topics, and/or potential product offerings to be included in the sales event, and the like. Input can be received and provided to information processing system 230 via communications module 316 and network 220 . As another example, processor 314 may receive and provide user input to information handling system 230 via communications module 316 and network 220 requesting the publication of a generated sale event. As yet another example, processor 314 receives user input entering a product query via input device 320 and transmits the product query and data associated with the product query to the information processing system via network 220 and communication module 316 . 230 can be provided.

Processor 314 of user terminal 210 may be configured to manage, process and/or store information and/or data received from input device 320, other user terminals, information processing system 230 and/or external systems. . Information and/or data processed by processor 314 may be provided to information processing system 230 via communications module 316 and network 220 . Processor 314 of user terminal 210 can transfer and output information and/or data to input/output device 320 via input/output interface 318 . For example, processor 314 can display the received information and/or data on the screen of the user terminal.

Processor 334 of information handling system 230 may be configured to manage, process and/or store information and/or data received from multiple user terminals 210 and/or multiple external systems. Information and/or data processed by processor 334 may be provided to user terminal 210 via communications module 336 and network 220 . In one embodiment, processor 334 of information processing system 230 generates trends associated with at least one of a particular population or a particular time period based on data associated with multiple product queries received from multiple user terminals 210 . A sale event can be generated that includes featured products that are extracted and associated with the extracted trends.

The processor 334 of the information processing system 230 outputs the information processed via the output device 320 such as a device capable of display output of the user terminal 210 (eg, touch screen, display, etc.), a device capable of audio output (eg, speaker). and/or data can be configured to be output. For example, the processor 334 of the information processing system 230 provides the generated (or published) sale event (or information about the sale event) to the user terminal 210 via the communication module 336 and the network 220, and sends the sale event to the user terminal 210. can be configured to be output by a device capable of outputting a display. As another example, the processor 334 of the information processing system 230 may transmit, via the communication module 336 and the network 220 , the user terminal 210 with candidate topics for the sale event, candidate target consumers for the sale event, candidate trend keywords for the sale event, Candidate titles of the event and/or candidates of planned products included in the bargain event can be provided, and can be configured to be output by a device capable of display output of the user terminal 210 .

FIG. 4 is a flowchart illustrating a sale event generation method 400 according to one embodiment of this disclosure. In one embodiment, sale event generation method 400 can be performed by a processor (eg, at least one processor of an information handling system). The sale event generation method 400 may begin by the processor receiving data associated with a plurality of product queries (S410). For example, the processor can receive data associated with multiple product queries from multiple user terminals. Additionally or alternatively, the processor can receive data associated with multiple product queries from internal and/or external storage (eg, a product query database, etc.). Here, the data associated with a product query includes the product query itself (for example, search query, text indicating a search keyword, image, etc.), the date and time when the product query was input, and the user terminal where the product query was input. (e.g., user terminal location, IP information, etc.), information of the user who entered the product query (e.g., user account or profile information, user's age, gender, occupation, etc.).

The processor can extract trends associated with at least one of a specific population or a specific time period based on data associated with multiple product queries (S420). In one embodiment, the processor clusters the data associated with the plurality of product queries into a plurality of clusters, and among the plurality of clusters, one or more clusters are associated with at least one of a particular population or a particular time period. One or more trending keywords (eg, search queries, topics, dates, etc.) can be extracted. For example, the processor can use statistical thematic embedding models to cluster data associated with multiple product queries into multiple clusters.

The processor can then generate a sale event including the featured product associated with the extracted trend (S430). For example, the processor can determine one or more products associated with one or more trend keywords among the plurality of products as planned products, and generate a bargain sale event including the planned products. In one embodiment, the processor determines a sale event topic indicative of the trending subject based on the one or more trending keywords, and determines one or more of the plurality of products based on the sale event topic. One or more products associated with trend keywords can be determined as planned products. To this end, the processor can use the language model to transform one or more trending keywords into a sale event topic that indicates the trending subject matter.

For example, the processor uses the planned product extraction model to extract one or more products associated with one or more trend keywords that are matched with the topic of the special sale event from among the plurality of products, One or more products can be determined as planned products. Additionally or alternatively, the processor extracts at least one product from among the plurality of products that matches the subject of the sale event using the planned product extraction model, and creates a list including the extracted at least one product. can be output via the user terminal. At this time, the processor may determine one or more products as planned products according to the user's selection of one or more products from the output list.

