CN113298610A - Information recommendation and acquisition method, equipment and storage medium - Google Patents

Information recommendation and acquisition method, equipment and storage medium Download PDF

Info

Publication number
CN113298610A
CN113298610A CN202110169012.7A CN202110169012A CN113298610A CN 113298610 A CN113298610 A CN 113298610A CN 202110169012 A CN202110169012 A CN 202110169012A CN 113298610 A CN113298610 A CN 113298610A
Authority
CN
China
Prior art keywords
inventory
information
resource
user
inventory resource
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110169012.7A
Other languages
Chinese (zh)
Inventor
马光锐
郑欢
邓玉明
戚赟炜
荣鹰
陈督
张勋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN202110169012.7A priority Critical patent/CN113298610A/en
Publication of CN113298610A publication Critical patent/CN113298610A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0283—Price estimation or determination
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0641—Shopping interfaces
    • G06Q30/0643—Graphical representation of items or shoppers

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Algebra (AREA)
  • Game Theory and Decision Science (AREA)
  • Human Resources & Organizations (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The embodiment of the application provides an information recommendation and acquisition method, equipment and a storage medium. In the embodiment of the application, the front-end recommendation and the inventory information of the supply chain end are combined aiming at the inventory resources which can be traded on line, and the multi-dimensional information such as the shortage information, the current inventory information and the preference information of a target user to the inventory resources is fused at the same time, so that the information recommendation method based on the inventory balance is realized.

Description

Information recommendation and acquisition method, equipment and storage medium
Technical Field
The present application relates to the field of internet technologies, and in particular, to a method, an apparatus, and a storage medium for recommending and acquiring information.
Background
The rapid development of the retail industry promotes a new retail mode with integration of all channels and multiple scenes. Under a new retail mode, the machine learning-based method is increasingly applied to the estimation links of commodity Click Through Rate (CTR), Click conversion Rate (CVR) or Gross transaction Volume (GMV), and the estimation accuracy Rate is continuously improved; after the estimation is finished, personalized commodity recommendation can be performed for the user according to the multi-objective requirements, and indexes such as CTR, CVR or GMV are tried to be maximized.
Currently, it is a common practice to estimate the click probability of a commodity, such as CTR or CVR, according to historical data, sort the commodities according to the click probability, and select the commodity with the highest click probability to recommend to a user. However, since the user has randomness in accessing the APP or the webpage, the user may not be able to accurately recommend the product to the user based on only the click probability, for example, an invalid product may be recommended to the user, so that the user may not purchase the recommended product, and the user experience may be reduced.
Disclosure of Invention
Various aspects of the present application provide an information recommendation and acquisition method, device, and storage medium, so as to recommend inventory resources to a user more accurately, reduce a recommendation probability of invalid resources, and improve user experience.
The embodiment of the application provides an information recommendation method, which comprises the following steps: receiving a page request sent by terminal equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating a page request operation; acquiring the shortage information and the current inventory information of at least one inventory resource which can be traded on line, and predicting the preference information of the target user on the at least one inventory resource; selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource; and sending the information of the target inventory resource to the terminal equipment so that the terminal equipment can display the information of the target inventory resource on a page requested by the target user.
An embodiment of the present application further provides an information obtaining method, including: responding to page request operation, and sending a page request to server equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating the page request operation; receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by the target user; the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
An embodiment of the present application further provides a server device, including: a memory and a processor; the memory for storing computer programs or instructions; the processor, coupled with the memory, to execute the computer program or instructions to: receiving a page request sent by terminal equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating a page request operation; acquiring the shortage information and the current inventory information of at least one inventory resource which can be traded on line, and predicting the preference information of the target user on the at least one inventory resource; selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource; and sending the information of the target inventory resource to the terminal equipment so that the terminal equipment can display the information of the target inventory resource on a page requested by the target user.
An embodiment of the present application further provides a terminal device, including: a memory, a processor, and a display; the memory for storing computer programs or instructions; the processor, coupled with the memory, to execute the computer program or instructions to: responding to page request operation, and sending a page request to server equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating the page request operation; receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by the target user; the display is used for displaying the page requested by the target user; the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
Embodiments of the present application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, can cause the processor to implement the steps in the information recommendation or acquisition method provided by the embodiments of the present application.
Embodiments of the present application further provide a computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the processor is caused to implement the steps in the information recommendation or acquisition method provided by the embodiments of the present application.
In the embodiment of the application, the front-end recommendation and the inventory information of the supply chain end are combined aiming at the inventory resources which can be traded on line, and the multi-dimensional information such as the shortage information, the current inventory information and the preference information of a target user to the inventory resources is fused at the same time, so that the information recommendation method based on the inventory balance is realized.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a block diagram of a transaction data processing system according to an exemplary embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating estimation of shadow prices of commodities according to an exemplary embodiment of the present application;
fig. 3 is a flowchart illustrating an information recommendation method according to an exemplary embodiment of the present application;
fig. 4 is a schematic flowchart of an information obtaining method according to an exemplary embodiment of the present application;
fig. 5 is a schematic structural diagram of a server device according to an exemplary embodiment of the present application;
fig. 6 is a schematic structural diagram of a terminal device according to an exemplary embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the existing new retail mode, when commodity recommendation is carried out on a user, the problem that the commodity cannot be accurately recommended to the user exists. In view of the problems in the prior art, the embodiment of the present application provides an information recommendation method, which can perform personalized recommendation not only for commodities which can be traded online and have inventory information in the e-commerce field, but also for other resource objects which can be traded online and have inventory information. In the embodiment of the application, commodities which can be traded on line and have inventory information in the e-commerce field and other resource objects which can be traded on line and have inventory information are collectively called inventory resources, and aiming at the inventory resources which can be traded on line, front-end recommendation and inventory information at the supply chain end are combined, and meanwhile, multi-dimensional information such as scarcity information of the inventory resources, current inventory information and preference information of target users to the inventory resources is fused, so that the information recommendation method based on inventory balance is realized.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic structural diagram of a transaction data processing system according to an exemplary embodiment of the present application. As shown in fig. 1, the transaction data processing system 100 includes: a terminal device 101, a server device 102, and an inventory management device 103; the server apparatus 102 is communicatively connected to the terminal apparatus 101 and the inventory management apparatus 103.
The communication connection between the server device 102 and the terminal device 101 and the inventory management device 103 may be a wireless connection or a wired connection. If the server device 102 is communicatively connected to the terminal device 101 or the inventory management device 103 through a mobile network, the network format of the mobile network may be any one of 2G (gsm), 2.5G (gprs), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G + (LTE +), 5G, WiMax, and the like.
In this embodiment, the transactional data processing system 100 may provide some tradable resource objects to the user, including but not limited to: the resource objects can be various commodities in the e-commerce field, and can also be other goods, raw materials or labor resources which can be traded online, electronic film and television resources, and the like. These resource objects, which all have a quantity attribute, may be stored and managed by the resource repository 104 and may be referred to as inventory resources. If the resource object is a physical object such as various tangible goods, goods or raw materials, the resource warehouse can be a physical warehouse or a containing space which is provided with certain positions; if the resource object is a virtual object such as an electronic movie resource, the resource repository may be a storage space with a certain information storage or recording function, for example, a disk, a hard disk, a database, or the like. The inventory management device 103 corresponds to the resource warehouse 104, and is mainly responsible for maintaining and managing inventory information of inventory resources stored in the resource warehouse 104, for example, current inventory information can be dynamically updated according to a transaction process of the inventory resources, and is responsible for managing matters and information related to replenishment, such as determining replenishment time and replenishment quantity, and sending a replenishment notice. The stock information of the stock resources mainly refers to the stock quantity of the stock resources.
In this embodiment, under the cooperation of the terminal device 101 and the server device 102, the user may perform a transaction operation on the inventory resource provided by the transaction data processing system 100, where the transaction operation includes an online transaction operation or an online and offline combined transaction operation. The terminal device used by the user can be, for example, a smart phone, a tablet computer, a personal computer, a wearable device, and the like. The terminal device 101 is installed with application software for a user to perform online and/or offline transactions, where the application software may be an application program (APP), a client, an applet, or an SDK, and the application software run by the terminal device 101 may provide online and/or offline transaction functions for the user. For example, taking commodities in the e-commerce field as an example, a user may initiate an online transaction operation through the APP on the terminal device 101, for example, the user may select commodities, add a shopping cart, place orders online, pay online, and the like, and may issue an evaluation online. Or, the user may also select and purchase the offline commodity in the online offline store, scan the two-dimensional code or barcode information of the selected and purchased offline commodity through the APP scanning function on the terminal device 101, obtain attribute information such as the price of the offline commodity, form an electronic order for online payment, and implement online and offline combined transaction operation. Or, the user can select the commodities sold by the online and offline store on line, order the commodities on line, complete payment, and then the store personnel are responsible for picking the commodities from the online and offline store and distributing the commodities to the user, so that online and offline combined transaction operation is realized.
The server device 102 may be a server that performs resource transaction processing in a network virtual environment, and generally refers to a server that performs online resource transaction using a network, and a user generally needs to register identity information in the server to perform transaction behaviors such as purchase and exchange of inventory resources using a registered account number, for example, a transaction server of each e-commerce platform or an online transaction website, or a third-party server. In physical implementation, the server device 102 may be any device capable of providing computing services, responding to service requests, and performing processing, and may be, for example, a conventional server, a cloud host, a virtual center, a server array, or the like. The server mainly comprises a processor, a hard disk, a memory, a system bus and the like, and is similar to a general computer framework.
