CN116523556A

CN116523556A - Method and system for predicting merchant demand

Info

Publication number: CN116523556A
Application number: CN202310599472.2A
Authority: CN
Inventors: 周洋
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2023-05-25
Filing date: 2023-05-25
Publication date: 2023-08-01

Abstract

According to the method and the system for predicting the business demands, the computing equipment of the payment platform can automatically predict the target demands of the target business through the pre-trained business demand prediction model based on the business management data set of the target business, the target demands comprise the business demands aiming at the payment platform, and compared with the manual analysis of the business demands, the method and the system for predicting the business demands are more accurate in the prediction of the business demands, high in prediction timeliness and high in demand prediction yield, and meanwhile manual pressure is relieved.

Description

Method and system for predicting merchant demand

Technical Field

The present disclosure relates to the field of data processing, and in particular, to a method and system for predicting a merchant demand.

Background

Merchants act as important customers of the paymate, which are an important component of the basic disc of payment, and thus the needs of the merchant are very important to the paymate. Taking an offline merchant as an example, the offline merchant is taken as an important asset of a payment platform and is also a service object of offline sales personnel, so that how to effectively identify or predict the demands of the offline merchant, and then help the offline sales personnel and operators to locate the business problem of the merchant, and the targeted establishment of operation services and solutions meeting the demands of the merchant is an important problem. At the same time, looking at industry dynamics, the digitized understanding of customer needs is the first place for professional sales personnel to use the appeal.

Accordingly, there is a need for a method that can effectively predict merchant demand. The statements in this background section merely provide information to the inventors and may not represent prior art to the present disclosure nor may they represent prior art to the filing date of the present disclosure.

Disclosure of Invention

The method and the system for predicting the demands of the merchants can accurately predict the demands of the merchants.

In a first aspect, the present description provides a method of predicting merchant demand for a computing device of a paymate, comprising: determining a store operating dataset of a target merchant, the store operating dataset comprising data characterizing a store operating state; and outputting a target demand predicted for the target merchant through a pre-trained merchant demand prediction model based on the store operation data set, wherein the target demand comprises an operation demand for the payment platform.

In some embodiments, the outputting, based on the store operating dataset, the target demand predicted for the target merchant by a pre-trained merchant demand prediction model includes: determining a feature set of the target merchant based on the store operating dataset; and inputting the feature set into a merchant demand analysis model, and outputting the target demand, wherein the merchant demand prediction model comprises the merchant demand analysis model.

In some embodiments, the feature set includes at least one of a target representation, a target business feature, a target hierarchy feature, a demand feedback feature of the target merchant.

In some embodiments, the store business data set includes store attribute data and commodity business data, the target representation is obtained by inputting the store attribute data and the commodity business data into a merchant representation model, and the merchant demand prediction model includes the merchant representation model.

In some embodiments, the store attribute data includes at least one of a merchant identification, a store location, and a store area of the target merchant.

In some embodiments, the commodity business data includes at least one of a business industry, a commodity name, and a commodity classification of the target merchant.

In some embodiments, the store operating dataset further comprises demand feedback data that is feedback of the target merchant's own demand for the payment platform

In some embodiments, the store business data set includes order data for the target merchant to generate a plurality of trade orders within a preset period of time, the target business characteristics being derived by inputting the order data into a merchant trade model and a consumption preference model, respectively, the merchant demand forecast model including the merchant trade model and the consumption preference model.

In some embodiments, the target business characteristics include merchant transaction characteristics including a total amount of merchandise transactions over the preset period of time, and/or a total amount of orders over the preset period of time, and/or consumption preference characteristics including at least one of: peak period of consumption, average amount of consumption per order, and free-selling goods.

In some embodiments, the target hierarchical features are derived by inputting the target representation and the target business features into a merchant hierarchical model, the target hierarchical features characterizing the liveness of the target merchant in using the paymate, the merchant demand prediction model comprising the merchant hierarchical model.

In some embodiments, the outputting, based on the store operation dataset, the target demand predicted for the target merchant by a pre-trained merchant demand prediction model further comprises: and outputting recommended operation information matched with the target demand based on the marketing strategy model, wherein the merchant demand prediction model comprises the marketing strategy model.

In some embodiments, the recommended operation information includes at least one of equity operation information and equipment paving operation information, where the equity operation information includes equity value, equity type, and equity distribution mode that can be obtained by the paymate completing the order, and the equipment paving operation information includes equipment recommended to pave the paymate.

In some embodiments, the recommended business information is obtained by inputting the target demand into the marketing strategy model.

In some embodiments, the recommended business information is obtained by inputting the target demand and at least one of the target representation, the target business feature, and the target hierarchical feature into the marketing strategy model.

In some embodiments, the determining the store operating dataset of the target merchant comprises: and identifying multi-mode data to obtain the store management data set, wherein the multi-mode data at least comprises two of video, image, audio, text and geographic position.

In a second aspect, the present specification also provides a system for predicting merchant demand, comprising: at least one storage medium storing at least one set of instructions for effecting predictions of merchant demand; and at least one processor communicatively coupled to the at least one storage medium, wherein the at least one processor reads the at least one instruction set and implements the method of predicting merchant demand of the first aspect when the system for predicting merchant demand is running.

According to the method and the system for predicting the business demand, which are provided by the specification, the computing equipment of the payment platform can automatically predict the target demand of the target business based on the business operation data set of the target business through the pre-trained business demand prediction model, the target demand comprises the business demand aiming at the payment platform, compared with the manual analysis of the business demand, the method and the system for predicting the business demand are more accurate, and are high in prediction timeliness and demand prediction yield, and meanwhile, the manual pressure is relieved. Meanwhile, the merchant demand prediction model can also match corresponding recommended operation information for target demands, and assist operators in formulating more efficient and scientific operation strategies. Meanwhile, the computing equipment can predict the target requirement by identifying multi-mode data such as video, image, audio, text and the like of the target merchant, and the problems of low quality and poor authenticity of manually collected data are solved. Meanwhile, the computing equipment of the payment platform can update the merchant demand prediction model in real time based on real-time order data of the target merchant so as to adjust the predicted target demand in real time, and accuracy of demand prediction is guaranteed.

Additional functionality of the business need prediction methods and systems provided herein will be set forth in part in the description that follows. The following numbers and examples presented will be apparent to those of ordinary skill in the art in view of the description. The inventive aspects of the business need prediction methods and systems provided herein may be fully explained by practicing or using the methods, devices, and combinations described in the detailed examples below.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present description, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 illustrates a schematic view of an application scenario of a business need prediction system provided in accordance with some embodiments of the present description;

FIG. 2 illustrates a hardware architecture diagram of a computing device provided in accordance with some embodiments of the present description;

FIG. 3 illustrates a flow chart of a method of predicting merchant demand provided in accordance with some embodiments of the present description;

FIG. 4 illustrates a schematic diagram of a multi-modal data collection and identification provided in accordance with some embodiments of the present disclosure;

FIG. 5 illustrates a flow chart of a method of predicting merchant demand provided in accordance with some embodiments of the present description; and

FIG. 6 illustrates a schematic diagram of a business demand prediction model output target demand provided in accordance with some embodiments of the present description.