In one embodiment, the planned product extraction model calculates a matching score for each of the plurality of products with the theme of the special sale event based on information about each of the plurality of products, and based on the calculated matching score, at least It may be a model trained to extract one or more products and output the product list. Further, the planned product extraction model may be a model trained to extract a predetermined number of products from among a plurality of products and output a product list including the extracted predetermined number of products. Additionally or alternatively, the featured product extraction model may include an embedded generative model trained to generate embedded data for each of the plurality of products and embedded data for the topic of the sale event. At this time, the planned product extraction model calculates a matching score for each of the plurality of products with the topic of the special sale event based on the embedded data regarding each of the plurality of products and the embedded data regarding the topic of the special sale event. It may be a model trained to extract a predetermined number of products among the plurality of products based on the matching score and output a product list containing the extracted predetermined number of products.

Additionally or alternatively, the planned product extraction model is such that a matching score between the embedded data regarding the subject of the learning special sale event and the embedded data regarding the products included in the learning sale event is the It can be a model trained with a triplet loss function that constrains the matching score between the embedded data and the embedded data for items not included in the learning sale event to be greater. Here, the learning sale event may indicate a sale event that has actually progressed in the past and/or a sale event that is artificially generated to learn the planned product extraction model.

FIG. 5 is a diagram illustrating an example of extracting trending keywords 540 from data 510 associated with multiple product queries according to one embodiment of this disclosure. A user group containing users with similar characteristics (e.g., age, gender, occupation, interests, etc.) is likely to visit during a similar period (e.g., within a week, within a month, a specific season such as spring, consecutive holidays, or anniversaries). A product query searched for during a specific time period, such as a specific time period, may include the same or similar user intent (eg, shopping intent or search intent) shared among the users. In this way, product queries input or searched with the same or similar user intentions can be clustered into one or more groups, and user product searches or product shopping trends are extracted from the groups containing the same or similar user intentions. be.

For example, a processor (e.g., at least one processor of an information processing system) generates one or more crowds containing product queries with the same or similar user intent (e.g., search intent, shopping intent, etc.), It is also possible to extract trends for specific populations and/or specific times from one or more crowds. Here, the specific group may indicate a group of users who have input or searched product queries included in one or more groups. Additionally or alternatively, the specific group may indicate the gender, age, occupational group, etc. of users who have entered or searched multiple product queries included in one or more groups. In addition, product queries can include meaningful queries (eg, refined, filtered and/or pre-processed shopping queries) among queries that users have entered or searched for shopping.

In one embodiment, the processor can generate multiple clusters 530 by clustering data 510 associated with multiple product queries into multiple clusters 530 . The processor can then extract one or more trending keywords 540 associated with at least one of a particular group or a particular time period from one or more of the plurality of crowds 530 . For example, the processor extracts one or more trend keywords 540 from one or more crowds out of the plurality of crowds 530, and extracts one or more trend keyword-specific features (e.g., searching for a large number of the trend keywords). Specific populations and/or characteristic time periods associated with one or more trending keywords 540 can be determined based on demographics, gender, age, time of year, etc.). That is, the processor can extract trends associated with at least one of a particular population or a particular time period based on data 510 associated with a plurality of product queries received.

The processor also uses the statistical thematic embedding model 520 to cluster the data 510 associated with the plurality of product queries into a plurality of clusters 530, and among the plurality of clusters 530, one from one or more clusters. The above trend keywords 540 can be extracted. Here, the statistical thematic embedding model 520 is a statistical or probability-based clustering model learned to cluster product queries with common shopping intent (or search intent or query intent) into one or more crowds. and/or embedded models. As shown, the processor inputs data 510 associated with the plurality of product queries into a statistical thematic embedding model 520 to generate a plurality of embeddings corresponding to each of the data 510 associated with the plurality of product queries. Can generate data. The processor can cluster the generated embedded data into clusters 530 . The processor can then extract one or more trending keywords 540 from one or more of the crowds 530 .