In the embodiment, the process of the user performing the online transaction or the online and offline combined transaction through the terminal device 101 and the server device 102 is the same or similar for any type of inventory resources. For ease of understanding and description, in the embodiment of the present system, the operation principle of the system 100 of the present embodiment is described by taking the example that the inventory resource is a commodity in the e-commerce field. In the e-commerce field, based on the system 100 of this embodiment, a user can browse information of goods provided by a merchant on line through the terminal device 101, and determine which goods to buy on line. Alternatively, in the case of online shopping, the user may browse the information of the product provided by the merchant online through the terminal device 101, and then online shop for the product in the shopping mall. No matter which shopping mode is adopted, in order to enable a user to more conveniently and efficiently obtain needed commodity information from a plurality of commodity information through the terminal equipment 101, the server-side equipment 102 further has an information recommendation function, commodity information which meets the requirements of the user better can be recommended to the user through the terminal equipment 101, the quantity of the commodity information which the user needs to browse is reduced, the user does not need to browse and select one by one in a plurality of commodity information banks, the commodity information can be directly selected from a small quantity of recommended commodity information, accordingly, less time is spent, the needed commodity information can be selected more efficiently, and the user experience is improved.
Specifically, as shown in fig. 1 ((r)), a user initiates a page request operation through the terminal device 101, the terminal device 101 sends the page request to the server device 102 in response to the page request operation, and the page request carries a user identifier, where the user identifier is used to identify the user initiating the page request operation, so that the server device 102 can perform personalized information recommendation for the user accordingly. Further, as shown in fig. 1, the server device 102 performs personalized commodity recommendation for the user, and returns information of a recommended target commodity to the terminal device 101, and a description is subsequently developed about a personalized recommendation process. Further, as shown in the sixth step in fig. 1, after receiving the information of the target product returned by the server device 102, the terminal device 101 displays the requested page to the user, and at the same time displays the information of the target product on the page requested by the user, the probability that the target product contains the favorite or required product of the user is higher, so that the user can conveniently and quickly select and purchase the required product from the target product, the time for browsing the product is saved, and the shopping efficiency and the shopping experience are improved.
Further optionally, when the terminal device 101 displays the information of the target commodity on the page requested by the user, the information of the target commodity may be displayed on the page requested by the user in a random display manner; or, the information of the target commodity can be displayed on the page requested by the user in sequence from high to low according to the preference degree of the user on the target commodity; or, the information of the target commodity can be displayed on the page requested by the user in sequence according to the sequence of the shortage of the target commodity from low to high; or, the information of the target commodity can be displayed on the page requested by the user in sequence according to the sequence of the current stock information of the target commodity from high to low; or, the priority of the target commodity may be calculated by simultaneously combining at least two of the current stock information of the target commodity, the shortage information, and the preference information of the user for the target commodity, and the information of the target commodity may be displayed on the page requested by the user in sequence from high to low according to the priority of the target commodity.
In this embodiment, the terminal device 101 is installed with shopping application software, and by running the application software, the terminal device 101 can provide an application page to a user, where the application page may be an APP page, an applet page, or a browser page according to a type of the application software. In an optional embodiment, no matter what type of application software is, a recommendation function may be embedded in each page provided by the application software, that is, no matter which page is requested by the user, the terminal device 101 and the server device 102 cooperate with each other to perform personalized commodity recommendation for the user, and the terminal device 101 displays information of a recommended target commodity on the page requested by the user. In addition, a recommendation function may be embedded in a part of the pages, so that only when the user requests a specific page in which the recommendation function is embedded, the terminal device 101 and the server device 102 cooperate with each other to perform personalized commodity recommendation for the user, and the terminal device 101 displays information of a recommended target commodity on the specific page requested by the user. In this embodiment, the specific page embedded with the recommendation function is not limited, and may be, for example, a home page, a shopping cart page, a group page, or a user detail page provided by application software. The following examples illustrate:
for example, a recommendation function may be embedded in the application home page, and when the user requests the application home page, the terminal device 101 may send a page request to the server device 102 in response to an operation initiated by the user to request the application home page, so as to request the server device 102 to perform personalized information recommendation for the user and return information of a recommended target commodity; the terminal device 101 renders the application home page, and displays the information of the target commodity returned by the server device 102 on the application home page while displaying the application home page to the user, so as to guide the user to select and purchase the required commodity from the target commodity. An example of the application home page on which the target product information is displayed is shown in fig. 1. It should be noted that, a user may request to start the application software by clicking an icon of the application software on the terminal device 101, and if the user defaults to enter the application home page when the application software is started through the icon of the application software, the user clicking the icon of the application software is equivalent to initiating an operation of requesting the application home page. Or, in a case that the application software provides a home page jump function, the user may click a navigation tag pointing to the home page on the current application page to initiate an operation of jumping to the application home page, which is also equivalent to initiating an operation of requesting the application home page.
For another example, a recommendation function may be embedded in the shopping cart page, and when the user requests the shopping cart page, the terminal device 101 may send a page request to the server device 102 in response to an operation initiated by the user to request the shopping cart page, so as to request the server device 102 to perform personalized information recommendation for the user and return information of a recommended target commodity; the terminal device 101 renders the page of the shopping cart, and displays the page of the shopping cart to the user, and at the same time, displays the information of the target product returned by the server device 102 on the page of the shopping cart to guide the user to select and purchase the required product from the target product. An example of a shopping cart page on which target commodity information is displayed is shown in fig. 1. Optionally, in the case that the application software provides a shopping cart page jumping function, the user may also click a navigation tag pointing to the shopping cart page on the current application page, and initiate an operation of jumping to the shopping cart page, which is equivalent to initiating an operation of requesting the shopping cart page. Alternatively, a shopping cart icon or a floating window may be displayed on each page, and a user clicking on the shopping cart icon or the floating window may initiate an operation of requesting a shopping cart page.
For another example, a recommendation function may be embedded in a group page of the shopping software, and when the user requests the group page, the terminal device 101 may send a page request to the server device 102 in response to an operation initiated by the user to request the group page, so as to request the server device 102 to perform personalized information recommendation for the user and return information of a recommended target commodity; the terminal device 101 renders a group page, and displays information of the target product returned by the server device 102 on the group page while displaying the group page to the user, so as to guide the user to select and purchase the required product from the target product. Optionally, in a case that the application software provides a group page jump function, the user may also click a navigation tag pointing to the group page on the current application page, and initiate an operation of jumping to the group page, where the operation is equivalent to initiating an operation of requesting the group page.
For another example, a recommendation function may be embedded in the personal details page of the user, and when the user requests the personal details page, the terminal device 101 may send a page request to the server device 102 in response to an operation initiated by the user to request the personal details page, so as to request the server device 102 to perform personalized information recommendation for the user and return information of a recommended target product; the terminal device 101 renders the personal detail page, and displays the personal detail page to the user, and at the same time, displays the information of the target commodities returned by the server device 102 on the personal detail page so as to guide the user to select and purchase the required commodities from the target commodities. Optionally, in a case where the application software provides the personal detail page jump function, the user may click a navigation tag pointing to the personal detail page, for example, a "my" tag commonly found in various applications, on the current application page, and initiate an operation of jumping to the personal detail page, which is equivalent to initiating an operation of requesting the personal detail page.
In any of the above manners, when the user requests the page embedded with the recommendation function, the terminal device 101 may respond to the page request operation, send a page request to the server device 102, and carry the user identifier in the page request; for the server device 102, it may receive a page request sent by the terminal device 101, obtain a user identifier from the page request, and determine a user initiating the page request according to the user identifier. For convenience of differentiation and description, in the embodiments described below in the present application, a user who initiates a page request operation is referred to as a target user. After determining the target user, the server device 102 may perform personalized commodity recommendation for the target user.
In this embodiment, the server device 102 determines at least one commodity that can participate in the online transaction, where the commodities may be all commodities provided by the merchant or part of commodities provided by the merchant, and this is not limited. The commodities come from a resource warehouse of a merchant, and the resource warehouse may be one or more, and may be a store warehouse or an area-level warehouse, which is not limited herein. Further, in order to perform personalized commodity recommendation on the target user, personalized requirements of the target user can be considered, and commodity recommendation is performed on the target user in a targeted manner. In addition, in the embodiment, in consideration of application requirements of online and offline transaction fusion, in the process of recommending commodities to the target user, in addition to consideration of personalized requirements of the target user, inventory information of marketing terminals is combined, marketing strategies at the front end of a merchant are combined with inventory decisions at the back end, and front-end and back-end joint optimization recommendation can be performed in the view of global yield management.
Specifically, as shown in fig. 1, after receiving the page request, the server device 102 obtains the shortage information and the current inventory information of at least one commodity on the one hand; and on the other hand, predicting the preference information of the target user for at least one commodity. Wherein the current inventory information may be obtained in real time from the inventory management device 103. The information about the shortage of at least one commodity may be acquired in real time, or may be acquired offline in advance and stored, for example, the information about the shortage of at least one commodity may be acquired periodically and stored, so that the latest stored information about the shortage of at least one commodity may be acquired directly each time the commodity is used. Accordingly, the preference information of the target user for the at least one commodity can be predicted in real time, or can be predicted offline in advance and stored, so that the newly stored preference information of the target user for the at least one commodity can be directly obtained in each use. In fig. 1, the example of predicting the preference information of the target user for at least one product in real time is shown, but the present invention is not limited thereto. Then, as shown in the fourth step in fig. 1, the server device 102 selects a target product from at least one product according to the shortage information of the at least one product, the current inventory information, and the preference information of the target user for the at least one product; further, as indicated by a fifth in fig. 1, the server device 102 returns the information of the selected target product to the terminal device 101, so that the terminal device 101 displays the information of the target product on the page requested by the target user, as indicated by a sixth in fig. 1.
When recommending commodities to users, if the preferences of the users on the commodities are simply based on the preferences of the users on the commodities, most users prefer the same commodity, and the commodities are recommended to most users, so that under the condition that the inventory of the commodities is insufficient, the condition of invalid commodity recommendation occurs, namely, a part of users are recommended with the commodities but cannot purchase the recommended commodities due to the insufficient inventory, the shopping experience of the users is seriously influenced, the trust of the users on the recommendation system is also reduced, and the advantages of the recommendation system cannot be fully exerted. In the embodiment of the application, the preference of the target user, the shortage degree of the commodities and the inventory information are combined at the same time, so that other commodities can be recommended to the user in a compromise mode under the conditions of shortage of the commodities and insufficient inventory no longer according to the preference of the user to the commodities, the commodities can be recommended to the user more reasonably, the condition of invalid commodity recommendation can be reduced, and the shopping experience of the user can be guaranteed while the shopping efficiency of the user is improved.