Detailed Description

The following description is presented to enable one of ordinary skill in the art to make and use the invention, and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the disclosure. Thus, the present description is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the claims.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. For example, as used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. The terms "comprises," "comprising," "includes," and/or "including," when used in this specification, are taken to specify the presence of stated integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

These and other features of the present specification, as well as the operation and function of the related elements of structure, as well as the combination of parts and economies of manufacture, may be significantly improved upon in view of the following description. All of which form a part of this specification, reference is made to the accompanying drawings. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the description. It should also be understood that the drawings are not drawn to scale.

The flowcharts used in this specification illustrate operations implemented by systems according to some embodiments in this specification. It should be clearly understood that the operations of the flow diagrams may be implemented out of order. Rather, operations may be performed in reverse order or concurrently. Further, one or more other operations may be added to the flowchart. One or more operations may be removed from the flowchart.

Before describing the specific embodiments of the present specification, the application scenario of the present specification will be described as follows:

the method provided by the specification can be used for predicting the demand of online merchants and also predicting the demand of offline merchants. Taking an offline merchant as an example, in order to consolidate partnership with customers, sales personnel of a paymate (e.g., a payroll) need to frequently visit the offline merchant. In the visit, the sales personnel can check the business condition of the merchant, inquire about problems and difficulties encountered by the merchant in the process of using the product or service of the payment platform, know business appeal fed back by the merchant and the like, and meanwhile, the sales personnel can collect and record the data by using the mobile workbench. The computing device of the paymate may obtain these data from the mobile workstation, which may be analyzed by a pre-trained business demand prediction model to predict the actual demands of the business. Operators of the payment platform can formulate corresponding operation strategies for the real demands and implement the operation strategies in a landing mode to serve merchants, so that the real demands of the merchants are met. In this process, it is a very important link to accurately predict the actual needs of merchants. In order to accurately predict the demands of merchants, various factors of the merchants need to be considered in all directions, the data volume of one merchant is huge, certain rules necessarily exist between huge data and the real demands of the merchant, and the merchant demand prediction model in the specification can mine the rules and provide the final real merchant demands.

For convenience of description, the present specification explains terms appearing in the context description:

and (3) a client: merchants who sign up with the payment platform in cooperation with the service are completed.

Potential customers: merchants who have not contracted with the paymate for service are the intended customers of the paymate.

BD (Business Development, business expansion): sales representatives, referred to herein primarily as off-line sales personnel, have the responsibility of providing services, such as order collection and operation, to merchants by touching the merchant off-line, thereby driving the merchant to benefit.

Middle station operator/operator: refers to personnel responsible for operation of a certain product and making an operation policy.

And (3) expanding: refers to the process of visiting and contracting on business intention cooperation by staff of a payment platform.

Multimodal data: refers to different types of data, and by combining the multi-mode data together, the deep learning model capacity can be enhanced, and the accuracy is improved. Multimodal data in this specification includes at least two of video, image, audio, text, and geographic location.

LBS: location-based services (Location Based Services, LBS) are devices that utilize various types of positioning techniques to obtain the current location of a positioning device, and in this specification refer primarily to the geographic location of a merchant.

And (3) an image-text recognition platform: a platform for accurate image content recognition services is provided.

Semantic recognition platform: platform for natural language processing (Natural Language Processing, NLP).

An algorithm platform: the platform for training and verifying the service can be used for machine learning, deep learning, convolutional neural network and other algorithms.

And (3) moving a workbench: the BD can record the visit content through the mobile workbench in the visiting business process, and a visiting service APP can be installed on the BD.

Fig. 1 illustrates an application scenario diagram of a business need prediction system 001 provided according to some embodiments of the present disclosure. As shown in fig. 1, the application scenario may include a system 001, sales person 110, target merchant 120, and network 400. The system 001 may include a mobile station 200, a terminal 300, and a server 500.

Sales force 110 of the paymate may carry mobile workstation 200 to visit target merchant 120. The paymate may be a payment bridge for online transactions between the target merchant 120 and consumers of the target merchant 120. When the consumer pays for an order by the target merchant 120, the computing device of the payment platform may acquire a payment account of the target merchant 120 and a payment account of the consumer, and transfer money corresponding to the order from the payment account to the payment account, thereby completing the transaction of the target merchant 120 and the consumer for the order. To enhance collaboration with the customer, the paymate may provide business services, such as payment services, to target merchants 120. Specifically, the target merchant 120 may use a cashing device (such as a face-brushing payment device) provided by the payment platform to settle each order, or may lay a payment two-dimensional code of the payment platform for the consumer to pay by scanning the code, etc. It should be noted that, the merchant data obtained in the present description is authorized by the target merchant 120, and does not relate to the merchant privacy.

Target merchant 120 may be a customer or potential customer of the paymate. The target merchant 120 may be a merchant of any industry, such as in the catering industry, the target merchant 120 may be a store-less street booth, store-with restaurant, etc.; in the retail industry, the target merchant 120 may be a retail store, a supermarket, or the like.

Mobile station 200, terminal 300 and server 500 all belong to the devices of the paymate. In some embodiments, the system 001 may include a mobile workstation 200 and a server 500. At this point, the computing device of the paymate may be mobile workstation 200 or server 500. When the computing device of the paymate is mobile workstation 200, mobile workstation 200 can both record the visit content when visiting target merchant 120 and predict the target demand. Wherein the visit content can be multimodal data such as video, images, audio, text, geographic location, and the like. In this case, the method of predicting merchant demand may be performed on mobile workstation 200. At this time, the mobile station 200 may store data or instructions for performing the method of predicting a business demand described in the present specification, and may execute or be used to execute the data or instructions. In some embodiments, mobile station 200 may include a hardware device having data information processing functions and programs necessary to drive the hardware device to operate. The server 500 at this time is used to provide a background service for the mobile station 200, such as a background server that provides support for pages displayed on the mobile station 200. When the paymate computing device is server 500, mobile station 200 may record the visit content and send the visit content to server 500, server 500 may predict the target demand and may send the target demand to mobile station 200 for display. In this case, the method of predicting merchant demand may be performed on the server 500. At this time, the server 500 may store data or instructions for performing the method of predicting a business demand described in the present specification, and may execute or be used to execute the data or instructions. In some embodiments, the server 500 may include a hardware device having a data information processing function and a program necessary to drive the hardware device to operate.