In one embodiment, the processor can use the statistical thematic embedding model 520 to embed each product query into an embedding space based on the data 510 associated with the multiple product queries. The processor can then cluster the multiple product queries into multiple clusters 530 based on the distance between each product query in the embedding space. For example, the processor can cluster multiple product queries into multiple crowds 530 by generating crowds containing product queries with close distances relative to a reference product query (e.g., seed keyword) in the embedding space. . Here, the reference product query may include product queries or keywords indicating trendiness.

The processor can determine a reference product query to cluster the product queries into the crowds 530 . That is, the processor can determine a trending reference product query (or focus product query) based on data 510 associated with multiple product queries. In one embodiment, the processor identifies popular keywords for a particular time of year (e.g., summer, winter, etc.) and/or a particular population (e.g., a particular age or gender) based on data 510 associated with multiple product queries. The extracted popular keywords can be determined as the reference product query. For example, the processor can extract keywords reflecting temporal characteristics based on data 510 associated with multiple product queries and determine them as reference product queries (eg, seed keywords and main intent keywords). Here, the reference product query may include a popular keyword corresponding to a current or future specific point in time or period, so that the processor can generate special sales corresponding to a current or future specific point in time or period desired by the operator. Can generate events.

In order to determine the reference product query, the processor constructs and analyzes time series data based on the product query inflow volume by gender/age group to identify high seasonality, trend ( trend), you can select product queries with product query traffic. In the case of a current highly seasonal product query (or keyword), the past and future that are temporally adjacent to the present may also maintain similar seasonality patterns. Therefore, the processor replaces the time-series data of product queries in the future (for example, one week later or two weeks later) when the product query inflow volume cannot be observed, instead of the product query time-series data for the same past period (for example, the same date one year ago or Time-series data collected during the same season, etc.) can be analyzed to determine a baseline product query.

As an example, the processor inputs valid product queries obtained by refinement and/or filtering into the statistical thematic embedding model 520, thereby embedding each product query on the embedding space and generating a reference product query (or keyword) (e.g., , “baby coveralls”), a crowd including product queries (or keywords) (for example, “mother’s bag”, “newborn child car seat”, “baby carrier”, etc.) that are close to each other in the embedded space can be generated. The processor then extracts one or more trending keywords 540 from the crowd and determines the specific group and/or specific time period associated with the crowd and/or the one or more trending keywords 540. Trends associated with particular populations and/or particular times can be extracted.

One or more trend keywords 540 extracted from the crowd are common features such as keywords included in product queries included in the crowd, themes indicated by the crowd, and product queries included in the crowd. , the user population associated with the crowd, the time of year associated with the crowd, and the like. For example, the one or more trend keywords 540 extracted from the "baby coveralls" crowd are the product queries included in the "baby coveralls" crowd (i.e., the query set of the crowd): "mother's bag", "newborn child seat”, “carrying strap”, and the like. Additionally or alternatively, one or more trending keywords 540 extracted from the "baby coveralls" crowd may be the subject of the "baby coveralls" crowd (e.g., common keywords such as product queries contained in the "baby coveralls" crowd). characteristics, categories, relationships, query intent, etc.). Additionally or alternatively, the one or more trending keywords 540 extracted from the “baby coveralls” crowd may be “April,” etc., which is the time when many product queries included in the “baby coveralls” crowd were entered. can contain. Additionally or alternatively, the one or more trending keywords 540 extracted from the "baby coveralls" crowd may be based on age, gender, marital status, etc. of a group of users, such as users who have entered numerous product queries included in the "baby coveralls" crowd. "30s", "female", "married", etc., which are attributes related to the presence or absence of a person, can be included. Based on one or more trending keywords 540 extracted from the crowd in this manner, a trending topic (or sale event topic) (eg, "newborn delivery theme") for the crowd can be generated.

The data 510 associated with multiple product queries can be in various forms of data. For example, data 510 associated with multiple product queries may include various forms of data such as text data, vector data, and the like.

FIG. 6 is a diagram illustrating an example of determining a sale event topic 630 that indicates a trending topic based on trending keywords 610, according to one embodiment of the present disclosure. A processor (eg, at least one processor of an information handling system) can determine a sale event topic 630 that indicates a trending topic based on one or more trending keywords 610 . For example, the processor can include a sale event title generator (or sales event title generator) 620 that generates a title for the sale event.