In each of the above or following embodiments of the present application, when predicting preference information of a target user for at least one inventory resource in real time or in advance, the server device 102 may obtain portrait data of the target user according to a user identifier; the portrait data of the target user includes at least: basic attribute information of the target user and historical transaction behavior data of the target user; the basic attribute information of the target user includes, but is not limited to: the target user's consumption abilities, educational background, etc.; the historical transaction behavior data of the target user includes, but is not limited to: the type of the historical transaction behavior such as purchase, shopping cart, payment, comment, etc., the attributes of the goods involved in the historical transaction behavior, the time of the historical transaction behavior, the frequency of occurrence, etc. Further, the server device 102 predicts preference information of the target user for at least one commodity in real time or in advance according to the portrait data of the target user. Furthermore, the server device 102 may predict, in real time or in advance, the preference information of the target user for the at least one commodity according to the portrait data of the target user and the basic attribute information of the at least one commodity. The basic attribute information of each commodity includes, but is not limited to: categories, brands, specifications, prices, brand identity, sales of the goods, etc. Preferably, considering that historical transaction behavior data in the target user image data dynamically changes with the time, the latest image data of the target user can be acquired each time commodity recommendation needs to be carried out on the target user, and preference information of the target user for at least one commodity can be predicted in real time according to the latest image data or the latest image data and basic attribute information of at least one commodity at the same time, so that the accuracy of a prediction result can be improved.
In an alternative embodiment, a preference prediction model may be trained in advance, and the server device 102 may invoke the preference prediction model to predict, in real time or in advance, the preference information of the target user for the at least one commodity. Specifically, the server device 102 may input portrait data of the target user and basic attribute information of at least one commodity into the preference prediction model, and output preference information of the target user for the at least one commodity by the preference prediction model.
Further alternatively, the CVR and CTR of the target user for the at least one item may be used to represent the target user's preference information for the at least one item, but is not limited thereto. For example, the CVR or CTR of the target user for the at least one commodity may also represent the target user's preference information for the at least one commodity. Of course, other parameters than CVR and CTR, such as the preference degree of the target user for the merchandise or the matching degree between the target user and the merchandise, may be used to represent the preference information of the target user for each merchandise. In the case that the CVR and CTR of the target user for the at least one commodity are used to represent the preference information of the target user for the at least one commodity, the server device 102 may predict the CVR and CTR of the target user for the at least one commodity according to the portrait data of the target user. Further alternatively, in the case of using a preference prediction model, the preference prediction model may be a model capable of predicting CVR and CTR at the same time, or may include both the CVR prediction model and the CTR prediction model.
After obtaining the CVR and CTR of the target user for at least one commodity, the server device 102 may select the target commodity from the at least one commodity according to the scarcity information of the at least one inventory resource, the current inventory information, and the CVR and CTR of the target user for the at least one commodity, and return the information of the target commodity to the terminal device 101, so that the terminal device 101 displays the information of the target commodity on the page requested by the target user.
In the above or the following embodiments of the present application, one way of selecting a target inventory resource from at least one product according to the shortage information of the at least one product, the current inventory information, and the preference information of the target user for the at least one product includes:
first, the expected profit of at least one commodity is determined according to the scarcity information of at least one commodity and the preference information of the target user for the at least one commodity.
Optionally, in a case where the CVT and CTR of the target user for the goods are used to represent the preference information of the target user for the goods, the above-mentioned manner of determining the expected profit of at least one kind of goods according to the scarcity information of at least one kind of goods and the preference information of the target user for at least one kind of goods includes: correcting the price attribute of at least one commodity according to the shortage degree information of the at least one commodity to obtain the corrected price of the at least one commodity; the expected revenue for the at least one item is determined based on the target user's CVR and CTR for the at least one item and the revised price for the at least one item.
Next, the expected revenue for the at least one item is corrected based on the current inventory information for the at least one item.
Optionally, the correcting the expected profit for the at least one product according to the current inventory information of the at least one product includes: generating an inventory penalty factor of at least one commodity according to the current inventory information and the initial inventory information of the at least one commodity; the expected revenue of the at least one commodity is corrected using the inventory penalty factor for the at least one commodity. For example, for each commodity, the inventory penalty factor of the commodity can be calculated according to the ratio of the current inventory information of the commodity to the initial inventory information; further, the product stock penalty factor is multiplied by the expected yield of the product to obtain a corrected expected yield of the product.
And finally, selecting the target commodity from the at least one commodity according to the corrected expected income of the at least one commodity.
Optionally, the at least one commodity may be sorted according to the order of the corrected expected profit of the at least one commodity from large to small, and then, several commodities with the top sorting may be selected as the target commodity. Alternatively, a product whose corrected expected profit is within the set profit range may be selected as the target product. The embodiment of the present application does not limit the specific implementation manner of selecting the target product according to the corrected expected income of at least one product.
After the target commodity is obtained, the server-side device 102 may return the information of the target commodity to the terminal device 101, so that the terminal device 101 may display the information of the target commodity on a page requested by the target user, and thus, the user may select and purchase a desired commodity from the target commodity, which is time-saving and efficient, and the merchant may also obtain the maximum benefit. Alternatively, when displaying the information of the target product, the terminal device 101 may display the information of the target product on the page requested by the user in order from high to low of the corrected expected profit according to the corrected expected profit of the target product. The information of the target product includes, but is not limited to: the picture, name, price, weight or quantity of the target commodity, and relevant preferential information or preferential strategies.
In the above or below embodiments of the present application, the shadow price of the commodity may be used to reflect the shortage information of the commodity. The shadow price of the commodity can reflect the real situation of the shortage of the commodity, and the shadow price of the shortage commodity is higher, namely, the higher the shadow price of a certain commodity is, the higher the shortage degree of the commodity is. Further, in this embodiment of the application, the server device 102 may use historical preference information of the historical user on the at least one commodity as a data base, and use a linear programming method to estimate the shadow price of the at least one commodity. Further optionally, under the condition that the number of the historical users is large, the data base of the historical preference information of the historical users on the at least one commodity is large, and therefore, the historical preference information of the historical users on the at least one commodity can be sampled to obtain the historical preference information of the sampled historical users on the at least one commodity; based on the historical preference information of the sampling historical user on at least one commodity, the preference of the user arriving in the future on the at least one commodity is estimated, so that the shadow price of the at least one commodity is obtained, the calculation amount can be reduced, and the calculation efficiency is improved.
Furthermore, considering that the shortage of at least one commodity is influenced by various factors and is possibly dynamically changed, the shadow price of the at least one commodity can be estimated in real time based on the historical preference information of the sampling historical user on the at least one commodity every time a page request is received, or the shadow price of the at least one commodity can be periodically estimated based on the historical preference information of the sampling historical user on the at least one commodity, so that the accuracy of commodity recommendation based on the shadow price of the commodity can be improved, and the defect that the commodity recommendation effect cannot be accurately depicted only by calculating the price once can be avoided. Of course, in consideration of the problems of computational efficiency and computational resources, it is preferable that the shadow price of at least one commodity is estimated periodically. In view of the above, a detailed embodiment of the server device 102 periodically predicting the shadow price of at least one commodity by using a linear programming method, as shown in fig. 2, includes the following operations:
sampling operation: and sampling historical preference information of the historical user on at least one commodity to obtain the historical preference information of the sampled historical user on the at least one commodity.
In the sampling operation, historical preference information of a part of historical users on at least one commodity can be periodically extracted from historical preference information of the historical users on the at least one commodity, and since the extracted historical preference information of the part of historical users on the at least one commodity is used for guiding a subsequent recommendation process, the historical preference information is representative and can reflect the preference information of users on the at least one commodity in the future as far as possible. A future arrival user refers to a user who initiates a page request through the terminal apparatus 101 in the future.
In an alternative embodiment, historical CVR and historical CTR of the historical user for the at least one item may be used to represent historical preference information of the historical user for the at least one item. Based on this, the CTR and CVR of part of the historical users can be extracted from the historical data periodically and expressed as
Figure BDA0002938467080000141
And
Figure BDA0002938467080000142
the index h indicates history data, the index i indicates the ith commodity, and the index u indicates the sampling history user, then
Figure BDA0002938467080000143
The CTR of the sampling history user u for the ith commodity is shown,
Figure BDA0002938467080000144
indicating the sample history the CVR of user u for the ith item.
Further optionally, for the current time window, the number of users that may arrive in the current time window may be predicted according to the number of historical users that appear in the historical synchronization time window, and is denoted as kcur(ii) a According to the number k of users possibly arrived in the current time windowcurAnd sampling from historical users appearing in the historical contemporaneous time window to obtain sampled historical users. For example, the current timeThe window is 7:00-8:00 a.m., historical users who appeared in the time of 7:00-8:00 a.m. each morning in the last week may be sampled, or historical users who appeared in the time of 7:00-8:00 a.m. each morning in the last 10 days may be sampled. Optionally, the number of sampling history users is kcurBut is not limited thereto. Further, historical preference information of the sampling historical user on at least one commodity, such as CVR and CTR, is obtained from historical preference information of the historical user on the at least one commodity.
And (3) linear programming model construction operation: and constructing a linear programming model which takes the recommendation probability of the at least one commodity as a decision variable and takes the maximum expected income of the sampled historical user to the at least one commodity in the current time window as a target on the basis of the extracted historical preference information of the historical user to the at least one commodity and the price attribute of the at least one commodity.