Wherein the mobile station 200 may comprise an image acquisition device. In the event that approval by the target merchant 120 is obtained, the sales person 110 may use the image capturing device to capture video or images of the target merchant 120, such as capturing video or images of the store environment, store fronts, merchandise display conditions, and situations where the sales person 110 is talking to a staff (e.g., responsible person) of the target merchant 120. Mobile station 200 may also record, for example, audio when sales person 110 is talking to the responsible person of target merchant 120. The mobile station 200 may have a text recording function, and the sales person 110 may record the contents fed back by the responsible person in the form of text using the mobile station 200. Of course, the person in charge may input the content to be expressed by himself/herself in the form of text using the mobile station 200. The mobile table 200 may also have a positioning function, and the position of the mobile table 200 may be positioned. When mobile station 200 is at target merchant 120, the geographic location of target merchant 120 may be located and recorded.

In some embodiments, the system 001 may include a mobile workstation 200, a terminal 300, and a server 500, in which case the computing device of the paymate may be the terminal 300. The mobile station 200 may record the visit content and send the visit content to the terminal 300, and the terminal 300 may predict and display the target demand. The server 500 is used to provide a background service for the terminal 300, for example, a background server that provides support for pages displayed by a target APP on the terminal 300. In this case, the method of predicting merchant demand may be performed on terminal 300. At this time, the terminal 300 may store data or instructions for performing the method of predicting a merchant demand described in this specification, and may execute or be used to execute the data or instructions. In some embodiments, the terminal 300 may include a hardware device having a data information processing function and a program necessary to drive the hardware device to operate. As shown in fig. 1, the terminal 300 may be communicatively connected to a server 500. In some embodiments, the server 500 may be communicatively coupled to a plurality of terminals 300. In some embodiments, the terminal 300 may interact with the server 500 through the network 400 to receive or transmit messages, etc.

In some embodiments, mobile station 200 and/or terminal 300 may be a mobile device, a tablet, a notebook, a built-in device of a motor vehicle, or the like, a vending machine, a vending cabinet, or any combination thereof. In some embodiments, the mobile device may include a smart home device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home device may include a smart television, a desktop computer, or the like, or any combination. In some embodiments, the smart mobile device may include a smart phone, personal digital assistant, gaming device, navigation device, etc., or any combination thereof. In some embodiments, the virtual reality device or augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality patch, augmented reality helmet, augmented reality glasses, augmented reality patch, or the like, or any combination thereof. For example, the virtual reality device or the augmented reality device may include google glass, head mounted display, VR, or the like. In some embodiments, the built-in devices in the motor vehicle may include an on-board computer, an on-board television, and the like. In some embodiments, mobile station 200 and/or terminal 300 may have one or more of the following functions: NFC (Near Field Communication ), WIFI (Wireless Fidelity, wireless fidelity), 3G/4G/5G, POS (Point Of Sale) machine card swiping function, two-dimensional code scanning function, bar code scanning function, bluetooth, infrared, SMS (Short Message Service), MMS (Multimedia Message Service, multimedia message).

In some embodiments, mobile station 200 and/or terminal 300 may be equipped with one or more Applications (APP) capable of providing sales force 110 with the ability to interact with the outside world via network 400, as well as an interface. The APP includes, but is not limited to: web browser-like APP programs, search-like APP programs, chat-like APP programs, shopping-like APP programs, video-like APP programs, financial-like APP programs, instant messaging tools, mailbox clients, social platform software, and the like. In some embodiments, a visiting service APP may be installed on the mobile station 200. Sales person 110 can obtain multi-modal data such as video, image, audio, text, geographic location, etc. of target merchant 120 through the visit service APP. In some embodiments, the visited service APP is also capable of identifying multimodal data. In some embodiments, the terminal 300 may have a target APP installed thereon. The staff of the payment platform can trigger a merchant demand prediction request through the target APP. The target APP may perform a method of predicting the merchant demand in response to the merchant demand prediction request.

The server 500 may be a server providing various services. The server 500 may interact with the terminal 300 or the mobile station 200 through the network 400. The server 500 may be communicatively connected to a plurality of terminals 300 or a plurality of mobile stations 200, and receive data transmitted from the terminals 300 or the mobile stations 200.

Taking the example of server 500 interacting with terminal 300, network 400 serves as a medium for providing a communication connection between terminal 300 and server 500. The network 400 may facilitate the exchange of information or data. As shown in fig. 1, the terminal 300 and the server 500 may be connected to the network 400 and transmit information or data to each other through the network 400. In some embodiments, the network 400 may be any type of wired or wireless network, or a combination thereof. For example, network 400 may include a cable network, a wired network, a fiber optic network, a telecommunications network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Public Switched Telephone Network (PSTN), a bluetooth network, a ZigBee network, a Near Field Communication (NFC) network, or the like. In some embodiments, network 400 may include one or more network access points. For example, the network 400 may include a wired or wireless network access point, such as a base station or an internet switching point, through which one or more components of the terminal 300 and the server 500 may be connected to the network 400 to exchange data or information.

It should be understood that the number of mobile stations 200, terminals 300, servers 500, and networks 400 in fig. 1 are merely illustrative. There may be any number of mobile stations 200, terminals 300, servers 500, and networks 400, as desired for implementation.

It should be noted that, the method for predicting the needs of the merchant may be performed entirely on the mobile workbench 200, may be performed entirely on the terminal 300, or may be performed entirely on the server 500.

Fig. 2 illustrates a hardware architecture diagram of a computing device 600 provided in accordance with some embodiments of the present description. Computing device 600 may perform the method of predicting merchant demand described herein. The method of predicting merchant demand is described elsewhere in this specification.

As shown in fig. 2, computing device 600 may include at least one storage medium 630 and at least one processor 620. In some embodiments, computing device 600 may also include a communication port 650 and an internal communication bus 610. Meanwhile, computing device 600 may also include I/O component 660.

Internal communication bus 610 may connect the various system components including storage medium 630, processor 620, and communication ports 650.

I/O component 660 supports input/output between computing device 600 and other components.

The communication port 650 is used for data communication between the computing device 600 and the outside world, for example, the communication port 650 may be used for data communication between the computing device 600 and the network 400. The communication port 650 may be a wired communication port or a wireless communication port.

The storage medium 630 may include a data storage device. The data storage device may be a non-transitory storage medium or a transitory storage medium. For example, the data storage devices may include one or more of magnetic disk 632, read Only Memory (ROM) 634, or Random Access Memory (RAM) 636. Storage medium 630 may store at least one set of instructions for implementing predictions of merchant demand. The instructions are computer program code that may include programs, routines, objects, components, data structures, procedures, modules, etc. that perform the methods of predicting merchant demand provided herein. The storage medium 630 may also store a business demand prediction model for implementing a business demand prediction method. At this point, the model may be one or more sets of instructions stored in the storage medium 630 that execute the corresponding instructions and are executed by the processor 620 in the computing device 600. Of course, the model may also be part of a circuit, hardware device, or module in computing device 600. For example, the business demand prediction model may be a hardware device/module in computing device 600 that implements business demand prediction. At this point, processor 620 may have stored therein at least one set of instructions or instruction sets for controlling the model.

The at least one processor 620 may be communicatively coupled with at least one storage medium 630 and a communication port 650 via an internal communication bus 610. The at least one processor 620 is configured to execute the at least one instruction set. When the computing device 600 is running, the at least one processor 620 may read the at least one instruction set and, as indicated by the at least one instruction set, perform the method of predicting merchant demand provided herein. Processor 620 may perform all of the steps involved in the method of predicting merchant demand. The processor 620 may be in the form of one or more processors, and in some embodiments, the processor 620 may include one or more hardware processors, such as microcontrollers, microprocessors, reduced Instruction Set Computers (RISC), application Specific Integrated Circuits (ASICs), application specific instruction set processors (ASIPs), central Processing Units (CPUs), graphics Processing Units (GPUs), physical Processing Units (PPUs), microcontroller units, digital Signal Processors (DSPs), field Programmable Gate Arrays (FPGAs), advanced RISC Machines (ARM), programmable Logic Devices (PLDs), any circuit or processor capable of executing one or more functions, or the like, or any combination thereof. For illustrative purposes only, only one processor 620 is depicted in the computing device 600 in this specification. It should be noted, however, that computing device 600 may also include multiple processors, and thus, operations and/or method steps disclosed in this specification may be performed by one processor as described herein, or may be performed jointly by multiple processors. For example, if the processor 620 of the computing device 600 performs steps a and B in this specification, it should be understood that steps a and B may also be performed by two different processors 620 in combination or separately (e.g., a first processor performs step a, a second processor performs step B, or the first and second processors perform steps a and B together).

Fig. 3 illustrates a flow chart of a method P100 for predicting merchant demand provided in accordance with some embodiments of the present description. As previously described, computing device 600 may perform a method of predicting merchant demand P100 described herein. Specifically, the processor 620 may read the instruction set stored in its local storage medium and then execute the method P100 for predicting merchant demand described in this specification according to the specification of the instruction set. As shown in fig. 3, the method P100 may include:

s120: a store operating dataset for the target merchant 120 is determined.

The store operating dataset includes data characterizing a store operating state. The sales person 110, when visiting the target merchant 120, can record the visiting content using the mobile workstation 200 with consent from the target merchant 120. Mobile workstation 200 may gather multimodal data for target merchant 120, which may include at least two of video, image, audio, text, and geographic location. For example, sales person 110 may use mobile workstation 200 to capture video and/or images of target business 120, such as video and/or images of a shop interior and a shop exterior door facing. Sales force 110 may also talk to staff (e.g., responsible) of target merchant 120, such as asking target merchant 120 for problems with the business store, problems with the products and/or services offered using the paymate, which complaints are currently made with the paymate's products and/or services, and so forth. Mobile station 200 may record the conversation content for audio. Sales person 110 may also enter text into the mobile station to record the conversation content. Mobile station 200 may also locate and obtain the geographic location of target merchant 120. FIG. 4 illustrates a schematic diagram of a multi-modal data collection and identification provided in accordance with some embodiments of the present description. As shown in fig. 4, BD personnel of the paymate obtain 5 modalities of data such as text appeal, visit audio, business video, business photograph, and geographic location while using the mobile station 200.

The visiting content is collected by the method, data collection quality is improved through multi-mode data, meanwhile, collection dimensions are more diversified, information which can be utilized for target demand prediction is richer, and accordingly accuracy of target demand prediction is improved. Moreover, the capacity of different sellers is uneven, the collection quality of the visit content depends on the professional capacity of the sellers to a certain extent, and if the visit content is collected through single-mode characters, the problems that the quality of the visit content fed back by the sellers is not high, the visit content depends on the professional capacity of individuals and the authenticity is doubtful exist. By collecting video, images and/or audio, the data is more objective, the authenticity of the visit content is improved, and the quality of the abundant multi-mode data is higher than that of the single-mode data.

In some embodiments, mobile station 200 may upload the multimodal data to computing device 600 for identification. For example, the computing device 600 may perform image-text recognition on the video and/or the image through the image-text recognition platform, and may further perform verification recognition on the geographic location, so as to obtain store attribute data and commodity management data. The store operating data set includes the store attribute data and the commodity operating data. The computing device 600 may perform image-text recognition through an algorithm such as CNN, GAN, HOG on the image-text recognition platform. In some embodiments, mobile station 200 may also identify the multimodal data itself and send the identification to computing device 600. For ease of description, the following description will be presented with computing device 600 identified as an example.

The store attribute data may include one of a merchant identifier, a store location, and a store area of the target merchant 120, and may further include other store information such as a store owner name (e.g., a name of a responsible person), a number of staff members, and the like. The merchant identifier can be a shop name or a combination of the shop name and other information, such as a combination of the shop name and a shop position, so that the problem of inaccurate target demand prediction caused by information alignment errors due to shop name renaming is avoided. The merchant identification may also be a merchant ID encoded by computing device 600 for target merchant 110, such as merchant No. 1, merchant No. 2, merchant No. 3, and so on.

The shop location may be obtained through verification and identification of the geographical location, and the shop location and the geographical location may be the same or different. There may be a discrepancy between the geographic location where mobile station 200 is located and the actual store location of target merchant 110, e.g., the geographic location of the next-door store to target merchant 110 is located. The geographic location of mobile station 200 may also be less accurate, e.g., to the province, city, district, etc. to which target merchant 110 belongs. Thus, computing device 600 may perform verification of the geographic location, such as verifying that the geographic location is accurate in conjunction with a merchant identification, or manually. Of course, computing device 600 may not verify the geographic location, but instead directly treat the geographic location as the store location, which is not limited by the present embodiments. The store area may be the size of the store space of target merchant 110 or may be a planar area of the store floor, which may be indicative of the size of target merchant 110. In order to attract consumers, the store door shots of some merchants are novel and unique, so in order to preserve certain variability of the target merchant 110, when the mobile workbench 200 shoots the door shots of the store of the target merchant 110, the computing device 600 may also directly preserve the images of the door shots and determine the images as store attribute data instead of performing image-text recognition on the door shots.

The commodity operation data may include at least one of an operation industry, a commodity name, and a commodity classification of the target merchant 120, and may further include a commodity price, a commodity placement position, and the like. Wherein the business industry is, for example, the catering industry, the retail industry, the clothing industry, the beauty industry, etc. The business industry may be determined by computing device 600 based on store name, commodity name, and/or commodity classification. For the same name of commodity, computing device 600 may retain only one commodity name to remove redundancy. The commodity classification may be a classification of the commodity in the target merchant 120, for example, in a small business, the commodity classification may include beverage, snack, seasoning, instant, household, and the like. The commodity classification may be determined by computing device 600 based on a commodity name and/or a commodity classification reference table, which may be a classification table recognized in the business industry to which target merchant 120 belongs.

The computing device 600 may also convert the audio to text, use both the audio-converted text and the recorded complaint text as recognition text, and perform semantic recognition and understanding via the semantic recognition platform, thereby obtaining the demand feedback data. Upon semantic recognition and understanding, the computing device 600 may extract keywords and sentence dependencies in the text via a semantic recognition platform, thereby obtaining the demand feedback data. Wherein the algorithm for semantic recognition and understanding may be one or more of depth LSTM, RNN, multi-headed attention mechanism. Prior to semantic recognition and understanding, computing device 600 may perform redundancy elimination operations on the audio-converted text and/or recorded offending text to avoid duplicate text as much as possible, thereby reducing the burden of demand prediction.

The demand feedback data may be feedback content of the target merchant 110 for its own demands, such as participation demands of marketing campaigns participating in the paymate, improvement demands on products and/or services of the paymate, and so on. For example, the recognition text is 'double eleven activities in the former period' with good effect, and the trade GMV of our store is promoted. To participate in the next twenty-two marketing campaign, "keywords extracted by the computing device 600 include" twenty-one, "" campaign effects, "" good, "" boost, "" transaction GMV, "" participate, "" twenty-two, "and" marketing campaign, "and the extracted inter-sentence dependencies are the relationships between" twenty-one "and" twenty-two, "and the resulting demand feedback data may be" want to participate in the twenty-two marketing campaign in order to boost the transaction GMV. For another example, the target merchant 120 is a street stall, the recognition text is "i am out at night now, the light is not good, the customer scans the two-dimension code too long, business is affected, i want the customer to pay faster. The requirement feedback data can be "want to solve the code scanning problem caused by poor light at night".

As shown in fig. 4, the multi-modal data includes feedback text content, feedback video content, feedback audio content, and feedback image content, and the computing device 600 processes the multi-modal data by applying the operations of the image-text visual platform, the NLP platform, and the audio-to-text to obtain the demand feedback content, the merchant identification, the location of the merchant, the area of the merchant, the business industry, the name of the merchant, the category of the merchant, the door photograph, and the responsible person.

It should be noted that, the image-text recognition platform and the semantic recognition platform may be software on the computing device 600 or may be an independent hardware device, and if the hardware device is an independent hardware device, the computing device 600 may send the video and/or the image to the hardware device of the image-text recognition platform to perform image-text recognition, and send the recognition text to the hardware device of the semantic recognition platform to perform semantic recognition and understanding.

With consent from the target merchant 120, the computing device 600 may obtain order data for the target merchant 120 in real-time. For example, target merchant 120 may be credited using a cashier device of a paymate, and computing device 600 of the paymate may obtain order data for target merchant 120 in real-time. As such, the store operating dataset may include order data for the target merchant 120 to generate multiple trade orders within a preset period of time. The preset period of time may be one day, one week, one month, one year, etc. The order data includes, for example, the amount of money consumed, the goods consumed, and/or the time of consumption per order, etc.

The computing device 600 may store data in the identified store operating dataset, such as store attribute data and commodity operating data in a merchant archive, demand feedback data in a merchant feedback demand store, and order data in a merchant transaction database. To effectively manage merchant data, prior to storage, computing device 600 may concatenate the data to be stored through a join syntax based on the merchant identification (e.g., merchant ID) of target merchant 120. For example, as shown in FIG. 4, computing device 600 obtains a merchant archive and a merchant feedback demand store via identification and storage of multimodal data, and performs join operations on the multidimensional data that needs to be stored prior to storage.

If the acquired multi-modal data is directly stored and the multi-modal data is manually analyzed, the manual failure of effective analysis and identification can be caused because of the fact that the visit content is more, mixed and unstructured; moreover, nearly tens of thousands of sales personnel off-line, the volume of the visit content fed back is very large, and human analysis and identification have no sustainability. The computing device 600 of the present disclosure may automatically identify the multimodal data using an algorithm, thereby avoiding the problems of poor efficiency and unsustainable human analysis and identification. Moreover, the present description also enables the precipitation of valuable databases, such as merchant archives, merchant feedback demand libraries, and merchant transaction databases, facilitating subsequent business developments.

S140: outputting a target demand predicted for the target merchant 120 by a pre-trained merchant demand prediction model based on the store operating dataset.

In some embodiments, the target demand may be an operational demand for a paymate. For example, it may be desirable to engage in a marketing campaign related to paymate, such as a holiday promotion campaign, a double 11 campaign, a double 12 campaign, and so forth. For example, there is a need to add payment means for paymate, such as adding paymate development cash-back devices, adding services to use paymate cash-back (i.e., a consumer may use paymate APP to make a payment at target merchant 120), and so forth. For example, a payment product of a payment platform needs to be replaced/added, such as an entity two-dimensional code (the consumer can use the payment platform APP to scan the two-dimensional code for payment), a code scanning device (the target merchant 120 can use the code scanning device to scan the electronic two-dimensional code of the payment platform presented by the consumer), and a self-checkout IOT device (the consumer can self-checkout the commodity on the device). As another example, there is a need to improve paymate products, such as repairing cashier devices that are often stuck. In some embodiments, the target demand may not be an operational demand for a paymate. For example, a large quantity of items of merchandise is sold by a target merchant 120, whose target demand is to sell the products for sale. As another example, the target merchant 120 needs to promote daily average GMV (Gross Merchandise Volume, total amount of deals).

In some embodiments, the merchant demand prediction model may be a single model. Computing device 600 may input a store business data set into the merchant demand prediction model, outputting a target demand without determining the feature set. For example, computing device 600 may input all data in the store business data set into the merchant demand prediction model, outputting a target demand. In this way, the efficiency of demand prediction is improved by omitting the step of determining the feature set. For another example, computing device 600 may input demand feedback data in the store business data set into the merchant demand prediction model, outputting a target demand. In this way, the merchant demand prediction model may obtain the target demand based on the demand dictated by the target merchant 120, and because the authenticity of the demand dictated by the target merchant 120 is higher, the accuracy of the target demand is higher, and at the same time, the speed and efficiency of predicting the target demand based on the demand feedback data are faster.

In some embodiments, the merchant demand prediction model may be a fusion model fused from a plurality of sub-models. The plurality of sub-models may be, for example, a business representation model, a business transaction model, a business layering model, a consumption preference model, a business demand analysis model, a marketing strategy model, and the like. The business demand prediction model can be obtained by fusing any one or more of a business demand analysis model and the rest of sub-models. For example, the business demand prediction model may be obtained by fusing a business demand analysis model, a business portrait model and a consumption preference model, may be obtained by fusing a business demand analysis model, a business transaction model and a business layering model, may be obtained by fusing a business demand analysis model, a business portrait model, a business transaction model and a consumption preference model, may be obtained by fusing a business demand analysis model, a business portrait model, a business transaction model and a marketing strategy model, and may be obtained by fusing all 6 sub-models of a business demand analysis model, a business portrait model, a business transaction model, a business layering model, a consumption preference model and a marketing strategy model. Wherein the network structure of each sub-model in the business portrayal model, the business transaction model, the consumption preference model, the business layering model, the marketing strategy model may be a combination of one or more of LSTM, CNN, GPT. The network structure of the merchant demand prediction model may be GPT-2 or GPT-3.

Wherein the fusion may be a fusion training, i.e., computing device 600 may perform fusion training on the plurality of sub-models to obtain a trained business demand prediction model. The fusion training may be a stacked fusion training, i.e. the output of one or more sub-models for which training is completed is used as training data for training another sub-model. For example, a business representation model and a business transaction model are trained first, and then the outputs of the trained business representation model and the trained business transaction model are used as training data to train a business layering model. The functional advantages of the multiple sub-models can be integrated through the stacking fusion training, and the business demand prediction model obtained through the fusion training has all functions of the sub-models participating in the fusion training. For example, after the above 6 sub-models are fused, the business demand prediction model has the function of determining the target portrait, business transaction characteristics, consumption preference characteristics, target hierarchical characteristics, recommended operation information and target demand of the target business 120. It should be noted that, the execution subject of the training process and the execution subject of the prediction method for executing the merchant demand referred to in this specification may be the same or different, and both execution subjects are the same in this specification and are described by taking the computing device 600 as an example.

The following describes a method for outputting target demand by taking a method for predicting a demand of a merchant as an example, where the model is obtained by fusing the 6 sub-models, and fig. 5 shows a flowchart of a method S140 for predicting a demand of a merchant according to some embodiments of the present disclosure, and as shown in fig. 5, the method S140 may include:

s142: a feature set of the target merchant 120 is determined based on the store operating dataset.

FIG. 6 illustrates a schematic diagram of a business demand prediction model output target demand provided in accordance with some embodiments of the present description. As shown in fig. 6, computing device 600 may input the store attribute data and the commodity operation data into a pre-trained merchant representation model, outputting a target representation of target merchant 120 that may be understood as a label of numerous store attribute data and commodity operation data. The feature set may include the target representation. Since the store attribute data and the commodity operation data are the contents of the identified multi-mode data, are numerous and heterogeneous, are not related to each other, and cannot well characterize the target merchant 120, the store attribute data and the commodity operation data can be classified, combined and extracted into valuable words through the merchant portrait model, so that a highly refined tag is obtained. For example, the responsible person in FIG. 4 may not have value for predicting target demand, and therefore may be filtered out by the merchant portrayal model. For another example, the number of commodity names may be very large, and representative commodity names may be refined by the merchant portrayal model. In this way, computing device 600 can utilize refined merchant tags to predict target demand, improving the efficiency of demand prediction. The training data of the merchant portrayal model during training can comprise store attribute sample data and commodity operation sample data of a plurality of training sample merchants.

It should be noted that, the computing device 600 may perform feature extraction on the store attribute data and the commodity management data through a feature platform to obtain corresponding feature vectors, and store the feature vectors into a merchant feature library, so as to input the merchant feature library into the merchant portrait model to determine the target portrait. Thus, the target demand is predicted by the feature vector in the merchant feature library, and the efficiency of demand prediction can be improved. The feature platform may be a software algorithm on the computing device 600 or a hardware device for feature extraction. When the feature platform is a hardware device, computing device 600 may interact with the feature platform.

As shown in fig. 6, computing device 600 may input the order data for target merchant 120 into a pre-trained merchant transaction model, outputting merchant transaction characteristics for target merchant 120. The feature set may include the merchant transaction features. The merchant transaction characteristic may include a total amount of merchandise transactions within the preset period of time and/or a total amount of orders within the preset period of time. For example, the merchant transaction model outputs a total of two tens of thousands of items for one week and a total of 500 orders for one week according to the order data generated within one week of the target merchant 120. The merchant transaction characteristic may also include other transaction characteristics within the predetermined period of time, such as a total amount of merchandise transactions on weekdays, a total amount of merchandise transactions on holidays, and so forth. Wherein the merchant transaction model may be trained from historical order data of a training sample merchant, which may include target merchants 120.

As shown in fig. 6, computing device 600 may input the order data into a pre-trained consumption preference model, outputting consumption preference characteristics of target merchant 120. The feature set may include the consumption preference feature. The consumption preference feature may include at least one of the following within the preset period of time: peak period of consumption, average amount of consumption per order, and free-selling goods. For example, the consumption preference model outputs the order data generated within a week of the target merchant 120 with the peak period of consumption between 18:00-20:00 evening for the week, the average amount of consumption per order being 40 yuan, the free selling goods being tomatoes and vegetables. The consumption preference feature may also include other preference features within the preset period of time, such as, for example, free-selling items in each commodity category, etc. Wherein the consumption preference model may be trained from the historical order data.

Computing device 600 may input one or more of the target representation, the merchant transaction feature, the consumption preference feature into a merchant hierarchical model to output target level features. The feature set may include the target-level features. As shown in fig. 6, computing device 600 inputs the target representation, merchant transaction features, and consumption preference features into a pre-trained merchant hierarchical model, outputting target hierarchy features. Wherein the target tier characteristic characterizes how active the target merchant 120 is in using the paymate, such as high active, medium active, low active, to be activated, and the like. The training data of the merchant hierarchical model may include one or more of merchant image samples, merchant transaction sample features, consumption preference sample features.

The feature set may include at least one of a target representation, a target business feature, a target tier feature, a demand feedback feature of the target merchant 120. The target business features include merchant transaction features and/or consumption preference features. The demand feedback characteristics may be the same as the demand feedback data described above. It should be noted that, the data in the feature set may be obtained by other methods, such as a mathematical formula, which is not limited in this specification, instead of the above model.

S144: inputting the feature set into a merchant demand analysis model, and outputting the target demand.

The merchant demand prediction model includes the merchant demand analysis model. Computing device 600 may input the feature set into the merchant demand analysis model, outputting the target demand. As shown in FIG. 6, computing device 600 inputs the target representation, target business characteristics, target tier characteristics, and demand feedback characteristics into the pre-trained business demand analysis model, outputting the target demand. The business demand analysis model can be obtained through training of a manual demand library, training data and corresponding real demands of the training data can be included in the manual demand library, the training data can be one or more of business portrait samples, business sample characteristics, business level samples and demand feedback sample characteristics, and the business sample characteristics comprise business transaction sample characteristics and/or consumption preference sample characteristics.

It should be noted that, the merchant demand prediction model may also output recommended operation information matched with the target demand. Wherein the recommended operation information may include one of equity operation information and equipment paving operation information. The rights management information comprises a rights value, a rights type and a rights distribution mode which can be obtained by completing an order through the payment platform. The equity value may be an equity amount, such as 1-gram, 8.8-gram, etc. The equity type may be a ticket, cash, red envelope, etc. The rights distribution means may be return, discount, deduction, etc. The equipment laydown management information may be recommended information about the products/services of the paymate, such as recommending equipment to lay the paymate, such as: the current entity two-dimensional code of the target merchant 120 is recommended to be updated into a code scanning device, or the target merchant 120 is recommended to use the latest cash register device of the payment platform, or the target merchant 120 is recommended to introduce a self-service settlement IOT device of the payment platform, and the like.

In some embodiments, the recommended business information is derived based on the target demand. For example, the computing device 600 may input the target requirements into a pre-trained business strategy model, and output the recommended business information, i.e., the matched recommended business information is given according to the target requirements. For example, the target requirement is "a cashier device to be developed by adding a paymate", and the corresponding recommended operation information may be "a latest cashier device to be set for the target merchant 120".

In some embodiments, the recommended business information may be derived based on the target demand and at least one of the target representation, the target business feature, and the target hierarchy feature. As shown in fig. 6, computing device 600 may input the target representation, the merchant transaction feature, the consumption preference feature, the target tier feature, and the target demand into a pre-trained business strategy model, outputting the recommended business information. For example, if the target merchant 120 is merchant a, the target requirement is "participate in a double 12 offline large promotion" and the corresponding recommended operation information may be "participate in a full 10 return 1-membered shopping coupon activity". For another example, if the target merchant 120 is merchant B, the target requirement is "participate in the double 12-line large promotion", and the corresponding recommended operation information may be "provide full consumption 500 enjoying 8.8-fold payment rights". For another example, if the target merchant 120 is merchant C, the target requirement is "participate in a double 11 online large promotion" and the corresponding recommended operation information may be "participate in a shopping full 1000 deduction 100 red pack activity".

The operation strategy model can be obtained through training of a strategy training library. The policy training library may include a historical marketing policy library and a historical marketing equity feature library. The data in the policy training library may be training data and corresponding real labels, i.e. real business information. The training data may be a business need sample, or may be one or more of the business image sample, business transaction sample characteristics, consumption preference sample characteristics, business tier sample, and the business need sample.

The business demand prediction model in the specification not only can predict target demands, but also can match and recommend operation information for the target demands, and is efficient and scientific. And, the operator of the payment platform can directly use the recommended operation information to operate or refer to the recommended operation information to formulate the final floor execution operation information, so that the pressure of the operator is reduced.

After the target merchant 120 performs the recommended business information or the fulfillment business information on the floor, the computing device 600 may obtain the most up-to-date order data for the target merchant to verify, via GMV, the accuracy of the merchant demand prediction model for the target demand prediction. The computing device 600 may determine that the target merchant 120 performs the recommended operation information on the floor or a GMV for a period of time after the performance of the operation information, such as obtaining a GMV for one month, determining whether the GMV after the performance of the floor is higher than the GMV before the performance of the floor, and when so, determining that the accuracy of the merchant demand prediction model on the target demand prediction is high; when the target demand prediction model is lower than the target demand prediction model, the accuracy of the target demand prediction model can be determined to be low, and the target demand prediction model can be retrained.

Because the store operation data set of the target merchant 120 may change, especially the order data is continuously increased, in order to ensure accuracy of the prediction of the merchant demand prediction model, the computing platform 600 may continuously obtain the latest data of the target merchant 120 and update the store operation data set, and iteratively train the merchant demand prediction model using the updated store operation data set.

It should be noted that, the accuracy of the merchant demand prediction model may be verified, so that it is determined that the merchant demand prediction model is capable of accurately understanding the merchant demand by integrating multi-mode data. When validated, computing device 600 may implement the predictive methods of the present description through the merchant demand prediction model.

In addition, computing device 600 may store a store operating dataset for each merchant affiliated with the paymate, and may also store predicted needs for each merchant, so that paymate staff may query the needs of individual merchants on terminal 300 or mobile workstation 200. For example, a sales person enters the ID of the merchant to be queried on mobile station 200, which may give the merchant a demand.

When computing device 600 utilizes information of business industries of each merchant in predicting merchant demand, a plurality of merchant demands corresponding to a plurality of merchants in the same business industry may be obtained and a demand corresponding to the business industry may be determined based on the plurality of merchant demands. In some embodiments, the computing device 600 may present a first aggregate demand based on the plurality of merchant demands, e.g., with the most numerous of the plurality of demands being the first aggregate demand, and in turn, the aggregate demand being the demand of the business industry. In some embodiments, computing device 600 may treat all of the plurality of merchant requirements as corresponding requirements for the business industry. In this way, a payment platform operator may query the terminal 300 or mobile station 200 for a business industry requirement, such as a catering industry requirement. The business industry may be an industry within a geographic area, such as a restaurant industry within a certain area of a certain city.

When computing device 600 utilizes information for each merchant's merchant tier in predicting merchant demand, multiple merchant demands corresponding to multiple merchants at the same tier may be obtained and the demand corresponding to that tier determined based on the multiple merchant demands. In some embodiments, the computing device 600 may present a second aggregate demand based on the plurality of merchant demands, e.g., with the most numerous demands of the plurality of demands as the second aggregate demand, and in turn, the aggregate demand as the hierarchical demand. In some embodiments, computing device 600 may treat all of the plurality of merchant requirements as the level corresponding requirements. In this way, a paymate's staff may query the terminal 300 or mobile station 200 for the needs of one merchant level, such as the needs of a highly active (highly active) level of merchants.

The method of the specification can realize the requirement insight of a single merchant from the microcosmic aspect, can realize the macroscopic requirement insight from the aspects of merchant layering and business industry, meets the requirements of different staff, and becomes the fulcrum of scientific operation.

In summary, according to the method P100 and the system 001 for predicting the demand of the merchant provided in the present disclosure, the computing device 600 of the payment platform may automatically predict the target demand of the target merchant 120 based on the store management data set of the target merchant 120 through the pre-trained model for predicting the demand of the merchant, where the target demand includes the management demand for the payment platform. Meanwhile, the merchant demand prediction model can also match corresponding recommended operation information for target demands, and assist operators in formulating more efficient and scientific operation strategies. Meanwhile, the computing device 600 can predict the target requirement by identifying the multi-modal data such as video, image, audio, text and the like of the target merchant, so that the problems of low quality and poor authenticity of the artificially collected data are overcome. Meanwhile, the computing device 600 may update the merchant demand prediction model in real time based on the real-time order data of the target merchant to adjust the predicted target demand in real time, thereby ensuring the accuracy of demand prediction.

In another aspect, the present description provides a non-transitory storage medium storing at least one set of executable instructions for making predictions of merchant demand. When executed by a processor, the executable instructions direct the processor to perform the steps of the business need prediction method P100 described herein. In some possible implementations, aspects of the specification can also be implemented in the form of a program product including program code. The program code is for causing the prediction system 001 of merchant demand to perform the steps of the method P100 of predicting merchant demand described in this specification when the program product is run on the prediction system 001 of merchant demand. The program product for implementing the above method may employ a portable compact disc read only memory (CD-ROM) comprising program code and may run on the forecasting system 001 for merchant demand. However, the program product of the present specification is not limited thereto, and in the present specification, the readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system. The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. The computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable storage medium may also be any readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing. Program code for carrying out operations of the present specification may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the merchant demand prediction system 001, partially on the merchant demand prediction system 001, as a stand-alone software package, partially on the merchant demand prediction system 001, partially on a remote computing device, or entirely on the remote computing device.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

In view of the foregoing, it will be evident to a person skilled in the art that the foregoing detailed disclosure may be presented by way of example only and may not be limiting. Although not explicitly described herein, those skilled in the art will appreciate that the present description is intended to encompass various adaptations, improvements, and modifications of the embodiments. Such alterations, improvements, and modifications are intended to be proposed by this specification, and are intended to be within the spirit and scope of the exemplary embodiments of this specification.

Furthermore, certain terms in the present description have been used to describe embodiments of the present description. For example, "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present description. Thus, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined as suitable in one or more embodiments of the invention.

It should be appreciated that in the foregoing description of embodiments of the present specification, various features have been combined in a single embodiment, the accompanying drawings, or description thereof for the purpose of simplifying the specification in order to assist in understanding one feature. However, this is not to say that a combination of these features is necessary, and it is entirely possible for a person skilled in the art to label some of the devices as separate embodiments to understand them upon reading this description. That is, embodiments in this specification may also be understood as an integration of multiple secondary embodiments. While each secondary embodiment is satisfied by less than all of the features of a single foregoing disclosed embodiment.

Each patent, patent application, publication of patent application, and other material, such as articles, books, specifications, publications, documents, articles, and the like, in addition to any historical prosecution documents associated therewith, any identical or conflicting material to the present document or any identical historical prosecution document which may have a limiting effect on the broadest scope of the claims, is incorporated herein by reference for all purposes now or later associated with the present document. Furthermore, the terms in this document are used in the event of any inconsistency or conflict between the description, definition, and/or use of terms associated with any of the incorporated materials.

Finally, it is to be understood that the embodiments of the application disclosed herein are illustrative of the principles of the embodiments of the present specification. Other modified embodiments are also within the scope of this specification. Accordingly, the embodiments disclosed herein are by way of example only and not limitation. Those skilled in the art can adopt alternative arrangements to implement the application in the specification based on the embodiments in the specification. Therefore, the embodiments of the present specification are not limited to the embodiments precisely described in the application.

Claims

1. A method for predicting merchant demand, applied to a computing device of a paymate, comprising:

determining a store operating dataset of a target merchant, the store operating dataset comprising data characterizing a store operating state; and

outputting target demands predicted for the target merchants through a pre-trained merchant demand prediction model based on the store operation data set, wherein the target demands comprise operation demands for the payment platform.

2. The method of claim 1, wherein the outputting, based on the store business dataset, the predicted target demand for the target merchant by a pre-trained merchant demand prediction model comprises:

Determining a feature set of the target merchant based on the store operating dataset; and

inputting the feature set into a merchant demand analysis model, and outputting the target demand, wherein the merchant demand prediction model comprises the merchant demand analysis model.

3. The method of claim 2, wherein the feature set includes at least one of a target representation, a target business feature, a target level feature, a demand feedback feature of the target merchant.

4. The method of claim 3, wherein the store business data set includes store attribute data and commodity business data, the target representation is obtained by inputting the store attribute data and the commodity business data into a merchant representation model, and the merchant demand prediction model includes the merchant representation model.

5. The method of claim 4, wherein the store attribute data includes at least one of a merchant identification, a store location, and a store area of the target merchant.

6. The method of claim 4, wherein the commodity business data includes at least one of a business industry, a commodity name, and a commodity classification of the target merchant.

7. The method of claim 4, wherein the store operating dataset further comprises demand feedback data that is feedback of the target merchant's own demand for the payment platform.

8. The method of claim 3, wherein the store business data set includes order data for the target merchant to generate a plurality of transaction orders over a preset period of time, the target business characteristics being derived by inputting the order data into a merchant transaction model and a consumption preference model, respectively, the merchant demand prediction model including the merchant transaction model and the consumption preference model.

9. The method of claim 8, wherein the target business characteristics include merchant transaction characteristics and/or consumption preference characteristics,

the merchant transaction characteristic includes a total amount of merchandise transactions within the preset time period and/or a total amount of orders within the preset time period,

the consumption preference feature comprises at least one of the following within the preset period of time: peak period of consumption, average amount of consumption per order, and free-selling goods.

10. The method of claim 3, wherein the target hierarchical features are derived by inputting the target representation and the target business features into a merchant hierarchical model, the target hierarchical features characterizing the activity of the target merchant in using the paymate, the merchant demand prediction model comprising the merchant hierarchical model.

11. The method of claim 3, wherein the outputting, based on the store business dataset, the predicted target demand for the target merchant by a pre-trained merchant demand prediction model further comprises:

and outputting recommended operation information matched with the target demand based on a marketing strategy model, wherein the merchant demand prediction model comprises the marketing strategy model.

12. The method of claim 11, wherein the recommended operations information includes at least one of equity operations information and equipment deployment operations information, wherein the equity operations information includes equity values, equity types, and equity distribution patterns that can be obtained by the paymate completing an order, and the equipment deployment operations information includes equipment recommended to deploy the paymate.

13. The method of claim 11, wherein the recommended operations information is obtained by inputting the target demand into the marketing strategy model.

14. The method of claim 11, wherein the recommended business information is obtained by inputting the target demand and at least one of the target representation, the target business feature, and the target hierarchical feature into the marketing strategy model.

15. The method of claim 1, wherein the determining the store operating dataset for the target merchant comprises:

and identifying multi-mode data to obtain the store management data set, wherein the multi-mode data at least comprises two of video, image, audio, text and geographic position.

16. A system for predicting merchant demand, comprising:

at least one storage medium storing at least one set of instructions for effecting predictions of merchant demand; and

at least one processor communicatively coupled to the at least one storage medium,

wherein the at least one processor reads the at least one instruction set and implements the method of predicting merchant demand of any one of claims 1-15 when the system of predicting merchant demand is running.