In one embodiment, the sale event topic generator 620 can simply combine one or more of the input trend keywords 610 to generate a sale event topic 630 that indicates the trending topic. In another embodiment, the sale event topic generator 620 can use a language model to transform one or more trending keywords 610 into a sale event topic that indicates the topic of the trend. At this time, the sale event topic generator 620 can include a language model that, based on the input keyword, generates sentences in which the intent of the input keyword (eg, the topic associated with the keyword, etc.) is displayed. It may be an artificial neural network model and/or a statistical model trained to do so. That is, the language model may be a model trained to recognize relationships and meanings among a plurality of input trend keywords and generate sentences indicating trend themes.

As shown in the figure, the bargain sale event topic generating unit 620 generates the bargain event title "Wait for me" by inputting trend keywords 610 including "rash guard", "swimwear", "beachwear", and "June 27". ! Water Park” 630 can be generated. In one embodiment, the sale event topic generator 620 can generate multiple sentences based on one or more trend keywords 610 input. At this time, the processor may determine at least one of the plurality of sentences generated by the special sale event title generator 620 as the special sale event title 630 . For example, the processor outputs a plurality of sentences generated by the bargain sale event topic generating unit 620 as candidate bargain sale event titles via a user terminal (for example, a bargain sale event operator's terminal), In response to a user's (eg, sale event operator's) selection of at least one sentence, the selected at least one sentence may be determined as the sale event title 630 . Alternatively, the processor can receive text directly input by the user via the user terminal and determine it as the subject of the sale event.

FIG. 7 is a diagram illustrating an example of extracting one or more items 730 associated with a trend among a plurality of items based on a sale event topic 710 according to one embodiment of the present disclosure. A processor (for example, at least one processor of an information processing system) determines one or more products 730 associated with one or more trend keywords among a plurality of products as plan products, and includes plan products. Can generate special sale events. In one embodiment, the processor may extract one or more products 730 that match the theme of the bargain sale event from among the plurality of products using the planned product extraction model, and determine them as planned products. Here, the planned product extraction model may be a model trained to output one or more products matched with the input special sale event title. For example, the planned product extraction model extracts the title of a bargain sale event for each of a plurality of products based on information about each of the plurality of products (for example, product titles, categories, tags, introductions, reviews, images, etc.). Calculate the matching score (e.g., the distance in the embedding space), extract a predetermined number of products from among multiple products based on the calculated matching score, and extract a product list containing the predetermined number of extracted products It can be a model trained to output

In another embodiment, the processor uses the proposed product extraction model to extract at least one product from among the plurality of products that matches the topic of the special sale event, and extracts at least one product that matches the topic of the special sale event. A list containing one product can be output via the user terminal. The processor can then determine one or more products as the featured product in response to a user's (eg, event manager's) selection of one or more products in the output list. That is, the processor can provide planned product candidates to the user via the user terminal, and can determine one or more products selected by the user via the user terminal from among the planned product candidates as planned products. Additionally or alternatively, the processor can determine the product selected and/or input directly by the user via the user terminal as the proposed product.

As shown in the figure, the processor plans one or more products 730 output from the planned product extraction model 720 out of the plurality of products by inputting the title 710 of the bargain sale event to the planned product extraction model 720 . A product can be determined and a bargain sale event including the planned product can be generated. Here, the special sale event title 710 input to the planned product extraction model 720 includes not only the text of the special sale event title itself, but also the target consumers (for example, target group), subject, intention, and subtitle of the special sale event. , hashtags, etc. For example, when the processor inputs the special sale event title 710 “Wait! "Swimsuit", "Swimming cap", "Aqua shoes", "Pool bag", "Water park season ticket", etc., which are products with a high value (that is, products matched with the title of the bargain sale event) can be output. The processor can determine at least some of the plurality of output products as planned products. For example, the processor outputs a list including a plurality of products output from the planned product extraction model 720 as planned product candidates via the user terminal, and according to the user's selection of at least a part of the output multiple products , the selected product can be determined as the planned product.

FIG. 8 is a diagram illustrating an example of learning a planned product extraction model according to one embodiment of the present disclosure. The proposed product extraction model described above may include an embedding generation model trained to generate embedded data for each of a plurality of products and embedded data for a sale event topic. Here, the embedded generative model can generate embedded data for each of the plurality of products based on input information about each of the plurality of products. For example, an embedded generative model generates embedded data for each of multiple products based on each product's product name, category, seller, tags, brand, manufacturer, product description, options, price, image, etc. It can be a model that has been trained to produce. At this time, the planned product extraction model calculates a matching score for each of the plurality of products with the topic of the special sale event based on the embedded data regarding each of the plurality of products and the embedded data regarding the topic of the special sale event. An artificial neural network model and/or statistical model trained to extract a predetermined number of products from a plurality of products based on matching scores and output a product list including the extracted predetermined number of products. obtain.