In an alternative embodiment, CVR and CTR of the historical user for the commodity are used to represent historical preference information of the historical user for the commodity, and the objective function of the constructed linear programming model can be expressed by the following formula:
Figure BDA0002938467080000151
further, in the process of constructing the linear programming model, the constraint conditions of the linear programming model can be constructed by considering the average distribution of the commodity inventory and the constraint of the maximum recommendable commodity quantity at each time. Specifically, the method comprises the following steps: in combination with the number k of possible users arriving within the current time windowcurAnd current inventory information of the at least one commodity, determining the distribution amount of the at least one commodity in the current time window
Figure BDA0002938467080000152
According to the distribution amount of at least one commodity in the current time window
Figure BDA0002938467080000153
And the maximum recommendable inventory resource variety number K at a timeConstraints of the linear programming model. This constraint can be expressed as the following equations (2) and (3):
Figure BDA0002938467080000154
Figure BDA0002938467080000161
the above equation (1) represents that the expected profit of the sampling history user on N commodities is maximized in the current time window; wherein the content of the first and second substances,
Figure BDA0002938467080000162
indicating the expected revenue of the sample history user u for the ith good within the current time window,
Figure BDA0002938467080000163
the CTR of the sampling history user u for the ith commodity is shown,
Figure BDA0002938467080000164
CVR, r representing sampling history of user u for ith commodityiRepresenting the price attribute, x, of the ith itemiuRepresenting the recommendation probability of the ith commodity for the sampling history user u, and Ts representing the total number of the sampling history users, and optionally, Ts ═ kcurAnd N represents a total number of at least one commodity.
In the above formula (1), xiuIs a decision variable and satisfies
Figure BDA0002938467080000165
In the linear programming model of the present embodiment, the decision variable xiuIs not a binary decision variable of 0-1, but is relaxed to [0,1 ]]Continuous variables over the range, but the solution of the corresponding integer programming model is still a 0-1 variable.
The above formula (2) is a concrete representation of the balance requirement of the inventory of goods, which is allocated to each time window after the initial inventory at the beginning of the current time window is properly adjusted, andthe total quantity of the ith commodity sold to Ts users in the current time window is required to be less than or equal to the distribution quantity of the commodity in the current time window,
Figure BDA0002938467080000166
indicating the distribution amount of the ith product in the current time window.
The above formula (3) represents that the number of commodities that can be recommended to the user at a time does not exceed the maximum number, where K represents the maximum number of commodities that can be recommended to the user at a time.
Linear scale model solving operation: and solving the linear programming model based on a dual theory to obtain the shadow price of at least one commodity, wherein the shadow price of each commodity is a dual value of the recommendation probability of the commodity and reflects the shortage of the commodity. Wherein, the shortage degree of the commodity also reflects the popularity of the commodity. The embodiment of the present application is not described in detail with respect to the process of solving the linear programming model based on the dual theory.
After the shadow price of each commodity is obtained in the above manner, in the current time window, when a target user initiates a page request for a page embedded with a recommendation function, the server device 102 predicts CVR and CTR of the target user for at least one commodity according to a user identifier carried in the page request, on one hand, according to portrait data corresponding to the user identifier and basic attribute information of the at least one commodity, and records the CVR and CTR as
Figure BDA0002938467080000167
And
Figure BDA0002938467080000171
where the superscript cur denotes the current time window, the subscript i denotes the ith good, the subscript o denotes the target user,
Figure BDA0002938467080000172
indicating the CVR of the target user for the ith item,
Figure BDA0002938467080000173
and representing the CTR of the target user to the ith commodity. On the other hand, the server device 102 may obtain a shadow price, denoted as α, of at least one commodityiRepresenting the shadow price of the ith commodity; the price attribute of at least one commodity is corrected according to the shadow price of at least one commodity to obtain the corrected price of at least one commodity, and the corrected price of the ith commodity can be expressed as ri-αi(ii) a Further, determining expected income of at least one commodity according to the CVR and the CTR of the target user on the at least one commodity and the corrected price of the at least one commodity; the expected revenue for the ith good may be expressed as
Figure BDA0002938467080000174
Generating an inventory penalty factor of at least one commodity according to the current inventory information of the at least one commodity and the initial inventory information of the current time window, wherein the inventory penalty factor for the ith commodity can be expressed as
Figure BDA0002938467080000175
Wherein f (x) represents an inventory penalty factor function,
Figure BDA0002938467080000176
current inventory information representing the ith item, superscript t representing the current time,
Figure BDA0002938467080000177
initial inventory information representing the ith commodity in the current time window; correcting the expected income of at least one commodity according to the inventory penalty factor of at least one commodity to obtain the corrected expected income of at least one commodity, wherein the corrected expected income of the ith commodity can be expressed as
Figure BDA0002938467080000178
And sorting the at least one commodity in a descending order based on the corrected expected income of the at least one commodity, and recommending one or more commodities which are ranked most at the top as target commodities.
In the embodiment, the shadow price and the inventory balance are organically combined, and the modeling capability of linear programming, the theoretical basis of dual theory and the dynamic property of the inventory balance algorithm are integrated. When a linear programming model for calculating the shadow price is constructed, limited commodity inventory is reasonably scaled, the linear programming problem based on historical data is periodically solved to dynamically update the commodity shadow price, and an inventory penalty factor based on real-time inventory is introduced to adjust expected profit indexes of various commodities in real time, so that the conservatism caused by only considering inventory balance can be properly reduced, the defect that the shadow price is always static in a single period and cannot reflect user difference can be overcome, commodity recommendation can be more accurately carried out, and the profit of merchants can be prevented from being lost. In addition, due to the simplicity of the application of the inventory balance algorithm, the inventory information of the supply chain management perspective can be integrated on the premise of hardly increasing extra calculation load, and the personalized recommendation problem is considered from the global perspective. Furthermore, in the embodiment of the application, the characteristics of the supply chain end and the logistics operation are fully considered, so that the front-end commodity recommendation and the commodity inventory management are combined, from the perspective of a merchant, the commodity recommendation is not performed purely based on the purchase probability or the expected profit of a single product, but the commodity recommendation is performed by fusing the commodity inventory of the supply chain end, and the GMV of the merchant can be maximized by explicitly introducing real-time inventory level information to perform personalized commodity recommendation.
In practical application, most commodities have valid periods, some commodities have longer valid periods, some commodities have shorter valid periods, and the commodities need to be sold in the valid periods. Especially, some fresh products such as vegetables, fresh milk, meat and the like have short effective period and cannot be sold after rotting. The expiration date of a good has some effect on the shadow price of the good. Based on this, in some optional embodiments of the present application, in the process of calculating the shadow price of the commodity, residual value information of the commodity can be further introduced, and the residual value information is determined by the validity period of the commodity. Each commodity has a residual value information, the residual value information of different commodities is different, the longer the validity period of the commodity is, the larger the residual value information is, and conversely, the shorter the validity period of the commodity is, the smaller the residual value information is, even some commodities such as fish, fresh milk, meat and the like exist, and the residual value information may also be a negative value.
Alternatively, the residual value information of each commodity can be determined in advance according to the validity period of the commodity by adopting a corresponding residual value determination rule. For example, if the selling price of a commodity is 10 yuan in the morning, the selling price is reduced to 6 yuan in the evening, and the residual value information of the commodity is 6 yuan. Or the selling price of a commodity is 15 yuan in the morning, the commodity needs to be processed into other foods in the evening, and the selling price of the processed food is 5 yuan, so that the residual value information of the commodity is 5 yuan. The manner of determining the commodity residual value information here is merely an example, and is not limited thereto.
The process of estimating the shadow price of the commodity when the residual value information of the commodity is introduced is similar to the process of estimating the shadow price of the commodity when the residual value information of the commodity is not introduced as shown in fig. 2, and the difference is mainly in the construction process of a linear programming model. Under the condition of introducing residual value information of commodities, the process of constructing the linear programming model comprises the following steps:
constructing an objective function: generating a basic expected revenue function of the historical user for the at least one commodity in the current time window according to the historical preference information of the sampling historical user for the at least one commodity, the price attribute of the at least one commodity and the recommendation probability of the at least one commodity; generating a loss expected revenue function of the at least one commodity in a current time window according to the residual value information of the at least one commodity and the current inventory information of the at least one commodity; and taking the sum of the maximized basic expected revenue function and the lost expected revenue function as an objective function of the linear programming model. The objective function can be expressed as the following equation (4):
Figure BDA0002938467080000191
the above formula (1) represents maximizing the sum of the basic expected revenue function and the loss expected revenue function; wherein the content of the first and second substances,
Figure BDA0002938467080000192
a base expected revenue function is represented that is,
Figure BDA0002938467080000193
representing a loss expected revenue function; w is aiThe residual value information indicating the ith product,
Figure BDA0002938467080000194
indicating the remaining stock information of the ith product. For the description of other parameters, reference may be made to the foregoing embodiments, which are not repeated herein.
And (3) construction of constraint conditions: constraints of the linear programming model are constructed taking into account constraints of the average distribution of inventory of the goods, the remaining inventory of the goods and the maximum number of goods recommendable at a time. Specifically, the method comprises the following steps: in combination with the number k of possible users arriving within the current time windowcurAnd current inventory information of the at least one commodity, determining the distribution amount of the at least one commodity in the current time window
Figure BDA0002938467080000195
According to the distribution amount of at least one commodity in the current time window
Figure BDA0002938467080000196
And constructing the constraint conditions of the linear programming model by using the residual inventory information of at least one commodity and the recommended inventory resource amount K at most each time. This constraint can be expressed as the following equations (5) and (6):
Figure BDA0002938467080000197
Figure BDA0002938467080000198
the above formula (5) is a representation of the requirement for balance of inventory of goods, and the initial inventory at the beginning of the current time window is properly adjusted and then distributed to the subsequent time windows, and the ith goods are required to be sold to Ts in the current time windowThe sum of the total number of users and the remaining inventory information of the ith product should be less than or equal to the amount of the product allocated in the current time window,
Figure BDA0002938467080000199
indicating the distribution amount of the ith product in the current time window. Wherein, it is required to satisfy
Figure BDA00029384670800001910
Equation (6) is the same as equation (3), and is not described in detail herein.
Similarly, the linear programming model may be solved based on a dual theory to obtain a shadow price of the at least one commodity. After the shadow price of the at least one commodity is obtained, a commodity recommendation process is performed on the CVR and CTR of the at least one commodity based on the shadow price of the at least one commodity, the current inventory information, and the target user, which is the same as the foregoing embodiment, and is not described herein again.
In the above or below embodiments of the present application, the calculation function f (x) used for calculating the inventory penalty factor is not limited. f (x) may be any non-decreasing concave function, and satisfies f (0) ═ 0 and f (1) ═ 1. In an alternative embodiment, a combination of the above may be used
Figure BDA0002938467080000201
The calculation function can ensure that the competition ratio of the algorithm is at least up to
Figure BDA0002938467080000202
In the above embodiments of the present application
Figure BDA0002938467080000203
In the case of (a) in (b),
Figure BDA0002938467080000204
in another alternative embodiment, f (x) x may be used.
In the following optional embodiments of the present application, in the process of calculating the inventory penalty factor, in addition to the current inventory information and the initial inventory information of the product, other information having an influence on the inventory information may be considered, and the inventory penalty factor is calculated by fusing the multi-source information.
In an alternative embodiment S1, consider an initial inventory
Figure BDA0002938467080000205
Current inventory information
Figure BDA0002938467080000206
And predicting transaction information
Figure BDA0002938467080000207
Considering that at least one commodity may support multiple transaction channels, a transaction channel refers to a channel through which a user may purchase commodities on line, including but not limited to: applets and APPs developed by the merchants themselves, and APPs and applets of third parties collaborating with the merchants. The business self-developed small program, APP, the third party small program and the third party APP belong to different transaction channels. For the sake of distinction and description, a transaction channel used by a target user to initiate a page request is referred to as a target transaction channel, and a transaction channel different from the target transaction channel is referred to as another transaction channel.
In this case, the initial inventory of the current time window may be considered simultaneously when calculating the inventory penalty factor
Figure BDA0002938467080000208
Current inventory information
Figure BDA0002938467080000209
And the predicted trade information of the current commodities on other trade channels is recorded
Figure BDA00029384670800002010
The superscript j represents the target transaction channel, which may be 1,2,3,4, etc., the value of-j represents the other transaction channels, the following table i represents the ith good, the subscript t represents the current time,
Figure BDA00029384670800002011
and the predicted transaction information of the ith commodity on other transaction channels currently is shown. That is, the parameters x and x in the inventory penalty function f (x)
Figure BDA00029384670800002012
And
Figure BDA00029384670800002013
in this embodiment, the parameters x and x are not correlated
Figure BDA00029384670800002014
And
Figure BDA00029384670800002015
the relationship between them is defined. For example, the parameters x and
Figure BDA0002938467080000211
and
Figure BDA0002938467080000212
one relationship between can be expressed as:
Figure BDA0002938467080000213
the inventory penalty factor may be expressed as
Figure BDA0002938467080000214
But is not limited thereto. Wherein, y+＝max{0,y},
Figure BDA0002938467080000215
The prediction may be based on the predicted transaction information of the at least one commodity currently on the other transaction channel based on historical transaction information of the at least one commodity on the other transaction channel before using the predicted transaction information of the at least one commodity currently on the other transaction channel. For any commodity, its historical transaction information on other transaction channels includes but is not limited to: the transaction time (such as weekday, weekend, holiday), the transaction amount, the characteristics of the transaction commodity, the current traffic information of the transaction channel, the current marketing information of the transaction channel, and the like.
In an alternative embodiment S2, consider an initial inventory
Figure BDA0002938467080000216
Current inventory information
Figure BDA0002938467080000217
And predicting transaction information
Figure BDA0002938467080000218
Is inaccurate.
The method of claim 1, further comprising the step of considering predicted transaction information of at least one commodity currently in other transaction channels
Figure BDA0002938467080000219
The situation cannot be predicted accurately. In this case, a relatively robust processing approach may be employed. For example, predicted transaction information that is less numeric but is more likely to occur may be estimated. Optionally, the predicted transaction information of at least one commodity on other transaction channels can be predicted
Figure BDA00029384670800002110
Calculating the current predicted transaction information of at least one commodity on other transaction channels
Figure BDA00029384670800002111
According to the standard deviation, predicting the transaction information of at least one commodity on other transaction channels
Figure BDA00029384670800002112
Making corrections based on the corrected predicted transaction information
Figure BDA00029384670800002113
An inventory penalty factor is calculated. Alternatively, the following approach may be taken for predictionTransaction information
Figure BDA00029384670800002114
Making a correction to obtain
Figure BDA00029384670800002115
Wherein the content of the first and second substances,
Figure BDA00029384670800002116
the standard deviation is indicated.
Based on the above, in the present embodiment, the parameters x and x in the inventory penalty function f (x)
Figure BDA00029384670800002117
And
Figure BDA00029384670800002118
in this embodiment, the parameters x and x are not correlated
Figure BDA00029384670800002119
And
Figure BDA00029384670800002120
the relationship between them is defined. For example, the parameters x and
Figure BDA00029384670800002121
and
Figure BDA00029384670800002122
one relationship between can be expressed as:
Figure BDA00029384670800002123
the inventory penalty factor may be expressed as
Figure BDA0002938467080000221
But is not limited thereto. Wherein the content of the first and second substances,
Figure BDA0002938467080000222
in an alternative embodiment S3, consider the initialStock keeping
Figure BDA0002938467080000223
Current inventory information
Figure BDA0002938467080000224
Predicting transaction information
Figure BDA0002938467080000225
Inaccuracy and lag in offline and online synchronization of inventory information.
On the basis of the alternative embodiment S2, the problem of a certain delay in the synchronization of the offline inventory information to the online is further considered. For example, if the inventory manager updates the inventory information at 11:00 and synchronizes the updated inventory information to the server device 102 at 11:20, a certain number of products may be sold in the offline transaction channel during the period of 11:00-11:20, and if the inventory manager still processes the inventory information at 11:00 at 11:20, there will be an error. In view of this, in the process of calculating the inventory penalty factor, a "Protection Level" (Protection Level) in a period of synchronizing the online-off inventory information of the product to the online is further considered, so that the situation that no inventory exists when the user purchases the recommended product after the product recommendation is performed according to the delayed inventory information is prevented, and the shopping experience of the user is reduced. Taking the time period of 11:00-11:20 as an example, the time period is the period of online synchronization of offline inventory information. In this embodiment, the "protection level" during the period of online synchronization of the online inventory information of the product with the online is referred to as predicted transaction information during the period of online synchronization of the online inventory information of the product with the online inventory information. For the ith product, recording the predicted transaction information of the product in the period of synchronizing the current off-line inventory information to the on-line as PLi。
Alternatively, the predicted transaction information of each commodity in the period of the online synchronization of the current offline inventory information can be predicted according to the historical transaction information of each commodity in the period of the online synchronization of the historical offline inventory information of each commodity. For example, for the ith commodity, the sales volume of the commodity in the period of synchronizing the online inventory information of the commodity to the online in the recent month may be counted, and the sales volume of the commodity in the period of synchronizing the current online inventory information of the commodity to the online in the recent month may be predicted according to the sales volume of the commodity in the period of synchronizing the online inventory information of the commodity to the online in the recent month.
For each commodity, after the predicted transaction information of the commodity in the period that the current off-line inventory information is synchronized to the on-line is obtained, when the inventory penalty factor is calculated for the commodity, the quantity of the commodities sold in the period that the on-line inventory information is synchronized to the on-line can be removed (namely the predicted transaction information), so that when the commodity is recommended to the on-line user based on the inventory penalty factor, when the inventory level is low, the corresponding commodity cannot be pushed, and the situation that the user has no inventory when purchasing the recommended commodity is avoided.
Based on the above, in the present embodiment, the parameters x and x in the inventory penalty function f (x)
Figure BDA0002938467080000231
And PLiIn this embodiment, the parameters x and x are not correlated
Figure BDA0002938467080000232
And PLiThe relationship between them is defined. For example, the parameters x and
Figure BDA0002938467080000233
and PLiOne relationship between can be expressed as:
Figure BDA0002938467080000234
the inventory penalty factor may be expressed as
Figure BDA0002938467080000235
But is not limited thereto. Wherein, y+＝max{0,y},
Figure BDA0002938467080000236
In the above alternative embodiment S3, the predicted transaction information is also considered
Figure BDA0002938467080000237
And predicted transaction information PLiBut is not limited thereto. In another alternative embodiment of the present application, only initial inventory may be considered
Figure BDA0002938467080000238
Current inventory information
Figure BDA0002938467080000239
And predicted transaction information PLiConcerning consideration of only initial inventory
Figure BDA00029384670800002310
Current inventory information
Figure BDA00029384670800002311
And predicted transaction information PLiIn a manner similar to that described above, and will not be described in detail. That is, in the process of generating the inventory penalty factor of at least one commodity, at least one predicted transaction information corresponding to the at least one commodity may be obtained, where the at least one predicted transaction information includes predicted transaction information of the at least one commodity currently on other transaction channels and/or predicted transaction information of the at least one commodity during a period when the current offline inventory information is synchronized online; and then generating an inventory penalty factor of at least one commodity according to the current inventory information and the initial inventory information of at least one commodity and at least one predicted transaction information corresponding to the at least one commodity.
After the stock penalty factor of at least one commodity is obtained by adopting the above optional embodiments, the expected income of at least one commodity can be corrected based on the stock penalty factor of at least one commodity; and recommending the target commodity to the target user based on the corrected expected income of the at least one commodity. These operations are the same as the previous embodiments and are not described herein.
In the optional embodiments, additional predicted transaction information is introduced, and the inventory penalty factor can be calculated more accurately under the condition that the prediction has deviation, so that the accuracy of commodity recommendation based on the inventory penalty factor is improved, commodities are recommended to a user more reasonably, and the probability of invalid commodity recommendation is reduced.
Further, in the embodiment of the present application, offline simulation is performed on the embodiment of the present application, and the following beneficial results are obtained in the indexes of the embodiment of the present application, such as revenue and inventory, of the merchant:
1. the scheme of the embodiment of the application can give consideration to both the user preference and the stability of user arrival, and the GMV of the merchant is improved more obviously along with the poorer the user preference and the stability of intensity achieved by the user; in a simulation experiment, compared with the existing recommendation method, the GMV improved by the scheme of the embodiment of the application can reach 1% or even 2% at most.