A processor (e.g., at least one processor of an information processing system) can receive data about a learning sale event and/or receive information about each of a plurality of products to generate and/or learn a proposed product extraction model. . Here, the learned bargain event data may include bargain event data that has actually progressed in the past and/or bargain event data that is artificially generated for learning the planned product extraction model. In addition, the data regarding the bargain event may include the title of the bargain event, planned products, target consumers, bargain event time, and the like. Also, learning sale events can be received from storage devices internal and/or external to the information handling system.

The processor is configured to output one or more items associated with the sale event subject when the sale event subject is input based on the learned sale event data and information regarding each of the plurality of items. , can learn the planning product extraction model. In one embodiment, the processor calculates a matching score for each of the plurality of commodities with the topic of the bargain sale event by inputting the subject of the bargain sale event, and based on the calculated matching score, determines the matching score of the plurality of commodities. To extract a predetermined number (for example, 10) of products (for example, products whose matching score ranking is from 1st to 10th) and output a product list containing the extracted predetermined number of products, You can learn the planning product extraction model.

In one embodiment, the processor uses an embedded data generation model (e.g., a language model such as BERT (Bidirectional Encoder Representations from Transformers)) to generate embedded data about the topic of the learning sale event, included in the learning sale event. You can generate embedded data about products and embedded data about products that are not included in learning sale events. For example, the processor can generate embedded data related to the subject of the learning sale event by inputting the subject of the learning sale event to the embedded data generation model, and the products included in the learning sale event (for example, learning information about the products included in the learning sale event), you can generate embedded data about the products included in the learning sale event, and the products not included in the learning sale event (e.g., the products not included in the learning sale event By entering information about products), you can generate embedded data about products that are not included in learning sale events. The processor then determines that a matching score between the embedded data about the topic of the learning sale event and the embedded data about the products included in the learning sale event is equal to the embedded data about the topic of the learning sale event and the embedded data about the topic of the learning sale event. The proposed product extraction model can be trained using a triplet loss function that forces the matching score to be greater than the matching score between embedded data for products not included in . Here, the matching score between embedded data can be inversely proportional to the distance between embedded data on the embedding space.

Furthermore, in order to improve the performance of the proposed product extraction model, a sampling process (e.g., negative sampling) for extracting products that are not included in learning bargain events can be performed as learning data. In one embodiment, the processor selects, among a plurality of products that are not included in the learning sale event, products that include some similar features to products included in the learning sale event as learning data (i.e., learning data). negative data). For example, when learning a plan product extraction model using a “shaver” bargain sale event as a learning bargain sale event, the processor selects a plurality of products that are not included in the “shaver” bargain sale event and selects the same category as the shaver (for example, “ "Digital category") can be extracted as a "product not included in learning bargain sale event" for learning the planned product extraction model. At this time, the planned product extraction model can learn to extract planned products by considering more detailed features of the products.

As shown in the figure, the processor generates a topic A (referred to as “anchor”) 812 of the learning sale event, a product B included in the learning sale event (that is, a planned product of the learning sale event) (“positive (positive)") 814, and product C not included in the learning special sale event (that is, any product that is not the planned product of the learning special sale event) (called "negative") 816 are embedded in the generative model. Inputs at 810 can generate embedded data for each of Topic A, Item B, and Item C of the learning sale event. After that, the processor moves the embedded data 822 related to the learning special sale event topic A in the direction in which the distance between the embedded data 822 related to the learning special sale event topic A and the embedded data 824 related to the product B becomes closer in the embedding space. , and the embedded data 826 related to product C, the planned product extraction model can be learned.

FIG. 9 is a diagram illustrating an example of a user (eg, operator) creating a sale event, according to one embodiment of the present disclosure. A user terminal (eg, at least one processor of the user terminal) can provide a user (eg, an operator) with a user interface for creating a sale event. In one embodiment, the user terminal can display subject matter candidates 912 for sale events on a display. For example, the user terminal can present on the display candidates for a particular population and/or a particular time of year to target a sale event. The user terminal can then receive user input selecting at least one of the potential sales event subject matter 912 and forward the request and/or data associated therewith to the information processing system.

In one embodiment, the information processing system selects a user-selected sales event subject (i.e., a user-selected specific group and/or specific One or more trend keywords can be extracted from one or more crowds associated with the time period) and output as trend keyword candidates 914 via the user terminal. Here, one or more crowds may be a crowd including many product queries input by a specific group selected by the user and/or a crowd including multiple product queries input at a specific time. The user terminal can display trending keyword candidates 914 on a display, receive user input selecting at least one trending keyword of trending keyword candidates 914, and transmit requests and/or data associated therewith to an information processing system. can be transferred.

In one embodiment, the information processing system can generate one or more sentences based on at least one trending keyword selected by the user, and output them as candidate titles 916 for the sale event via the user terminal. For example, the information processing system may include not only the user-selected at least one trending keyword, but also keywords associated with the subject of the user-selected sale event, keywords associated with one or more crowds, and/or trending keywords. One or more phrases or sentences can be generated based on similar keywords and output via the user terminal as topic candidates 916 for the bargain event. The user terminal can display the potential sale event topics 916 on a display, receive user input selecting at least one of the potential sales event topics 916, and process requests and/or data associated therewith. can be transferred to the system.

In one embodiment, the information processing system extracts at least one product that matches the title of the bargain sale event selected by the user from among the plurality of products, and creates a list including the extracted at least one product. Can be output via the user terminal as candidate 922 . Thereby, the user terminal can display a list including at least one product (that is, a planned product candidate) on the display, and the user input to select at least one product from the list (that is, a planned product candidate) as a planned product. Requests and/or data associated therewith can be forwarded to an information handling system. In addition, the information processing system can receive requests for generating and/or publishing special sale events from user terminals.

As described above, the information processing system generates and generates a special sale event including the special sale event topic, trend keyword, sale event topic, and/or proposed product selected by the user based on the received request and/or data. / Or can be issued. Additionally or alternatively, the user can directly enter the subject of the sale event, trending keywords, subject of the sale event and/or featured product without selecting from a pool of candidates. That is, the information processing system can generate and/or issue a bargain sale event based on the bargain sale event theme, trend keyword, bargain event title, and/or plan product input by the user. The information handling system can provide a sale event to consumers by generating and publishing the sale event.

As shown in a first operation 910 , the user terminal selects period candidates (“this week”, “next week”, “two weeks from now”, “three weeks from now”) and/or group candidates as topic candidates 912 for the bargain event. ("early teens", "late teens", "early 20s", "30s", "40s", "50s", "60s", "male", "female") can be output via In response to a user input that selects at least one of the output bargain event subject candidates 912 (e.g., “2 weeks later”, “40s”, “women”), the user terminal Trend keyword candidates extracted from the first crowd associated with the group (“home building”, “lighting”, “curtain”, “interior”) 914 and trend keyword candidates extracted from the second crowd (“parker , “hood”, “winter clothes”) 914 can be output via the user interface. Here, the first crowd and/or the second crowd are a crowd associated with many product queries entered during the selected period and/or a crowd associated with many product queries entered by the selected group. can be

The user terminal, in response to a user input for selecting at least one of the output trend keyword candidates 914 (“building a house”, “lighting”, “interior”), displays a bargain sale generated based on the selected trend keyword. Event topic candidates (“How about building a house with lighting?”, “Interior items for this winter”) 916 can be output via the user interface. In response to the user input selecting at least one of the sale event topic candidates 916 output from the first action 910 ("How about building a house with lights?"), the user terminal proceeds to a second action 920. As shown, it contains one or more products (“A floor lamp,” “B table lamp,” “C table lamp”) that match the theme of the selected sale event. Plan product candidates 922 can be output. Based on user input selecting at least one item (“A floor lamp”) from the output proposed item candidates 922, a bargain event can be generated that includes the selected at least one item. Furthermore, the special sale event can be issued to consumers by user input requesting the issuance of the generated special sale event.

FIG. 10 is a diagram illustrating an example in which one or more potential sale events 1000 are provided to a user (eg, operator) according to one embodiment of the present disclosure. According to the embodiments described above, the information processing system can generate one or more sale events. The information processing system can provide one or more bargain sale events generated to the user via the user terminal as bargain event candidates 1000 . The user terminal can display information about one or more sale events corresponding to the sale event candidates 1000 on the display.

As shown in the figure, the user terminal displays "Winter Health Care Food", "Winter Uniform Styling", "Prepare New Year's Gifts in Advance", and "Let's Go to the Ski Resort" as bargain event candidates 1000 for the user. can be output through the interface. For example, the user terminal can identify target consumers for each of the bargain event candidates (e.g., “57-63 year old male”, “18 year old female”, “45-55 year old female”, “20-30 year old male”), Time, images, hashtags (“#Health”, “#Seniors”, “#Students”, etc.), titles of sale events, etc. can be output via the user interface. Further, the user terminal can display a detailed configuration of the selected sale event on the display in response to a user input for selecting at least one sale event from the sale event candidates 1000 .

When the user selects at least one sale event from the sale event candidates 1000 and fulfills a request to publish the selected sale event, the sale event is generated and published, and can be provided to consumers. In FIG. 10, the user terminal outputs four bargain event candidates, but is not limited to this, and can output a different number of bargain event candidates.

The method described above may be provided as a computer program for execution on a computer. The media may be of continuous storage or temporary storage for execution or download of the computer-executable program. In addition, the medium may be various recording means or storage means in the form of single or multiple hardware combined, but it is not limited to a medium directly connected to a computer system, distributed on a network. can exist. Examples of media include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. (magneto-optical medium), ROM, RAM, flash memory, etc., and configured to store program instructions. Examples of other media include recording media and storage media managed by application stores that distribute applications, sites that supply and distribute various software, and servers.

The methods, acts, or techniques of the disclosure may be embodied by various means. For example, such techniques may be implemented in hardware, firmware, software, or any combination thereof. A person of ordinary skill in the art should understand that the various exemplary logic blocks, modules, circuits and algorithmic steps described in the present disclosure can be embodied in electronic hardware, computer software, or a combination of both. be. To clearly illustrate such mutual alternatives of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software will vary depending on the particular application and design requirements imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementations should not be interpreted as departing from the scope of the present disclosure.

In a hardware implementation, the processing unit used to perform the techniques may be one or more of ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PLDs). , field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, and other electronic units designed to perform the functions described in this disclosure , a computer, or a combination thereof.

Thus, the various exemplary logic blocks, modules and circuits described by this disclosure may be general purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or , may be embodied or performed in any combination, including those designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but, in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be embodied in a combination of computing devices, such as a DSP and a microprocessor, multiple microprocessors, one or more microprocessors associated with a DSP core, or any other combination of configurations. .

In a firmware and/or software implementation, the techniques are stored on computer readable media such as RAM, ROM, NVRAM, PROM, EPROM, EEPROM, flash memory, compact discs (CDs), magnetic or optical data storage devices, and the like. It can be embodied by a stored command. The instructions may be executable by one or more processors and may cause the processor(s) to perform certain aspects of the functions described in this disclosure.

Although the foregoing embodiments are described as utilizing aspects of the presently disclosed subject matter in one or more stand-alone computer systems, the disclosure is not so limited, and may be used in networked or distributed computing environments, such as. It can be embodied by any computing environment. Further, aspects of the subject matter of this disclosure may be embodied in multiple processing chips or devices, and storage may be similarly affected across multiple devices. Such devices may also include PCs, network servers and handheld devices.

In the present specification, the present disclosure has been described with some embodiments, but various modifications and changes can be made without departing from the present disclosure that can be understood by a person of ordinary skill in the technical field to which the disclosed invention belongs. It is possible. Also, such variations and modifications should be understood to fall within the scope of the claims appended hereto.

100 User 110 User Terminal 120 Screen

Claims (15)

A sale event generation method performed by at least one computing device comprising:
receiving data associated with multiple product queries;
extracting trends associated with at least one of a particular population or a particular time period based on data associated with the plurality of product queries; and
generating a special sale event including the proposed product associated with the extracted trend. extracting a trend associated with at least one of a specific population or a specific time period based on data associated with the plurality of product queries;
clustering data associated with the plurality of product queries into a plurality of clusters; and
2. The sale event generation of claim 1, comprising extracting one or more trending keywords associated with at least one of the specific group or specific period from one or more crowds among the plurality of crowds. Method. clustering the data associated with the plurality of product queries into a plurality of clusters;
3. The method of claim 2, comprising using a statistical thematic embedding model to cluster data associated with the plurality of product queries into a plurality of crowds. The step of generating a sale event including planned products associated with the extracted trends,
determining one or more products associated with the one or more trend keywords among a plurality of products as the planned products; and
3. The method of generating a sale event according to claim 2, comprising the step of generating a sale event including said planned product. The step of determining one or more products associated with the one or more trend keywords among the plurality of products as the planned products,
determining, based on the one or more trending keywords, a sale event title indicative of the trending subject matter; and
5. The method of generating a bargain sale event according to claim 4, further comprising the step of determining one or more products among said plurality of products as said plan product based on the title of said bargain sale event. Based on the one or more trending keywords, determining a sale event title indicative of the trending subject includes:
6. The method of claim 5, comprising using a language model to convert the one or more trending keywords into a sale event topic indicative of the subject matter of the trend. The step of determining one or more products among the plurality of products as the planned products based on the title of the bargain sale event,
extracting, from among the plurality of products, one or more products that match the title of the bargain sale event using a planned product extraction model; and
6. The bargain sale event generating method according to claim 5, comprising the step of determining said extracted one or more products as said plan product. The planned product extraction model calculates a matching score for each of the plurality of products with the title of the bargain sale event based on information on each of the plurality of products, and based on the calculated matching score, 8. The bargain sale event generation method according to claim 7, wherein a predetermined number of products are extracted from a plurality of products, and a product list including the extracted predetermined number of products is output. The step of determining one or more products among the plurality of products as the planned products based on the title of the bargain sale event,
A step of extracting at least one product that matches the title of the bargain sale event from among the plurality of products using a planned product extraction model;
outputting, via a user terminal, a list including at least one product that matches the topic of the sale event; and
6. The bargain sale event generation according to claim 5, further comprising a step of determining said one or more products as said plan product according to a user's selection of said one or more products from said output list. Method. In a method for learning a proposed product extraction model performed by at least one computer device,
receiving data about a learning sale event;
receiving information about each of the plurality of products; and
Based on the learned sale event data and information on each of a plurality of products, a sale event title is input to output one or more products associated with the sale event title, A plan commodity extraction model learning method, comprising: a step of learning the plan commodity extraction model. The step of learning the planned product extraction model so that when the topic of the special sale event is input, one or more products associated with the topic of the special sale event are output,
When the topic of the sale event is input, a matching score for each of the plurality of products with the topic of the sale event is calculated, and based on the calculated matching score, a predetermined item is selected among the plurality of products. 11. The plan product extraction model learning method according to claim 10, comprising the step of training the plan product extraction model so as to extract a number of products and output a product list containing the extracted predetermined number of products. Based on the learned bargain sale event data and the information about each of the plurality of products, the planning unit outputs one or more products out of the plurality of products by inputting the title of the bargain sale event. The step of learning the product extraction model is
A matching score between the embedded data about the subject of the learning special sale event and the embedded data about the product included in the learning special sale event is a matching score between the embedded data about the topic of the learning special sale event and the learning special sale event. 11. The proposed product extraction model of claim 10, comprising training the proposed product extraction model with a triplet loss function that constrains matching scores to be greater than embedded data for non-included products. learning method. 13. Computer program for executing the method according to any one of claims 1 to 12 on a computer. An information processing system,
communication module;
memory; and
at least one processor coupled to said memory and configured to execute at least one computer readable program contained in said memory;
The at least one program comprises:
receive data associated with multiple product queries;
extracting trends associated with at least one of a specific population or a specific period of time based on data associated with the plurality of product queries;
An information processing system, comprising instructions for generating a bargain sale event including planned products associated with the extracted trend. An information processing system,
communication module;
memory; and
at least one processor coupled to said memory and configured to execute at least one computer readable program contained in said memory;
The at least one program comprises:
Receive data about learning sale events,
receiving information about each of a plurality of products;
Based on the learned sale event data and information on each of a plurality of products, a sale event title is input to output one or more products associated with the sale event title, An information processing system including instructions for learning a planned product extraction model.

Images