2. Under the condition that the initial inventory amount is equivalent to the quantity of commodities expected to be purchased by a user on the same day, the improvement effect of the scheme of the embodiment of the application on the GMV index is particularly obvious.
Fig. 3 is a flowchart illustrating an information recommendation method according to an exemplary embodiment of the present application. As shown in fig. 3, the method includes:
31. receiving a page request sent by terminal equipment, wherein the page request comprises a user identifier which is used for identifying a target user initiating a page request operation;
32. acquiring the shortage information and the current inventory information of at least one inventory resource which can be transacted on line, and predicting the preference information of a target user on the at least one inventory resource;
33. selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource;
34. and sending the information of the target inventory resource to the terminal equipment so that the terminal equipment can display the information of the target inventory resource on a page requested by the target user.
In an alternative embodiment, obtaining scarcity information for at least one inventory resource operable on the line comprises: estimating the shadow price of at least one stock resource based on historical preference information of a historical user on the at least one stock resource which can be traded on line; wherein the shadow price of each inventory resource reflects the scarcity of that inventory resource.
Further optionally, estimating a shadow price of the at least one inventory resource based on historical user preference information for the at least one inventory resource that is tradable online, comprising: sampling historical preference information of the historical user on at least one inventory resource to obtain the historical preference information of the sampled historical user on the at least one inventory resource; and estimating the preference of the user arriving in the future on the at least one inventory resource based on the historical preference information of the sampled historical user on the at least one inventory resource so as to obtain the shadow price of the at least one inventory resource.
Further optionally, sampling the historical preference information of the historical user on the at least one inventory resource to obtain the historical preference information of the sampled historical user on the at least one inventory resource, including: aiming at the current time window, predicting the number of users which are possible to arrive in the current time window according to the number of historical users appearing in the historical synchronization time window; sampling from historical users appearing in a historical contemporaneous time window according to the number of users possibly arriving in the current time window to obtain sampled historical users;
and acquiring historical preference information of the sampled historical user on at least one type of inventory resource from the historical preference information of the historical user on the at least one type of inventory resource.
Further optionally, estimating the preference of the user to the at least one inventory resource in the future based on the historical preference information of the sampled historical user to the at least one inventory resource to obtain the shadow price of the at least one inventory resource, comprising:
based on the historical preference information of the sampling historical user on at least one stock resource and the price attribute of the at least one stock resource, constructing a linear programming model which takes the recommendation probability of the at least one stock resource as a decision variable and takes the maximum expected income of the sampling historical user on the at least one stock resource in the current time window as a target;
on the basis of a dual theory, the linear planning model is solved to obtain the shadow price of at least one inventory resource, and the shadow price of each inventory resource is a dual value of the recommendation probability of the inventory resource and reflects the shortage of the inventory resource.
Further optionally, in the process of constructing the linear programming model, the method further includes:
determining the allocation amount of at least one inventory resource in the current time window by combining the number of users which are possible to reach in the current time window and the current inventory information of at least one inventory resource;
and constructing the constraint condition of the linear programming model according to the distribution amount of at least one stock resource in the current time window and the recommended stock resource amount at most each time.
Further optionally, constructing a linear programming model with the recommendation probability of the at least one inventory resource as a decision variable and the expected profit of the sampled historical user on the at least one inventory resource within the current time window being the maximum target based on the historical preference information of the sampled historical user on the at least one inventory resource and the price attribute of the at least one inventory resource, includes:
generating a basic expected revenue function of the historical user to the at least one inventory resource in the current time window according to the historical preference information of the sampled historical user to the at least one inventory resource, the price attribute of the at least one inventory resource and the recommendation probability of the at least one inventory resource;
generating a loss expected revenue function of the at least one inventory resource in a current time window according to the residual value information of the at least one inventory resource and the current inventory information of the at least one inventory resource, wherein the residual value information of the inventory resource is determined according to the validity period of the inventory resource;
and taking the sum of the maximized basic expected revenue function and the lost expected revenue function as an objective function of the linear programming model.
In an optional embodiment, selecting the target inventory resource from the at least one inventory resource according to the scarcity information of the at least one inventory resource, the current inventory information, and the preference information of the target user for the at least one inventory resource includes:
determining expected income of at least one stock resource according to the shortage information of the at least one stock resource and the preference information of the target user on the at least one stock resource;
correcting the expected income of at least one inventory resource according to the current inventory information of the at least one inventory resource;
and selecting the target inventory resource from the at least one inventory resource according to the corrected expected income of the at least one inventory resource.
Further optionally, predicting the preference information of the target user for the at least one inventory resource comprises: and predicting the click through rate and the click conversion rate of the target user on at least one inventory resource based on the portrait data of the target user. Accordingly, determining the expected profit of the at least one inventory resource according to the scarcity information of the at least one inventory resource and the preference information of the target user for the at least one inventory resource comprises: correcting the price attribute of at least one stock resource according to the shortage degree information of at least one stock resource to obtain the corrected price of at least one stock resource; and determining the expected profit of the at least one inventory resource according to the click through rate and the click conversion rate of the target user on the at least one inventory resource and the corrected price of the at least one inventory resource.
Further optionally, modifying the expected revenue of the at least one inventory resource based on current inventory information of the at least one inventory resource comprises:
generating an inventory penalty factor of at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource;
and correcting the expected profit of the at least one inventory resource by using the inventory penalty factor of the at least one inventory resource.
Further optionally, generating an inventory penalty factor for the at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource, includes:
acquiring at least one predicted transaction information corresponding to at least one inventory resource, wherein the at least one predicted transaction information comprises predicted transaction information of the at least one inventory resource on other transaction channels at present and/or predicted transaction information of the at least one inventory resource in a period of synchronization of the current off-line inventory information to the on-line;
generating an inventory penalty factor of at least one inventory resource according to the current inventory information and the initial inventory information of at least one inventory resource and at least one predicted transaction information corresponding to at least one inventory resource;
the other transaction channels are transaction channels other than the target transaction channel in the multiple transaction channels supported by the at least one inventory resource, and the target transaction channel is a transaction channel used by the terminal device for initiating the page request.
Further optionally, obtaining the predicted transaction information of the at least one inventory resource currently on other transaction channels includes: predicting the current predicted transaction information of at least one inventory resource on other transaction channels based on the historical transaction information of the at least one inventory resource on other transaction channels;
accordingly, obtaining predicted transaction information for at least one inventory resource during a synchronization of current offline inventory information to online, comprises: and predicting the predicted transaction information of the at least one inventory resource in the online synchronization period of the current offline inventory information according to the historical transaction information of the at least one inventory resource in the online synchronization period of the historical offline inventory information.
Further optionally, before using the predicted transaction information of the at least one inventory resource currently on the other transaction channel, further comprising: and correcting the predicted transaction information of the at least one inventory resource on other transaction channels according to the standard deviation of the predicted transaction information of the at least one inventory resource on other transaction channels.
In an optional embodiment, the at least one inventory resource is a commodity, and accordingly, the page requested by the target user is a home page of the shopping application, a shopping cart page, a group page, or a user detail page.
In the method embodiment of the present application, the description is developed by using the inventory resource as a description object, but the process is the same as or detailed in the process of describing the commodity as an object, so that for detailed implementation and description of the above steps, reference may be made to the system embodiment, and details are not repeated here.
In the embodiment, for the stock resources which can be traded online, the front-end recommendation is combined with the stock information of the supply chain end, and meanwhile, the multi-dimensional information such as the shortage information, the current stock information and the preference information of the target user for the stock resources is fused, so that the information recommendation method based on the stock balance is realized.
Fig. 4 is a flowchart illustrating an information obtaining method according to an exemplary embodiment of the present application. As shown in fig. 4, the method includes:
41. responding to the page request operation, and sending a page request to the server-side equipment, wherein the page request comprises a user identifier which is used for identifying a target user initiating the page request operation;
42. receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by a target user; the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
For a detailed implementation process that the server device selects the target inventory resource from the information on the scarcity of the at least one inventory resource, the current inventory information, and the preference information of the target user on the at least one inventory resource, and returns the information on the target inventory resource to the terminal device, reference may be made to the foregoing embodiment, which is not described herein again.
It should be noted that the execution subjects of the steps of the methods provided in the above embodiments may be the same device, or different devices may be used as the execution subjects of the methods. For example, the execution subjects of steps 41 to 42 may be device a; for another example, the execution subject of step 41 may be device a, and the execution subject of step 42 may be device B; and so on.
In addition, in some of the flows described in the above embodiments and the drawings, a plurality of operations are included in a specific order, but it should be clearly understood that the operations may be executed out of the order presented herein or in parallel, and the sequence numbers of the operations, such as 41, 42, etc., are merely used for distinguishing different operations, and the sequence numbers do not represent any execution order per se. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.
Fig. 5 is a schematic structural diagram of a server device according to an exemplary embodiment of the present application. As shown in fig. 5, the server device includes: memory 51, processor 52, and communications component 53.
The memory 51 is used for storing computer programs and may be configured to store other various data to support operations on the server device. Examples of such data include instructions, messages, pictures, videos, etc. for any application or method operating on the server device.
The memory 51 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
A processor 52 coupled to the memory 51 for executing the computer program in the memory 51 for:
receiving a page request sent by a terminal device through a communication component 53, where the page request includes a user identifier, and the user identifier is used to identify a target user initiating a page request operation; acquiring the shortage information and the current inventory information of at least one inventory resource which can be transacted on line, and predicting the preference information of a target user on the at least one inventory resource; selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource; the information of the target inventory resource is sent to the terminal device through the communication component 53, so that the terminal device can display the information of the target inventory resource on the page requested by the target user.
In an alternative embodiment, processor 52, when obtaining the scarcity information of at least one inventory resource operable on the line, is specifically configured to: estimating the shadow price of at least one stock resource based on historical preference information of a historical user on the at least one stock resource which can be traded on line; wherein the shadow price of each inventory resource reflects the scarcity of that inventory resource.
Further optionally, the processor 52, when predicting the shadow price of the at least one inventory resource based on historical preference information of the historical user for the at least one inventory resource that can be traded online, is specifically configured to: sampling historical preference information of the historical user on at least one inventory resource to obtain the historical preference information of the sampled historical user on the at least one inventory resource; and estimating the preference of the user arriving in the future on the at least one inventory resource based on the historical preference information of the sampled historical user on the at least one inventory resource so as to obtain the shadow price of the at least one inventory resource.
Further, when sampling historical preference information of the historical user for the at least one inventory resource, the processor 52 is specifically configured to:
aiming at the current time window, predicting the number of users which are possible to arrive in the current time window according to the number of historical users appearing in the historical synchronization time window;
sampling from historical users appearing in a historical contemporaneous time window according to the number of users possibly arriving in the current time window to obtain sampled historical users;
and acquiring historical preference information of the sampled historical user on at least one type of inventory resource from the historical preference information of the historical user on the at least one type of inventory resource.
Further, the processor 52 is specifically configured to, when estimating the future user's preference of the at least one inventory resource based on the historical preference information of the sampled historical users for the at least one inventory resource to obtain the shadow price of the at least one inventory resource:
based on the historical preference information of the sampling historical user on at least one stock resource and the price attribute of the at least one stock resource, constructing a linear programming model which takes the recommendation probability of the at least one stock resource as a decision variable and takes the maximum expected income of the sampling historical user on the at least one stock resource in the current time window as a target;
on the basis of a dual theory, the linear planning model is solved to obtain the shadow price of at least one inventory resource, and the shadow price of each inventory resource is a dual value of the recommendation probability of the inventory resource and reflects the shortage of the inventory resource.
Further optionally, the processor 52, in constructing the linear programming model, is further configured to: determining the allocation amount of at least one inventory resource in the current time window by combining the number of users which are possible to reach in the current time window and the current inventory information of at least one inventory resource; and constructing the constraint condition of the linear programming model according to the distribution amount of at least one stock resource in the current time window and the recommended stock resource amount at most each time.
Further optionally, the processor 52 is specifically configured to, when constructing a linear programming model with the recommended probability of the at least one inventory resource as a decision variable and with the expected profit of the sampled historical user for the at least one inventory resource within the current time window being the maximum target:
generating a basic expected revenue function of the historical user to the at least one inventory resource in the current time window according to the historical preference information of the sampled historical user to the at least one inventory resource, the price attribute of the at least one inventory resource and the recommendation probability of the at least one inventory resource;
generating a loss expected revenue function of the at least one inventory resource in a current time window according to the residual value information of the at least one inventory resource and the current inventory information of the at least one inventory resource, wherein the residual value information of the inventory resource is determined according to the validity period of the inventory resource;
and taking the sum of the maximized basic expected revenue function and the lost expected revenue function as an objective function of the linear programming model.
In an optional embodiment, the processor 52, when selecting the target inventory resource from the at least one inventory resource according to the scarcity information of the at least one inventory resource, the current inventory information, and the preference information of the target user for the at least one inventory resource, is specifically configured to:
determining expected income of at least one stock resource according to the shortage information of the at least one stock resource and the preference information of the target user on the at least one stock resource;
correcting the expected income of at least one inventory resource according to the current inventory information of the at least one inventory resource;
and selecting the target inventory resource from the at least one inventory resource according to the corrected expected income of the at least one inventory resource.
Further optionally, the processor 52, when predicting the preference information of the target user for the at least one inventory resource, is specifically configured to: predicting the click through rate and the click conversion rate of the target user on at least one inventory resource based on the portrait data of the target user;
accordingly, the processor 52, when determining the expected profit for the at least one inventory resource based on the scarcity information of the at least one inventory resource and the preference information of the target user for the at least one inventory resource, is specifically configured to: correcting the price attribute of at least one stock resource according to the shortage degree information of at least one stock resource to obtain the corrected price of at least one stock resource; and determining the expected profit of the at least one inventory resource according to the click through rate and the click conversion rate of the target user on the at least one inventory resource and the corrected price of the at least one inventory resource.
Further optionally, when the expected profit of the at least one inventory resource is corrected according to the current inventory information of the at least one inventory resource, the processor 52 is specifically configured to: generating an inventory penalty factor of at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource; and correcting the expected profit of the at least one inventory resource by using the inventory penalty factor of the at least one inventory resource.
Further optionally, when the processor 52 generates the inventory penalty factor of the at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource, it is specifically configured to:
acquiring at least one predicted transaction information corresponding to at least one inventory resource, wherein the at least one predicted transaction information comprises predicted transaction information of the at least one inventory resource on other transaction channels at present and/or predicted transaction information of the at least one inventory resource in a period of synchronization of the current off-line inventory information to the on-line;
generating an inventory penalty factor of at least one inventory resource according to the current inventory information and the initial inventory information of at least one inventory resource and at least one predicted transaction information corresponding to at least one inventory resource;
the other transaction channels are transaction channels other than the target transaction channel in the multiple transaction channels supported by the at least one inventory resource, and the target transaction channel is a transaction channel used by the terminal device for initiating the page request.
Further optionally, the processor 52, when obtaining the predicted transaction information of at least one inventory resource currently on other transaction channels, is specifically configured to: predicting the current predicted transaction information of at least one inventory resource on other transaction channels based on the historical transaction information of the at least one inventory resource on other transaction channels;
accordingly, the processor 52, when obtaining the predicted transaction information of at least one inventory resource during the period of synchronizing the current offline inventory information to the online, is specifically configured to: and predicting the predicted transaction information of the at least one inventory resource in the online synchronization period of the current offline inventory information according to the historical transaction information of the at least one inventory resource in the online synchronization period of the historical offline inventory information.
Further optionally, the processor 52, prior to using the predicted transaction information of the at least one inventory resource currently on the other transaction channel, is further configured to: and correcting the predicted transaction information of the at least one inventory resource on other transaction channels according to the standard deviation of the predicted transaction information of the at least one inventory resource on other transaction channels.
In an optional embodiment, the page is a home page of a shopping application, a shopping cart page, a group page or a user detail page; the at least one inventory resource is a commodity.
Further, as shown in fig. 5, the server device further includes: power supply components 54, and the like. Only some of the components are schematically shown in fig. 5, and the server device is not meant to include only the components shown in fig. 5.
Accordingly, the present application further provides a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to implement the steps that can be performed by the server device in the above method embodiments.
Accordingly, the present application further provides a computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the processor is enabled to implement the steps that can be executed by the server device in the foregoing method embodiments.
Fig. 6 is a schematic structural diagram of a terminal device according to an exemplary embodiment of the present application. As shown in fig. 6, the terminal device includes: memory 61, processor 62 and communication component 63.
The memory 61 is used for storing computer programs and may be configured to store other various data to support operations on the terminal device. Examples of such data include instructions, messages, pictures, videos, etc. for any application or method operating on the terminal device.
The memory 61 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
A processor 62, coupled to the memory 61, for executing computer programs in the memory 61 for:
responding to page request operation, and sending a page request to server equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating the page request operation;
receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by the target user;
the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
Further, as shown in fig. 6, the terminal device further includes: a display 64, audio components 65, power components 66, and the like. Only some of the components are schematically shown in fig. 6, and the terminal device is not meant to include only the components shown in fig. 6.
Accordingly, the present application further provides a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to implement the steps that can be performed by the terminal device in the above method embodiments.
Accordingly, the present application also provides a computer program product, which includes a computer program/instruction, and when the computer program/instruction is executed by a processor, the processor is enabled to implement the steps that can be executed by the terminal device in the foregoing method embodiments.
The communication components of fig. 5 and 6 described above are configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as a WiFi, a 2G, 3G, 4G/LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
The display in fig. 6 described above includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.
The power supply components of fig. 5 and 6 described above provide power to the various components of the device in which the power supply components are located. The power components may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device in which the power component is located.
The audio component of fig. 6 described above may be configured to output and/or input an audio signal. For example, the audio component includes a Microphone (MIC) configured to receive an external audio signal when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in a memory or transmitted via a communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (18)

1. An information recommendation method, comprising:
receiving a page request sent by terminal equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating a page request operation;
acquiring the shortage information and the current inventory information of at least one inventory resource which can be traded on line, and predicting the preference information of the target user on the at least one inventory resource;
selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource;
and sending the information of the target inventory resource to the terminal equipment so that the terminal equipment can display the information of the target inventory resource on a page requested by the target user.
2. The method of claim 1, wherein obtaining scarcity information for at least one inventory resource operable on the line comprises:
estimating the shadow price of at least one stock resource which can be traded on line based on historical preference information of a historical user on the at least one stock resource; wherein the shadow price of each inventory resource reflects the scarcity of that inventory resource.
3. The method of claim 2, wherein estimating the shadow price of at least one inventory resource tradeable online based on historical user preference information for the at least one inventory resource comprises:
sampling historical preference information of the historical user on the at least one inventory resource to obtain historical preference information of the sampled historical user on the at least one inventory resource;
estimating the preference of the user arriving in the future for the at least one inventory resource based on the historical preference information of the sampled historical user for the at least one inventory resource so as to obtain the shadow price of the at least one inventory resource.
4. The method of claim 3, wherein sampling historical preference information of the historical user on the at least one inventory resource to obtain historical preference information of the sampled historical user on the at least one inventory resource comprises:
aiming at the current time window, predicting the number of users which are possible to arrive in the current time window according to the number of historical users appearing in the historical synchronization time window;
sampling from historical users appearing in the historical contemporaneous time window according to the number of users possibly arriving in the current time window to obtain sampled historical users;
and acquiring historical preference information of the sampled historical user on the at least one inventory resource from historical preference information of the historical user on the at least one inventory resource.
5. The method of claim 4, wherein estimating the future user's preference of the at least one inventory resource to obtain the shadow price of the at least one inventory resource based on the historical preference information of the sampled historical users for the at least one inventory resource comprises:
constructing a linear programming model which takes the recommendation probability of the at least one stock resource as a decision variable and takes the expected profit of the sampling historical user on the at least one stock resource in the current time window as the maximum target on the basis of the historical preference information of the sampling historical user on the at least one stock resource and the price attribute of the at least one stock resource;
and solving the linear programming model based on a dual theory to obtain the shadow price of the at least one inventory resource, wherein the shadow price of each inventory resource is a dual value of the recommendation probability of the inventory resource and reflects the shortage of the inventory resource.
6. The method of claim 5, wherein in constructing the linear programming model, further comprising:
determining the allocation amount of the at least one inventory resource in the current time window by combining the number of users which are possible to reach in the current time window and the current inventory information of the at least one inventory resource;
and constructing the constraint condition of the linear programming model according to the allocation amount of the at least one stock resource in the current time window and the recommended maximum stock resource amount at each time.
7. The method of claim 6, wherein constructing a linear programming model with the recommendation probability of the at least one inventory resource as a decision variable and the expected profit of the sampled historical user on the at least one inventory resource within the current time window as the maximum target is based on the historical preference information of the sampled historical user on the at least one inventory resource and the price attribute of the at least one inventory resource comprises:
generating a basic expected revenue function of the historical user for the at least one inventory resource in the current time window according to the historical preference information of the sampling historical user for the at least one inventory resource, the price attribute of the at least one inventory resource and the recommendation probability of the at least one inventory resource;
generating a loss expected revenue function of the at least one inventory resource in a current time window according to the residual value information of the at least one inventory resource and the current inventory information of the at least one inventory resource, wherein the residual value information of the inventory resource is determined according to the valid period of the inventory resource;
and taking the sum of the base expected profit function and the loss expected profit function as an objective function of the linear programming model in a maximized mode.
8. The method according to any one of claims 1-7, wherein selecting the target inventory resource from the at least one inventory resource based on the scarcity information of the at least one inventory resource, the current inventory information, and the preference information of the target user for the at least one inventory resource comprises:
determining expected income of the at least one inventory resource according to the scarcity information of the at least one inventory resource and the preference information of the target user on the at least one inventory resource;
correcting the expected income of the at least one inventory resource according to the current inventory information of the at least one inventory resource;
and selecting a target inventory resource from the at least one inventory resource according to the corrected expected income of the at least one inventory resource.
9. The method of claim 8, wherein predicting the target user's preference information for the at least one inventory resource comprises: predicting click through rate and click conversion rate of the target user on the at least one inventory resource based on the portrait data of the target user;
accordingly, determining the expected profit of the at least one inventory resource according to the scarcity information of the at least one inventory resource and the preference information of the target user for the at least one inventory resource comprises:
correcting the price attribute of the at least one stock resource according to the shortage information of the at least one stock resource to obtain the corrected price of the at least one stock resource;
and determining the expected profit of the at least one inventory resource according to the click through rate and the click conversion rate of the target user on the at least one inventory resource and the corrected price of the at least one inventory resource.
10. The method of claim 8, wherein modifying the expected revenue of the at least one inventory resource based on current inventory information of the at least one inventory resource comprises:
generating an inventory penalty factor of the at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource;
and correcting the expected profit of the at least one inventory resource by using the inventory penalty factor of the at least one inventory resource.
11. The method of claim 10, wherein generating an inventory penalty factor for the at least one inventory resource based on current inventory information and initial inventory information for the at least one inventory resource comprises:
acquiring at least one type of predicted transaction information corresponding to the at least one type of inventory resource, wherein the at least one type of predicted transaction information comprises predicted transaction information of the at least one type of inventory resource on other transaction channels at present and/or predicted transaction information of the at least one type of inventory resource in a period of synchronization of the current off-line inventory information to the on-line state;
generating an inventory penalty factor of the at least one inventory resource according to the current inventory information and the initial inventory information of the at least one inventory resource and at least one predicted transaction information corresponding to the at least one inventory resource;
the other transaction channels are transaction channels other than a target transaction channel in a plurality of transaction channels supported by the at least one inventory resource, and the target transaction channel is a transaction channel used by the terminal device for initiating the page request.
12. The method of claim 11, wherein obtaining predicted transaction information of the at least one inventory resource currently on other transaction channels comprises: predicting the predicted transaction information of the at least one inventory resource on the other transaction channel based on the historical transaction information of the at least one inventory resource on the other transaction channel;
acquiring the predicted transaction information of the at least one inventory resource during the period of synchronizing the current offline inventory information to the online, comprising: and predicting the predicted transaction information of the at least one inventory resource in the online synchronization period of the current offline inventory information according to the historical transaction information of the at least one inventory resource in the online synchronization period of the historical offline inventory information.
13. The method of claim 12, further comprising, prior to using the predicted transaction information of the at least one inventory resource currently on the other transaction channel:
and correcting the predicted transaction information of the at least one inventory resource on the other transaction channels according to the standard deviation of the predicted transaction information of the at least one inventory resource on the other transaction channels.
14. The method of any one of claims 1-7, wherein the page is a home page of a shopping application, a shopping cart page, a group page, or a user details page; the at least one inventory resource is a commodity.
15. An information acquisition method, comprising:
responding to page request operation, and sending a page request to server equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating the page request operation;
receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by the target user;
the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
16. A server-side device, comprising: a memory and a processor;
the memory for storing computer programs or instructions;
the processor, coupled with the memory, to execute the computer program or instructions to:
receiving a page request sent by terminal equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating a page request operation;
acquiring the shortage information and the current inventory information of at least one inventory resource which can be traded on line, and predicting the preference information of the target user on the at least one inventory resource;
selecting a target inventory resource from the at least one inventory resource according to the shortage information of the at least one inventory resource, the current inventory information and the preference information of the target user on the at least one inventory resource;
and sending the information of the target inventory resource to the terminal equipment so that the terminal equipment can display the information of the target inventory resource on a page requested by the target user.
17. A terminal device, comprising: a memory, a processor, and a display;
the memory for storing computer programs or instructions;
the processor, coupled with the memory, to execute the computer program or instructions to:
responding to page request operation, and sending a page request to server equipment, wherein the page request comprises a user identifier, and the user identifier is used for identifying a target user initiating the page request operation;
receiving information of the target inventory resources returned by the server equipment, and displaying the information of the target inventory resources on a page requested by the target user;
the display is used for displaying the page requested by the target user;
the target inventory resource is selected by the server-side equipment according to the shortage information of at least one inventory resource which can be traded online, the current inventory information and the preference information of the target user on the at least one inventory resource.
18. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 15.
CN202110169012.7A 2021-02-07 2021-02-07 Information recommendation and acquisition method, equipment and storage medium Pending CN113298610A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110169012.7A CN113298610A (en) 2021-02-07 2021-02-07 Information recommendation and acquisition method, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110169012.7A CN113298610A (en) 2021-02-07 2021-02-07 Information recommendation and acquisition method, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN113298610A true CN113298610A (en) 2021-08-24

Family

ID=77318907

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110169012.7A Pending CN113298610A (en) 2021-02-07 2021-02-07 Information recommendation and acquisition method, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN113298610A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113643099A (en) * 2021-08-30 2021-11-12 北京沃东天骏信息技术有限公司 Commodity data processing method, commodity data processing device, commodity data processing apparatus, storage medium, and program product
CN115909591A (en) * 2023-01-06 2023-04-04 北京国旺盛源智能终端科技有限公司 Goods selling management method, system and equipment based on point exchange cabinet

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113643099A (en) * 2021-08-30 2021-11-12 北京沃东天骏信息技术有限公司 Commodity data processing method, commodity data processing device, commodity data processing apparatus, storage medium, and program product
CN115909591A (en) * 2023-01-06 2023-04-04 北京国旺盛源智能终端科技有限公司 Goods selling management method, system and equipment based on point exchange cabinet
CN115909591B (en) * 2023-01-06 2023-05-05 北京国旺盛源智能终端科技有限公司 Goods selling management method, system and equipment based on point exchange cabinet

Similar Documents

Publication Publication Date Title
US10650434B2 (en) Predictive shopping
US9224167B2 (en) System and method for aiding user in online searching and purchasing of multiple items
US20160191450A1 (en) Recommendations Engine in a Layered Social Media Webpage
US11875380B2 (en) Method for matching influencers to brands and pairing influencer videos to products for intra-video transactions
US20200279191A1 (en) Personalized merchant scoring based on vectorization of merchant and customer data
JP6442535B2 (en) Information processing apparatus, information processing method, and information processing program
CN110348921B (en) Method and device for selecting store articles
US10185927B1 (en) Techniques for optimizing an inventory selection
US20130254025A1 (en) Item ranking modeling for internet marketing display advertising
US20120066055A1 (en) Generating a user interface based on predicted revenue yield
CN110766509A (en) Service order processing and takeout order recommending method and device
CN113298610A (en) Information recommendation and acquisition method, equipment and storage medium
CN114168843A (en) Search word recommendation method, device and storage medium
CN112215448A (en) Method and device for distributing customer service
CN105488163A (en) Information pushing method and apparatus
CN112053190A (en) Stall data management method and device, storage medium and equipment
US20240078523A1 (en) Systems and methods for e-commerce checkout with delay loading of checkout options
US10096045B2 (en) Tying objective ratings to online items
WO2016022436A1 (en) Facilitating intelligent and competitive consumer-favored online shopping experience at computing devices
CN117172866A (en) Order processing method, device, equipment and storage medium
CN115760296A (en) Page data processing and browsing method, terminal device and storage medium
CN107632989B (en) Method and device for selecting commodity objects, determining models and determining use heat
KR102415685B1 (en) Method for providing merchandise sales service using real-time messaging
US20130254019A1 (en) User level incremental revenue and conversion prediction for internet marketing display advertising
US10185980B1 (en) Efficiently computing a feature based on a plurality of variables

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination