WO2021175137A1 - 物品推荐系统、物品推荐的方法、计算机系统和介质 - Google Patents

物品推荐系统、物品推荐的方法、计算机系统和介质 Download PDF

Info

Publication number
WO2021175137A1
WO2021175137A1 PCT/CN2021/077669 CN2021077669W WO2021175137A1 WO 2021175137 A1 WO2021175137 A1 WO 2021175137A1 CN 2021077669 W CN2021077669 W CN 2021077669W WO 2021175137 A1 WO2021175137 A1 WO 2021175137A1
Authority
WO
WIPO (PCT)
Prior art keywords
brand
item
price
items
user
Prior art date
Application number
PCT/CN2021/077669
Other languages
English (en)
French (fr)
Inventor
张伟
尚鑫
祝光明
杨帆
司小婷
刘洪广
王伟然
兰江
黄义军
姜洪凯
钱雪娣
Original Assignee
北京沃东天骏信息技术有限公司
北京京东世纪贸易有限公司
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 北京沃东天骏信息技术有限公司, 北京京东世纪贸易有限公司 filed Critical 北京沃东天骏信息技术有限公司
Priority to US17/904,642 priority Critical patent/US20230099386A1/en
Priority to EP21764540.7A priority patent/EP4116912A4/en
Publication of WO2021175137A1 publication Critical patent/WO2021175137A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data
    • G06Q30/0204Market segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0283Price estimation or determination
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0623Item investigation
    • G06Q30/0625Directed, with specific intent or strategy
    • G06Q30/0629Directed, with specific intent or strategy for generating comparisons

Definitions

  • the present disclosure relates to the field of Internet technology, and more specifically, to an item recommendation system, a method for item recommendation through the item recommendation system, a computer system, and a computer-readable storage medium.
  • the present disclosure provides an item recommendation system, a method for item recommendation through the item recommendation system, a computer system, and a computer-readable storage medium.
  • An aspect of the present disclosure provides an item recommendation system, including: an item expansion module configured to respond to user input and expand the content input by the user to generate a collection of items that are of interest to the user, wherein the item The collection includes one or more items; a price radar module configured to monitor the preferential information of the items in the item collection; and a price monitoring module configured to calculate the actual price of the item based on the preferential information of the items monitored by the price radar module , And maintain the price change records of the items in the item set, and determine whether to push prompt information to the user based on the calculated actual prices of the items and the price change records.
  • the item expansion module includes: a brand similarity calculation module configured to respond to a user-inputted item of interest and expand the item of interest; and a category association calculation module configured to respond to a user-input category, Analyze the category, dig out the potential transaction scenarios that may exist in the category, and determine the items that may be involved in the potential transaction scenarios; wherein the items obtained by the expansion of the brand similarity calculation module are determined by the category association calculation module Of items make up the collection of items.
  • the brand similarity calculation module includes a brand map building unit configured to calculate a brand distance between any two brands, wherein the brand distance between any two brands is limited to The ratio of the number of the same categories involved in the two brands to the number of all categories involved in the two brands; compare the brand distance between any two brands with a first preset threshold, and The two brands whose brand distance is greater than the first preset threshold are determined as related brands; the association relationship between the two brands belonging to the related brands is constructed in the brand map; and the brand map is displayed visually.
  • the brand similarity calculation module further includes a brand grade division unit configured to: calculate an average price spillover ratio of each brand, wherein the average price spillover ratio of each brand is defined as follows Formula calculation:
  • the brand similarity calculation module further includes a brand connection determination unit configured to: multiply the final probabilities of any two brands to calculate the probability of association between any two brands; The probability of association between any two brands is compared with a second preset threshold to determine whether any two brands are associated; the probability of association between any two brands is greater than or equal to the second Two brands with a preset threshold are determined to be associated brands; and an association relationship is established for the brands determined to be associated brands in the brand map.
  • a brand connection determination unit configured to: multiply the final probabilities of any two brands to calculate the probability of association between any two brands; The probability of association between any two brands is compared with a second preset threshold to determine whether any two brands are associated; the probability of association between any two brands is greater than or equal to the second Two brands with a preset threshold are determined to be associated brands; and an association relationship is established for the brands determined to be associated brands in the brand map.
  • the category association calculation module includes: a category scene mining unit configured to construct a joint purchase category association map by analyzing historical orders within a preset time range, and mine a category based on the joint purchase category association map Or multiple consumption scenarios; and an item scenario query unit configured to query the common purchase category association graph to obtain categories associated with the content input by the user according to the content input by the user.
  • the price monitoring module determining whether to push prompt information to the user according to the calculated actual price of the item and the price change record includes: comparing the calculated actual price of the item with the price change Compare the historical prices in the records; and when the calculated actual price of the item is less than the historical price in the price change record, push prompt information to the user, or notify the price radar module to Users push preferential information.
  • the item recommendation system includes an item expansion module, a price radar module, and a price monitoring module.
  • the method includes: the item expansion module responds to a user The input content expands the content input by the user to generate a collection of items of interest to the user, wherein the collection of items includes one or more items; the price radar module monitors the status of the items in the collection of items Preferential information; and the price monitoring module calculates the actual price of the item based on the preferential information of the item monitored by the price radar module, and maintains the price change record of the item in the item set, and based on the calculated actual price of the item and the price The price change record determines whether to push prompt information to the user.
  • Another aspect of the present disclosure provides a computer system, including: one or more processors; a memory for storing one or more programs, wherein, when the one or more programs are used by the one or more When the processor is executed, the one or more processors are caused to implement the method as described above.
  • Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above-mentioned method when executed.
  • Another aspect of the present disclosure provides a computer program, which includes computer-executable instructions, which are used to implement the method as described above when executed.
  • Most of the items under the first category of items do not need to manually monitor the preferential information, so it at least partially overcomes the technical problem that users need to manually collect most of the items under the category of interest in the related technology, and then achieves the goal of providing consumers Real-time monitoring of item price changes for the items of interest and expansion items, and notifications when the price is the most favorable, so as to help consumers buy more favorable items in a more convenient manner.
  • FIG. 1 schematically shows an exemplary system architecture to which an item recommendation system and an item recommendation method can be applied according to an embodiment of the present disclosure
  • FIG. 3 schematically shows a flowchart of a method for recommending an item through an item recommendation system according to an embodiment of the present disclosure
  • Fig. 6 schematically shows a block diagram of a category association calculation module according to an embodiment of the present disclosure
  • FIG. 7 schematically shows a schematic diagram of a joint purchase category association graph according to an embodiment of the present disclosure.
  • the server 105 may be a server that provides various services, for example, a back-end management server (just an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103.
  • the background management server can analyze and process the received user request and other data, and feed back the processing result (for example, webpage, information, or data obtained or generated according to the user request) to the terminal device.
  • the item recommendation system 200 includes an item expansion module 210, a price radar module 220 and a price monitoring module 230.
  • the method includes operations S301 to S303.
  • the price radar module monitors the discount information of the items in the item set.
  • the price monitoring module can also periodically call the price radar module to calculate the price of the item and store the price.
  • the price monitoring module can maintain a record of changes in the historical price of an item, and define the verification period of the historical price. If the real-time price of the item is lower than the historical lowest price (or lower than the price when consumers pay attention to the item), the discount for this item can be pushed to the mobile terminal.
  • the item expansion module may include a brand similarity calculation module and a category association calculation module.
  • the brand similarity calculation module is configured to respond to the items of interest input by the user and expand the items of interest.
  • the item of interest input by the user is "Dabao Moisturizing Water", which can be expanded to obtain “Dabao Facial Cleanser” and so on.
  • the category input by the user is diaper
  • the category is analyzed
  • the potential transaction scenario may be baby products
  • the items that may be involved include milk bottles and milk powder.
  • Fig. 4 schematically shows a block diagram of a brand similarity calculation module according to an embodiment of the present disclosure.
  • the brand map building unit 401 is configured to calculate the brand distance between any two brands, where the brand distance between any two brands is limited to the individuals of the same category involved in the two brands. The ratio of the number and the number of all categories involved in the two brands; compare the brand distance between any two brands with the first preset threshold, and determine the two brands whose brand distance is greater than the first preset threshold It is an associated brand; the association relationship between two brands belonging to the associated brand is constructed in the brand map; the brand map is displayed visually.
  • the first preset threshold can be determined according to the number of items that need to be expanded. For example, when the number of items that need to be expanded is large, the first preset threshold can be set to be relatively small; when the items need to be expanded When the number is small, the first preset threshold can be set relatively large.
  • the present disclosure applies a brand graph to represent the relationship between brands.
  • a vertex on the brand graph can represent one brand, and an edge can represent that two brands are related.
  • the present disclosure there are three grades of high, medium, and low according to the price of the brand of the item.
  • the expanded items are more similar.
  • the present disclosure can construct a brand map of high, medium and low grades.
  • the present disclosure uses fuzzy mathematics to classify brands into three grades: high, medium, and low. Specifically, the average price spillover ratio of the brand is classified by fuzzy mathematics.
  • the brand grade dividing unit 402 is configured to calculate an average price overflow ratio of each brand, where the average price overflow ratio of each brand is limited to be calculated according to the following formula:
  • the probabilities of high, medium and low grades corresponding to a certain brand can be calculated, and then combined with the weights, the final probability P_defuzzification of the brand can be obtained.
  • the knowledge graph of which gear the given brand x belongs to is determined according to the finally calculated level(x).
  • the embodiment of the present disclosure innovatively constructs a brand knowledge graph, specifically, constructs brand graphs of different grades, and innovatively uses graph theory and fuzzy mathematics to help consumers lock items they will like and form up-sales.
  • the present disclosure can also create a category knowledge graph, so as to help consumers find items that meet their consumption scenarios and improve their consumption experience.
  • This disclosure is based on the price spillover ratio of brands and the similarity distance between brands, and uses fuzzy mathematics to classify brands to determine whether there is a connection between brands.
  • the brand map construction unit 401 is further configured to construct a brand map of different grade brands in the brand map according to the grade to which each brand belongs after the brand grade division unit determines the grade to which each brand belongs.
  • Fig. 5 schematically shows a schematic diagram of a brand map according to an embodiment of the present disclosure.
  • the brand of an item has three levels of high, medium, and low according to the price. Therefore, as shown in Figure 5, the overall brand relationship of items can be expressed by three brand knowledge maps of high, medium and low.
  • the brands Lancome and Rolex are not related, because the two brand items have no intersection.
  • brands and brands can be connected, such as Lancome and L'Oreal, because the two brand items have a relatively large intersection. All brands can be constructed into three knowledge graphs as shown in Figure 5. Each circle indicates a brand, the solid line indicates that the brands of the same grade are related, and the dotted line indicates the connection of different grades of brands.
  • the brand similarity calculation module 400 further includes a brand connection determination unit 403, configured to: multiply the final probabilities of any two brands to calculate the probability of association between any two brands; The probability of association between brands is compared with the second preset threshold to determine whether any two brands are associated; two brands with the probability of association between any two brands greater than or equal to the second preset threshold are determined to be associated Brand; establish an association relationship for the brand identified as an associated brand in the brand map.
  • a brand connection determination unit 403 configured to: multiply the final probabilities of any two brands to calculate the probability of association between any two brands; The probability of association between brands is compared with the second preset threshold to determine whether any two brands are associated; two brands with the probability of association between any two brands greater than or equal to the second preset threshold are determined to be associated Brand; establish an association relationship for the brand identified as an associated brand in the brand map.
  • any two brands may be brands of the same grade or brands of different grades.
  • the brand SKII and the brand Lancome are related brands.
  • connections can be used to characterize the relationship between related brands.
  • the threshold h is a set parameter.
  • brand_distance (x, y) is calculated by formula (1).
  • the similar item selection unit 404 is configured to expand the content input by the user based on the brand graph, and expand items from the brands associated with the content input by the user.
  • the brand map constructed as described above can be applied to expand items.
  • mining similar items can include the following three methods.
  • Method 1 In the brand x to which the item belongs, query other related items under the same three-level classification.
  • Method 2 From the same brand of brand x to which the item belongs, select items of the same item three-level classification of the related brand. For example, assuming that the brand of an item of interest entered by the consumer belongs to a mid-range brand, then in the mid-range brand map, select items of the same item three-level classification of the same item of the brand that is related (using formula 8 to calculate the related brand).
  • Method 3 In the more expensive grade of the item's brand x (for example, in the high-end brand map), select items in the same item three-level classification of the related brand to guide users to upgrade their consumption. For example, suppose that the brand of an item of interest entered by the consumer belongs to a mid-range brand, then in the high-end brand map, select the related (using formula 6 and formula 8) brand items of the same item three-level classification.
  • Fig. 6 schematically shows a block diagram of a category association calculation module according to an embodiment of the present disclosure.
  • the category association calculation module can search for similar collections of items in the three-level classification (or the four-level classification of items) input by the user.
  • the category correlation calculation module can mine the potential shopping scenes of the current items, and then consider the user's additional purchases, browsing and other historical purchase behaviors in this scenario, and combine the corresponding promotions and coupons to automatically select the item combination.
  • the category correlation calculation module The task is achieved by serially calling the following category scene mining unit and item scene query unit.
  • the category association calculation module 600 includes a category scene mining unit 601 and an item scene query unit 602.
  • the category scene mining unit 601 is configured to construct a joint purchase category association map by analyzing historical orders within a preset time range, and mine one or more consumption scenarios based on the joint purchase category association map.
  • a consumption scene can be represented by a combination of several different items. Since each item corresponds to a three-level classification of items, a consumption scene can be represented by a combination of several three-level classifications of items corresponding to the back of the item. If the vertices of the graph represent a three-level classification of items, and the edges of the graph represent that two three-level classifications of items belong to the same consumption scene, then graphics can be used to express the consumption scene.
  • FIG. 7 schematically shows a schematic diagram of a joint purchase category association map according to an embodiment of the present disclosure.
  • the common purchase category association graph (also referred to as the consumption scene knowledge graph), the vertices C1-C7 in the graph represent the three-level classification of different items.
  • C1-C5 represents a consumption scenario, because several nodes C1, C2, C3, C4 and C5 form a complete subgraph (any two vertices are connected by an edge).
  • a complete subgraph means that the five three-level categories C1-C5 are often purchased together, which means that these five three-level categories form a shopping scene. For example, milk bottles, milk powder, and diapers are potential maternal and child shopping scenes.
  • C6-C7 represents a relatively independent and small consumption scene.
  • the category scene mining unit 601 constructs a three-level classification common purchase association map by analyzing historical orders within a certain period of time, and the connection between the three-level classifications represents their co-purchase relationship and consumption scenarios in the historical orders.
  • the association graph is constructed by analyzing the joint purchase of historical orders, and the complete subgraph in the graph is searched through the intelligent optimization algorithm, so that potential shopping scenarios can be explored.
  • the item scene query unit 602 is configured to query the common purchase category association graph to obtain categories related to the content input by the user according to the content input by the user.
  • the common purchase category association graph constructed by the category scene mining unit, for an item category M of interest input by the consumer, mining similar items mainly includes the following steps.
  • any number of the modules, sub-modules, units, and sub-units, or at least part of the functions of any number of them may be implemented in one module. Any one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be split into multiple modules for implementation.
  • any one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), System-on-chip, system-on-substrate, system-on-package, application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware in any other reasonable way that integrates or encapsulates the circuit, or by software, hardware, and firmware. Any one of these implementations or an appropriate combination of any of them can be implemented.
  • one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be at least partially implemented as a computer program module, and when the computer program module is executed, the corresponding function may be performed.
  • any of the item expansion module 210, the price radar module 220, and the price monitoring module 230 can be combined into one module/unit/subunit for implementation, or any one of the modules/units/subunits can be split into Multiple modules/units/subunits. Or, at least part of the functions of one or more modules/units/subunits of these modules/units/subunits can be combined with at least part of the functions of other modules/units/subunits and integrated in one module/unit/subunit In the realization.
  • At least one of the item expansion module 210, the price radar module 220, and the price monitoring module 230 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array ( PLA), system-on-chip, system-on-substrate, system-on-package, application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way to integrate or package the circuit, or by software or hardware And any one of the three implementations of firmware or an appropriate combination of any of them.
  • at least one of the item expansion module 210, the price radar module 220, and the price monitoring module 230 may be at least partially implemented as a computer program module, and when the computer program module is run, it may perform a corresponding function.
  • the present disclosure also provides a computer system, including: one or more processors; a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors At this time, the one or more processors are caused to implement the method for recommending items.
  • FIG. 8 schematically shows a block diagram of a computer system suitable for implementing the above-described method according to an embodiment of the present disclosure.
  • the computer system shown in FIG. 8 is only an example, and should not bring any limitation to the function and scope of use of the embodiments of the present disclosure.
  • a computer system 800 includes a processor 801, which can be loaded into a random access memory (RAM) 803 according to a program stored in a read only memory (ROM) 802 or from a storage part 808 The program executes various appropriate actions and processing.
  • the processor 801 may include, for example, a general-purpose microprocessor (for example, a CPU), an instruction set processor and/or a related chipset and/or a special purpose microprocessor (for example, an application specific integrated circuit (ASIC)), and so on.
  • the processor 801 may also include on-board memory for caching purposes.
  • the processor 801 may include a single processing unit or multiple processing units for performing different actions of a method flow according to an embodiment of the present disclosure.
  • the processor 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804.
  • the processor 801 executes various operations of the method flow according to the embodiments of the present disclosure by executing programs in the ROM 802 and/or RAM 803. It should be noted that the program may also be stored in one or more memories other than ROM 802 and RAM 803.
  • the processor 801 may also execute various operations of the method flow according to the embodiments of the present disclosure by executing programs stored in the one or more memories.
  • the system 800 may further include an input/output (I/O) interface 805, and the input/output (I/O) interface 805 is also connected to the bus 804.
  • the system 800 may also include one or more of the following components connected to the I/O interface 805: an input part 806 including a keyboard, a mouse, etc.; including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker
  • the communication section 809 performs communication processing via a network such as the Internet.
  • the driver 810 is also connected to the I/O interface 805 as needed.
  • a removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that the computer program read from it is installed into the storage section 808 as needed.
  • the method flow according to the embodiment of the present disclosure may be implemented as a computer software program.
  • an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains program code for executing the method shown in the flowchart.
  • the computer program may be downloaded and installed from the network through the communication section 809, and/or installed from the removable medium 811.
  • the computer program executes the above-mentioned functions defined in the system of the embodiment of the present disclosure.
  • the systems, devices, devices, modules, units, etc. described above may be implemented by computer program modules.
  • the present disclosure also provides a computer-readable storage medium.
  • the computer-readable storage medium may be included in the device/device/system described in the above embodiment; or it may exist alone without being assembled into the device/ In the device/system.
  • the aforementioned computer-readable storage medium carries one or more programs, and when the aforementioned one or more programs are executed, the method according to the embodiments of the present disclosure is implemented.
  • the computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • it can include but not limited to: portable computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD- ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • a computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.
  • the computer-readable storage medium may include one or more memories other than the ROM 802 and/or RAM 803 and/or ROM 802 and RAM 803 described above.
  • each block in the flowchart or block diagram may represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more for realizing the specified logic function.
  • Executable instructions may also occur in a different order from the order marked in the drawings. For example, two blocks shown one after another can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved.
  • each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by It is realized by a combination of dedicated hardware and computer instructions.
  • Those skilled in the art can understand that the various embodiments of the present disclosure and/or the features described in the claims can be combined and/or combined in various ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the various embodiments of the present disclosure and/or the features described in the claims can be combined and/or combined in various ways. All these combinations and/or combinations fall within the scope of the present disclosure.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种物品推荐系统(200),通过物品推荐系统进行物品推荐的方法、一种计算机系统和一种计算机可读存储介质,系统(200)包括:物品扩展模块(210),配置为响应用户输入的内容,对用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中物品集合中包括一个或多个物品;价格雷达模块(220),配置为监控物品集合中的物品的优惠信息;以及价格监控模块(230),配置为基于价格雷达模块(230)监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。

Description

物品推荐系统、物品推荐的方法、计算机系统和介质
本公开要求在2020年03月02日提交中国专利局、申请号为202010137487.3、发明名称为“物品推荐系统、物品推荐的方法、计算机系统和介质”的中国专利申请的优先权,其全部内容通过引用结合在本公开中。
技术领域
本公开涉及互联网技术领域,更具体地,涉及一种物品推荐系统、一种通过物品推荐系统进行物品推荐的方法、一种计算机系统和一种计算机可读存储介质。
背景技术
随着电子商务技术的快速发展,消费者越来越习惯于在网上进行交易。每一个消费者都希望能够最方便的买到最优惠的物品。目前,线下超市和线上电商平台经营了海量物品,海量物品也带来海量促销优惠活动。由于促销优惠活动可能每天会变动,各种促销种类和规则相互交叠和限制,物品价格每天都在变化。因此在实际生活中,尤其是对于价格比较贵的物品,消费者常常在一段时期内每天检查一下自己喜欢的物品有没有降价,看看降价的力度满意不满意,然后再购买。
随着技术的发展,在相关技术中,消费者可以收藏自己喜欢或者有需求的物品,交易平台可以及时提醒用户收藏的物品的优惠活动。
但是,在实现本公开构思的过程中,发明人发现相关技术中至少存在如下问题:目前物品的种类和价格、促销种类和优惠券繁多,规则复杂,如果用户需要购买某一类物品,不仅需要手动收藏该类物品下的大部分物品,而且需要关注该类物品下的大部分物品的优惠信息,导致用户交易体验差。
发明内容
有鉴于此,本公开提供了一种物品推荐系统、一种通过物品推荐系统进行物品推荐的方法、一种计算机系统和一种计算机可读存储介质。
本公开的一个方面提供了一种物品推荐系统,包括:物品扩展模块,配置为响应用户输入的内容,对所述用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中所述物品集合中包括一个或多个物品;价格雷达模块,配置为监控所述物品集合中的物品的优惠信息;以及价格监控模块,配置为基于所述价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护所述物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和所述价 格变动记录确定是否向所述用户推送提示信息。
根据本公开的实施例,所述物品扩展模块包括:品牌相似计算模块,配置为响应用户输入的关注物品,对所述关注物品进行扩展;以及品类关联计算模块,配置为响应用户输入的品类,对所述品类进行分析,挖掘所述品类可能存在的潜在交易场景,并确定所述潜在交易场景可能涉及的物品;其中,所述品牌相似计算模块扩展得到的物品和所述品类关联计算模块确定的物品组成所述物品集合。
根据本公开的实施例,所述品牌相似计算模块包括品牌图谱构建单元,配置为:计算任意两个品牌之间的品牌距离,其中,所述任意两个品牌之间的品牌距离被限定为所述两个品牌所涉及相同分类的个数和所述两个品牌所涉及的所有分类的个数的比值;将所述任意两个品牌之间的品牌距离与第一预设阈值进行比较,并将品牌距离大于所述第一预设阈值的两个品牌确定为关联品牌;在品牌图谱中构建属于关联品牌的两个品牌之间的关联关系;以及可视化展示所述品牌图谱。
根据本公开的实施例,所述品牌相似计算模块还包括品牌档次划分单元,配置为:计算每个品牌的平均价格溢出比,其中,所述每个品牌的平均价格溢出比被限定为按照如下公式计算:
Figure PCTCN2021077669-appb-000001
通过模糊数学确定所述每个品牌的平均价格溢出比属于不同档次的概率;根据所述每个品牌的平均价格溢出比属于不同档次的概率和为每个档次分配的权重计算所述每个品牌的最终概率;以及根据所述每个品牌的最终概率确定所述每个品牌所属的档次。
根据本公开的实施例,所述品牌图谱构建单元,还配置为:在所述品牌档次划分单元确定所述每个品牌所属的档次后,在所述品牌图谱中根据所述每个品牌所属的档次构建关于不同档次品牌的品牌图谱。
根据本公开的实施例,所述品牌相似计算模块还包括品牌连接判定单元,配置为:将任意两个品牌的最终概率相乘,计算得到所述任意两个品牌之间关联的概率;将所述任意两个品牌之间关联的概率与第二预设阈值进行比较,确定所述任意两个品牌之间是否关联;将所述任意两个品牌之间关联的概率大于或等于所述第二预设阈值的两个品牌确定为关联品牌;以及在所述品牌图谱中为所述确定为关联品牌的品牌建立关联关系。
根据本公开的实施例,所述品牌相似计算模块还包括相似物品选择单元,配置为:基于所述品牌图谱对所述用户输入的内容进行扩展,从与所述用户输入的内容相关联的品牌中扩展物品。
根据本公开的实施例,所述品类关联计算模块包括:品类场景挖掘单元,配置为通过分析预设时长范围内的历史订单,构建共同购买品类关联图谱,根据所述共同购买品类关联图谱挖掘一个或多个消费场景;以及物品场景查询单元,配置为根据所述用户输入的内容从所述共同购买品类关联图谱中查询获得与所述用户输入的内容关联的品类。
根据本公开的实施例,所述价格监控模块根据计算得到的物品实际价格和所述价格变动记录确定是否向所述用户推送提示信息包括:将所述计算得到的物品实际价格与所述价格变动记录中的历史价格进行比较;以及在所述计算得到的物品实际价格小于所述价格变动记录中的历史价格的情况下,向所述用户推送提示信息,或者通知所述价格雷达模块向所述用户推送优惠信息。
本公开的另一个方面提供了一种通过物品推荐系统进行物品推荐的方法,所述物品推荐系统包括物品扩展模块、价格雷达模块和价格监控模块,所述方法包括:所述物品扩展模块响应用户输入的内容,对所述用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中所述物品集合中包括一个或多个物品;所述价格雷达模块监控所述物品集合中的物品的优惠信息;以及所述价格监控模块基于所述价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护所述物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和所述价格变动记录确定是否向所述用户推送提示信息。
本公开的另一个方面提供了一种计算机系统,包括:一个或多个处理器;存储器,用于存储一个或多个程序,其中,当所述一个或多个程序被所述一个或多个处理器执行时,使得所述一个或多个处理器实现如上所述的方法。
本公开的另一方面提供了一种计算机可读存储介质,存储有计算机可执行指令,所述指令在被执行时用于实现如上所述的方法。
本公开的另一方面提供了一种计算机程序,所述计算机程序包括计算机可执行指令,所述指令在被执行时用于实现如上所述的方法。
根据本公开的实施例,物品推荐系统可以基于用户输入的内容进行扩展,以生成用户感兴趣的物品集合,通过价格雷达模块对物品集合中的物品的优惠信息进行监控,通过价格监控模块基于价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。由于物品推荐系统可以自动扩展用户感兴趣的物品,并对扩展后的物品进行监控,使得即使目前物品的种类和价格、促销种类和优惠券繁多,规则复杂,用户也无需手动一一收藏感兴趣的一类物品下的大部分物品,无需人工监控优惠信息,所以至少部分地克服了相关技术中用户需要手动收藏感兴趣的一类物品下的大部分物品的技术问题,进而达到了为消 费者对其所关注的物品和扩展物品进行实时物品价格变动的监控,并且当价格最优惠时给以通知,从而帮助消费者以较为方便的方式买到较为优惠的物品的技术效果。
附图说明
通过以下参照附图对本公开实施例的描述,本公开的上述以及其他目的、特征和优点将更为清楚,在附图中:
图1示意性示出了根据本公开实施例的可以应用物品推荐系统及物品推荐方法的示例性系统架构;
图2示意性示出了根据本公开实施例的物品推荐系统的框图;
图3示意性示出了根据本公开实施例的通过物品推荐系统进行物品推荐的方法的流程图;
图4示意性示出了根据本公开实施例的品牌相似计算模块的框图;
图5示意性示出了根据本公开实施例的品牌图谱的示意图;
图6示意性示出了根据本公开实施例的品类关联计算模块的框图;
图7示意性示出了根据本公开实施例的共同购买品类关联图谱的示意图;以及
图8示意性示出了根据本公开实施例的适于实现推荐方法的计算机系统的框图。
具体实施方式
以下,将参照附图来描述本公开的实施例。但是应该理解,这些描述只是示例性的,而并非要限制本公开的范围。在下面的详细描述中,为便于解释,阐述了许多具体的细节以提供对本公开实施例的全面理解。然而,明显地,一个或多个实施例在没有这些具体细节的情况下也可以被实施。此外,在以下说明中,省略了对公知结构和技术的描述,以避免不必要地混淆本公开的概念。
在此使用的术语仅仅是为了描述具体实施例,而并非意在限制本公开。在此使用的术语“包括”、“包含”等表明了所述特征、步骤、操作和/或部件的存在,但是并不排除存在或添加一个或多个其他特征、步骤、操作或部件。
在此使用的所有术语(包括技术和科学术语)具有本领域技术人员通常所理解的含义,除非另外定义。应注意,这里使用的术语应解释为具有与本说明书的上下文相一致的含义,而不应以理想化或过于刻板的方式来解释。
在使用类似于“A、B和C等中至少一个”这样的表述的情况下,一般来说应该按照本领域技术人员通常理解该表述的含义来予以解释(例如,“具有A、B和C中至少一个的系统”应包括但不限于单独具有A、单独具有B、单独具有C、具有A和B、具有A和C、具有B 和C、和/或具有A、B、C的系统等)。在使用类似于“A、B或C等中至少一个”这样的表述的情况下,一般来说应该按照本领域技术人员通常理解该表述的含义来予以解释(例如,“具有A、B或C中至少一个的系统”应包括但不限于单独具有A、单独具有B、单独具有C、具有A和B、具有A和C、具有B和C、和/或具有A、B、C的系统等)。
在相关技术中,零售企业需要一种自动智能技术来帮助消费者对其所关注的物品(如海尔冰箱)、挖掘其所关注物品的其他近似品(如海尔冰柜)、替代品(如三星冰箱)和关联物品(冷冻食品)等进行实时物品价格变动的监控。一旦物品的价格够优惠,立刻及时的通知消费者,从而帮助消费者方便的买到最优惠的物品。
如果能够实现这一技术,具有重要意义,可以对消费者给以真正的关怀,提高用户体验,保护消费者权益;提高零售企业提供营收,实现真正的精细化经营。
然而,目前,广大零售企业没有可用的技术来实现这一商业需求,现有的技术一般依赖人工手工设置。具体技术难度有:需要对海量物品建立物品关联机制,从而帮助消费者以最优惠的价格买到所钟爱的物品或者最关联物品。海量物品的价格变动太多,验证物品的价格是否最优相对较难,需要发展一整套全新的自动化智能流程。
基于此,本公开的实施例提供了一种物品推荐系统,包括:物品扩展模块,配置为响应用户输入的内容,对用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中物品集合中包括一个或多个物品;价格雷达模块,配置为监控物品集合中的物品的优惠信息;以及价格监控模块,配置为基于价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。
图1示意性示出了根据本公开实施例的可以应用物品推荐系统及物品推荐方法的示例性系统架构。需要注意的是,图1所示仅为可以应用本公开实施例的系统架构的示例,以帮助本领域技术人员理解本公开的技术内容,但并不意味着本公开实施例不可以用于其他设备、系统、环境或场景。
如图1所示,根据该实施例的系统架构100可以包括终端设备101、102、103,网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型,例如有线和/或无线通信链路等等。
用户可以使用终端设备101、102、103通过网络104与服务器105交互,以接收或发送消息等。终端设备101、102、103上可以安装有各种通讯客户端应用,例如购物类应用、网页浏览器应用、搜索类应用、即时通信工具、邮箱客户端和/或社交平台软件等(仅为示例)。
终端设备101、102、103可以是具有显示屏并且支持网页浏览的各种电子设备,包括但 不限于智能手机、平板电脑、膝上型便携计算机和台式计算机等等。
服务器105可以是提供各种服务的服务器,例如对用户利用终端设备101、102、103所浏览的网站提供支持的后台管理服务器(仅为示例)。后台管理服务器可以对接收到的用户请求等数据进行分析等处理,并将处理结果(例如根据用户请求获取或生成的网页、信息、或数据等)反馈给终端设备。
需要说明的是,本公开实施例所提供的物品推荐方法一般可以由服务器105执行。相应地,本公开实施例所提供的物品推荐系统一般可以设置于服务器105中。本公开实施例所提供的物品推荐方法也可以由不同于服务器105且能够与终端设备101、102、103和/或服务器105通信的服务器或服务器集群执行。相应地,本公开实施例所提供的物品推荐系统也可以设置于不同于服务器105且能够与终端设备101、102、103和/或服务器105通信的服务器或服务器集群中。或者,本公开实施例所提供的物品推荐方法也可以由终端设备101、102、或103执行,或者也可以由不同于终端设备101、102、或103的其他终端设备执行。相应地,本公开实施例所提供的物品推荐系统也可以设置于终端设备101、102、或103中,或设置于不同于终端设备101、102、或103的其他终端设备中。
应该理解,图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要,可以具有任意数目的终端设备、网络和服务器。
图2示意性示出了根据本公开实施例的物品推荐系统的框图。
如图2所示,物品推荐系统200包括物品扩展模块210、价格雷达模块220和价格监控模块230。
物品扩展模块210配置为响应用户输入的内容,对用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中,物品集合中包括一个或多个物品。例如,物品集合中包括物品A,物品B和物品Z。
价格雷达模块220配置为监控物品集合中的物品的优惠信息。
价格监控模块230配置为基于价格雷达模块220监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。
下面参考图3,结合具体实施例对图2所示的系统做进一步说明。
图3示意性示出了根据本公开实施例的通过物品推荐系统进行物品推荐的方法的流程图。
如图3所示,该方法包括操作S301~S303。
在操作S301,物品扩展模块响应用户输入的内容,对用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中物品集合中包括一个或多个物品。
根据本公开的实施例,用户可以输入自己喜欢的物品和/或喜欢的品类。例如,用户可以输入自己喜欢的一个物品和喜欢的一种品类,其中,品类的划分可以根据当前主流商家划分方式进行划分。物品扩展模块可以对用户输入的内容进行扩展,以生成用户感兴趣的物品集合。例如,用户输入“兰蔻化妆水”,可以对该物品单品进行扩展,例如,扩展出“欧莱雅化妆水”,从而进一步挖掘出“兰蔻化妆水”相似的物品集合。需要说明的是,用户输入的内容并不限于输入物品单品和品类,还可以是其他内容,例如,还可以是关于物品的描述,例如,用户输入洗澡用品,物品扩展模块可以扩展出“舒肤佳”、“飘柔”和“大宝”等等。
在操作S302,价格雷达模块监控物品集合中的物品的优惠信息。
根据本公开的实施例,价格雷达模块可以监控单品的价格是否出现低价,或者出现有折扣力度更大的促销和优惠券,价格雷达模块可以及时将物品优惠信息触达到移动端300和价格监控模块。
在操作S303,价格监控模块基于价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。
根据本公开的实施例,例如,将计算得到的物品实际价格与价格变动记录进行比较,发现物品实际价格为历史最低,那么价格监控模块可以向用户使用的移动端300推送提示信息,移动端300上可以展示物品价格信息,提示用户此时物品的价格合适,可以进行交易。
根据本公开的实施例,价格监控模块可以计算物品集合中的物品实时价格,通过对物品促销优惠券的枚举计算,并返回该物品的实时最低价格。
根据本公开的实施例,价格监控模块根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息包括:将计算得到的物品实际价格与价格变动记录中的历史价格进行比较,在计算得到的物品实际价格小于价格变动记录中的历史价格的情况下,向用户推送提示信息,或者通知价格雷达模块向用户推送优惠信息。
根据本公开的实施例,针对用户感兴趣的物品集合中的物品,价格监控模块也可以定时调用价格雷达模块进行物品价格的计算,进行价格存储。价格监控模块可以维护物品历史价格的变动记录,并且定义历史价格的校验时期。如果物品的实时价格低于历史最低价格(或者低于消费者关注该物品时的价格),则可以将这一物品的优惠推送给移动端。通过本公开的实施例,可以确保消费者可以收到其感兴趣物品的较为优惠价格的通知,从而帮助消费者以较优价格购物,提升其消费体验。
根据本公开的实施例,可以为消费者对其所关注的物品和其所关注物品的扩展物品,进行实时物品价格变动的监控,并且当价格最优惠时给以通知,从而帮助消费者最方便的方式 买到最优惠的物品。
本公开不同于其他基于价格让利和消费者个性化的优惠让利,本公开针对消费者关注的物品,在一段时间上进行跟踪,利用物品在时间上价格波动所形成的优惠促进零售业的销售,在营销技术上是一个突破。
根据本公开的实施例,物品推荐系统可以基于用户输入的内容进行扩展,以生成用户感兴趣的物品集合,通过价格雷达模块对物品集合中的物品的优惠信息进行监控,通过价格监控模块基于价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和价格变动记录确定是否向用户推送提示信息。由于物品推荐系统可以自动扩展用户感兴趣的物品,并对扩展后的物品进行监控,使得即使目前物品的种类和价格、促销种类和优惠券繁多,规则复杂,用户也无需手动一一收藏感兴趣的一类物品下的大部分物品,无需人工监控优惠信息,所以至少部分地克服了相关技术中用户需要手动收藏感兴趣的一类物品下的大部分物品的技术问题,进而达到了为消费者对其所关注的物品和扩展物品进行实时物品价格变动的监控,并且当价格最优惠时给以通知,从而帮助消费者以较为方便的方式买到较为优惠的物品的技术效果。
根据本公开的实施例,物品扩展模块可以包括品牌相似计算模块和品类关联计算模块。
品牌相似计算模块配置为响应用户输入的关注物品,对关注物品进行扩展。
根据本公开的实施例,例如,用户输入的关注物品为“大宝保湿水”,可以扩展得到“大宝洗面奶”等等。
品类关联计算模块配置为响应用户输入的品类,对品类进行分析,挖掘品类可能存在的潜在交易场景,并确定潜在交易场景可能涉及的物品。
根据本公开的实施例,例如,用户输入的品类为尿不湿,对该品类进行分析,潜在交易场景可能是婴儿用品,可能涉及的物品包括奶瓶,奶粉。
根据本公开的实施例,品牌相似计算模块扩展得到的物品和品类关联计算模块确定的物品组成物品集合。
图4示意性示出了根据本公开实施例的品牌相似计算模块的框图。
根据本公开的实施例,品牌相似计算模块400可以对输入的物品,通过分析该物品的品牌,查找相似的物品集合。品牌相似计算模块400也可以分析各个品牌的物品价格,和所含物品覆盖的多级分类,使用模糊数学可以将品牌信息至少划分为高档,中档,低档三个档次;对应于三张高档,中档,低档品牌知识图谱,物品推荐系统利用图谱,可以查询相关联的物品集合。
如图4所示,品牌相似计算模块400可以包括品牌图谱构建单元401、品牌档次划分单 元402、品牌连接判定单元403和相似物品选择单元404。
根据本公开的实施例,品牌图谱构建单元401配置为:计算任意两个品牌之间的品牌距离,其中,任意两个品牌之间的品牌距离被限定为该两个品牌所涉及相同分类的个数和两个品牌所涉及的所有分类的个数的比值;将任意两个品牌之间的品牌距离与第一预设阈值进行比较,并将品牌距离大于第一预设阈值的两个品牌确定为关联品牌;在品牌图谱中构建属于关联品牌的两个品牌之间的关联关系;可视化展示品牌图谱。
根据本公开的实施例,以常见的三级分类为例,任意两个品牌之间的品牌距离的计算公式可以参见如下公式:
Figure PCTCN2021077669-appb-000002
当品牌之间的距离超过一定的阈值时,则可以认为两个品牌之间是相关联的。其中,第一预设阈值可以根据需要扩展出的物品数量进行确定,例如,当需要扩展出的物品数量较多时,可以将第一预设阈值设定的相对较小;当需要扩展出的物品数量较少时,可以将第一预设阈值设定的相对较大。
根据本公开的实施例,本公开应用品牌图谱来代表品牌之间的关系,品牌图谱上的顶点可以代表一个品牌,边可以代表两个品牌是相关联的。
根据本公开的实施例,由于物品的品牌根据价格存在高、中和低三个档次。为了使得构建出的品牌图谱具有层次性,扩展出的物品更加相似。本公开可以构建关于高、中和低三个档次的品牌图谱。
根据本公开的实施例,本公开使用模糊数学将品牌划分为高、中、低三个档次。具体而言,通过模糊数学对品牌的平均价格溢出比进行分档。
根据本公开的实施例,品牌档次划分单元402配置为:计算每个品牌的平均价格溢出比,其中,每个品牌的平均价格溢出比被限定为按照如下公式计算:
Figure PCTCN2021077669-appb-000003
通过模糊数学确定每个品牌的平均价格溢出比属于不同档次的概率;根据每个品牌的平均价格溢出比属于不同档次的概率和为每个档次分配的权重计算每个品牌的最终概率;以及根据每个品牌的最终概率确定每个品牌所属的档次。
假设对于给定品牌x,所有同三级分类所属的品牌的价格溢出比均值为m,标准差为d,用w表示平均价格溢出比overFlowRatio(x)的值,可以通过如下模糊数学确定品牌x的档位。
Figure PCTCN2021077669-appb-000004
Figure PCTCN2021077669-appb-000005
Figure PCTCN2021077669-appb-000006
通过上述定义的模糊数学公式(3)~(5),可以计算出某品牌对应的高、中和低档的概率,然后结合权重,即可得出品牌的最终概率P_defuzzification。
P_defuzzification(x)=Weihgt_low*P_low(x)+Weihgt_middle*P_middle(x)+Weihgt_high*P_high(x)   公式(6)
最后,品牌档次的划分方式如下:
Figure PCTCN2021077669-appb-000007
因此,对于给定品牌x,根据最后计算得到的level(x),来确定给定品牌x属于哪个档位的知识图谱。
本公开实施例创新性的构建了品牌知识图谱,具体地,构建了不同档次的品牌图谱,创新利用图论和模糊数学,帮助消费者锁定其会喜欢的物品,并且形成向上销售。同时,本公开还可以创建品类知识图谱,从而帮助消费者找到符合其消费场景的物品,提升消费体验。
本公开基于品牌的价格溢出比,品牌间的相似距离,并通过模糊数学对品牌进行划分档次,对品牌之间是否有连接进行了判定。创新性的将图论和模糊数学引入到了推荐系统中,通过图处理技术和智能优化算法进行相似物品的挖掘,可以实现物品所处的潜在场景的挖掘以及智能化的物品自动选品。
根据本公开的实施例,品牌图谱构建单元401还配置为在品牌档次划分单元确定每个品牌所属的档次后,在品牌图谱中根据每个品牌所属的档次构建关于不同档次品牌的品牌图谱。
图5示意性示出了根据本公开实施例的品牌图谱的示意图。
根据本公开的实施例,物品的品牌根据价格存在高、中和低三个档次。因此,如图5所示,物品的品牌总体关系可以由高、中和低三张品牌知识图谱来表达。在同一档位的品牌知识图谱中,比如高档品牌图中,品牌兰蔻和劳力士是不相联的,因为两个品牌物品分类无交集。不同一档位的品牌知识图谱中,品牌和品牌是可以相连的,比如兰蔻和欧莱雅,因为两个品牌物品分类交集比较大。所有的品牌可以构建成如图5所示的三张知识图谱,每个圆圈示意一个品牌,实线表示同档次的品牌之间是有关联的,虚线表征的是不同档次品牌的连接。
根据本公开的实施例,品牌相似计算模块400还包括品牌连接判定单元403,配置为:将任意两个品牌的最终概率相乘,计算得到任意两个品牌之间关联的概率;将任意两个品牌之间关联的概率与第二预设阈值进行比较,确定任意两个品牌之间是否关联;将任意两个品牌之间关联的概率大于或等于第二预设阈值的两个品牌确定为关联品牌;在品牌图谱中为确定为关联品牌的品牌建立关联关系。
根据本公开的实施例,任意两个品牌可以是同一档次的品牌,也可以是不同档次的品牌。如图5所示,品牌SKII和品牌兰蔻是关联品牌,在品牌图谱中可以利用连线表征关联品牌之间的关联关系。
对于任意品牌x和品牌y,可以由以下公式判定品牌x和品牌y是否是关联(连接的),标记为connectivity(x,y):
Figure PCTCN2021077669-appb-000008
其中,阈值h是设置的参数。当connectivity(x,y)=1时,则认为两个品牌是互相连接的,即连接x和y的边是存在的。否则,则认为是不连接的,即connectivity(x,y)=0。如果两个品牌相邻,则这样连接两个品牌边的最终权重计算如下:
edge(x,y)=connectivity(x,y)*brand_distance(x,y)   公式(9)
其中,brand_distance(x,y)由公式(1)计算所得。
根据本公开的实施例,相似物品选择单元404配置为基于品牌图谱对用户输入的内容进行扩展,从与用户输入的内容相关联的品牌中扩展物品。
根据本公开的实施例,可以应用上述构建的品牌图谱扩展物品,针对消费者所输入的感兴趣的一个物品A,假设物品A的品牌是x,进行相似物品的挖掘可以包含以下三种方式。
方式一:在该物品所属的品牌x中,查询其他相关联的同三级分类下的物品。
方式二:在物品所属品牌x的同档品牌中,选取相关联品牌的同一物品三级分类的物品。比如假设消费者所输入的感兴趣的一个物品的品牌属于中档品牌,那么在中档品牌图谱中, 选取相关联(利用公式8计算得到关联品牌)的品牌的同一物品三级分类的物品。
方式三:在物品所属品牌x的更贵一个档次(比如在高档品牌图谱中),选取相关联的品牌的同一物品三级分类的物品,引导用户进行消费升级。比如假设消费者所输入的感兴趣的一个物品的品牌属于中档品牌,那么在高档品牌图谱中,选取相关联的(利用公式6和公式8)品牌的同一物品三级分类的物品。
最后,可以汇总和输出以上三种方式所返回的所有物品。
图6示意性示出了根据本公开实施例的品类关联计算模块的框图。
品类关联计算模块可以对用户输入的物品的三级分类(或者物品的四级分类)进行查找相似的物品集合。
品类关联计算模块可以挖掘当前物品潜在的购物场景,然后在该场景下考虑用户的加购、浏览等历史购买行为,并结合相应的促销和优惠券,自动进行物品组合的选取,品类关联计算模块通过串行调用以下品类场景挖掘单元和物品场景查询单元实现其任务。
根据本公开的实施例,如图6所示,品类关联计算模块600包括品类场景挖掘单元601和物品场景查询单元602。
品类场景挖掘单元601配置为通过分析预设时长范围内的历史订单,构建共同购买品类关联图谱,根据共同购买品类关联图谱挖掘一个或多个消费场景。
消费者通常为了解决其生活中的一个消费场景而购物。因此一个订单通常包括几个不同的物品。常见的消费场景形成于现实中会经过成千上万的消费者购物行为。一个消费场景可以由几个不同的物品的组合来代表。由于每个物品都对应了一个物品三级分类,因此一个消费场景可以由物品背后所对应的几个物品三级分类的组合来代表。如果用图的顶点代表一个物品三级分类,用图的边代表两个物品三级分类同属于一个消费场景,那么可以用图形来表达消费场景。
图7示意性示出了根据本公开实施例的共同购买品类关联图谱的示意图。
如图7所示的共同购买品类关联图谱(也可以称之为消费场景知识图谱),图中顶点C1-C7代表不同的物品三级分类。C1-C5代表了一个消费场景,因为C1,C2,C3,C4和C5几个节点构成了一张完全子图(任意两个顶点都有一条边相连接)。完全子图意味着C1-C5这5个三级分类经常被一起购买,意味着这5个三级分类组成了一个购物场景。例如,奶瓶,奶粉,尿不湿就是一个潜在的母婴类购物场景。C6-C7代表了一个相对独立小的消费场景。
品类场景挖掘单元601通过分析一定时间内的历史订单,来构建一张三级分类共同购买关联图谱,三级分类之间的连接表征了它们在历史订单中被共同购买的关系和消费场景。
通过对历史订单的共同购买分析构建关联图,通过智能优化算法对图中的完全子图进行 搜索,可以对潜在的购物场景进行挖掘。
物品场景查询单元602配置为根据用户输入的内容从共同购买品类关联图谱中查询获得与用户输入的内容关联的品类。
根据本公开的实施例,通过品类场景挖掘单元构建的共同购买品类关联图谱,针对消费者所输入的感兴趣的一个物品品类M,进行相似物品的挖掘主要包含以下步骤。
在共同购买品类关联图谱中查询并且获得给定的品类M的关联的物品品类。在遍历每一个返回的关联的物品品类,选取相关联的品类的热销和好评物品。最后,输出以上两步所返回的所有物品。
通过本公开的实施例,可以为消费者实现对其所关注的物品和其所关注物品的扩展物品,进行实时物品价格变动的监控,并且给以通知,从而帮助消费者最方便的买到最优惠的物品。
通过创新建立品牌知识图谱(或者建立品类知识图谱),将消费者关注的物品扩展到管理的物品集合,实现物品推荐的多样化和场景化。通过持续监控用户可能感兴趣的物品,当发现出现低价或者有合适力度的促销时,及时触达到用户。为生成的面向场景的优惠组合专辑自动计算使用促销,优惠券后的价格,可以将实时优惠信息反馈给消费者。
根据本公开的实施例的模块、子模块、单元、子单元中的任意多个、或其中任意多个的至少部分功能可以在一个模块中实现。根据本公开实施例的模块、子模块、单元、子单元中的任意一个或多个可以被拆分成多个模块来实现。根据本公开实施例的模块、子模块、单元、子单元中的任意一个或多个可以至少被部分地实现为硬件电路,例如现场可编程门阵列(FPGA)、可编程逻辑阵列(PLA)、片上系统、基板上的系统、封装上的系统、专用集成电路(ASIC),或可以通过对电路进行集成或封装的任何其他的合理方式的硬件或固件来实现,或以软件、硬件以及固件三种实现方式中任意一种或以其中任意几种的适当组合来实现。或者,根据本公开实施例的模块、子模块、单元、子单元中的一个或多个可以至少被部分地实现为计算机程序模块,当该计算机程序模块被运行时,可以执行相应的功能。
例如,物品扩展模块210、价格雷达模块220和价格监控模块230中的任意多个可以合并在一个模块/单元/子单元中实现,或者其中的任意一个模块/单元/子单元可以被拆分成多个模块/单元/子单元。或者,这些模块/单元/子单元中的一个或多个模块/单元/子单元的至少部分功能可以与其他模块/单元/子单元的至少部分功能相结合,并在一个模块/单元/子单元中实现。根据本公开的实施例,物品扩展模块210、价格雷达模块220和价格监控模块230中的至少一个可以至少被部分地实现为硬件电路,例如现场可编程门阵列(FPGA)、可编程逻辑阵列(PLA)、片上系统、基板上的系统、封装上的系统、专用集成电路(ASIC),或可以通过对电路进行集成或封装的任何其他的合理方式等硬件或固件来实现,或以软件、硬件以及 固件三种实现方式中任意一种或以其中任意几种的适当组合来实现。或者,物品扩展模块210、价格雷达模块220和价格监控模块230中的至少一个可以至少被部分地实现为计算机程序模块,当该计算机程序模块被运行时,可以执行相应的功能。
本公开还提供了一种计算机系统,包括:一个或多个处理器;存储器,用于存储一个或多个程序,其中,当所述一个或多个程序被所述一个或多个处理器执行时,使得所述一个或多个处理器实现所述的物品推荐的方法。
图8示意性示出了根据本公开实施例的适于实现上文描述的方法的计算机系统的框图。图8示出的计算机系统仅仅是一个示例,不应对本公开实施例的功能和使用范围带来任何限制。
如图8所示,根据本公开实施例的计算机系统800包括处理器801,其可以根据存储在只读存储器(ROM)802中的程序或者从存储部分808加载到随机访问存储器(RAM)803中的程序而执行各种适当的动作和处理。处理器801例如可以包括通用微处理器(例如CPU)、指令集处理器和/或相关芯片组和/或专用微处理器(例如,专用集成电路(ASIC)),等等。处理器801还可以包括用于缓存用途的板载存储器。处理器801可以包括用于执行根据本公开实施例的方法流程的不同动作的单一处理单元或者是多个处理单元。
在RAM 803中,存储有系统800操作所需的各种程序和数据。处理器801、ROM 802以及RAM 803通过总线804彼此相连。处理器801通过执行ROM 802和/或RAM 803中的程序来执行根据本公开实施例的方法流程的各种操作。需要注意,所述程序也可以存储在除ROM 802和RAM 803以外的一个或多个存储器中。处理器801也可以通过执行存储在所述一个或多个存储器中的程序来执行根据本公开实施例的方法流程的各种操作。
根据本公开的实施例,系统800还可以包括输入/输出(I/O)接口805,输入/输出(I/O)接口805也连接至总线804。系统800还可以包括连接至I/O接口805的以下部件中的一项或多项:包括键盘、鼠标等的输入部分806;包括诸如阴极射线管(CRT)、液晶显示器(LCD)等以及扬声器等的输出部分807;包括硬盘等的存储部分808;以及包括诸如LAN卡、调制解调器等的网络接口卡的通信部分809。通信部分809经由诸如因特网的网络执行通信处理。驱动器810也根据需要连接至I/O接口805。可拆卸介质811,诸如磁盘、光盘、磁光盘、半导体存储器等等,根据需要安装在驱动器810上,以便于从其上读出的计算机程序根据需要被安装入存储部分808。
根据本公开的实施例,根据本公开实施例的方法流程可以被实现为计算机软件程序。例如,本公开的实施例包括一种计算机程序产品,其包括承载在计算机可读存储介质上的计算机程序,该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的实施例中,该 计算机程序可以通过通信部分809从网络上被下载和安装,和/或从可拆卸介质811被安装。在该计算机程序被处理器801执行时,执行本公开实施例的系统中限定的上述功能。根据本公开的实施例,上文描述的系统、设备、装置、模块、单元等可以通过计算机程序模块来实现。
本公开还提供了一种计算机可读存储介质,该计算机可读存储介质可以是上述实施例中描述的设备/装置/系统中所包含的;也可以是单独存在,而未装配入该设备/装置/系统中。上述计算机可读存储介质承载有一个或者多个程序,当上述一个或者多个程序被执行时,实现根据本公开实施例的方法。
根据本公开的实施例,计算机可读存储介质可以是非易失性的计算机可读存储介质。例如可以包括但不限于:便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开中,计算机可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
例如,根据本公开的实施例,计算机可读存储介质可以包括上文描述的ROM 802和/或RAM 803和/或ROM 802和RAM 803以外的一个或多个存储器。
附图中的流程图和框图,图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上,流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分,上述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意,在有些作为替换的实现中,方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如,两个接连地表示的方框实际上可以基本并行地执行,它们有时也可以按相反的顺序执行,这依所涉及的功能而定。也要注意的是,框图或流程图中的每个方框、以及框图或流程图中的方框的组合,可以用执行规定的功能或操作的专用的基于硬件的系统来实现,或者可以用专用硬件与计算机指令的组合来实现。本领域技术人员可以理解,本公开的各个实施例和/或权利要求中记载的特征可以进行多种组合和/或结合,即使这样的组合或结合没有明确记载于本公开中。特别地,在不脱离本公开精神和教导的情况下,本公开的各个实施例和/或权利要求中记载的特征可以进行多种组合和/或结合。所有这些组合和/或结合均落入本公开的范围。
以上对本公开的实施例进行了描述。但是,这些实施例仅仅是为了说明的目的,而并非为了限制本公开的范围。尽管在以上分别描述了各实施例,但是这并不意味着各个实施例中的措施不能有利地结合使用。本公开的范围由所附权利要求及其等同物限定。不脱离本公开 的范围,本领域技术人员可以做出多种替代和修改,这些替代和修改都应落在本公开的范围之内。

Claims (12)

  1. 一种物品推荐系统,包括:
    物品扩展模块,配置为响应用户输入的内容,对所述用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中所述物品集合中包括一个或多个物品;
    价格雷达模块,配置为监控所述物品集合中的物品的优惠信息;以及
    价格监控模块,配置为基于所述价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护所述物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和所述价格变动记录确定是否向所述用户推送提示信息。
  2. 根据权利要求1所述的系统,其中,所述物品扩展模块包括:
    品牌相似计算模块,配置为响应用户输入的关注物品,对所述关注物品进行扩展;以及
    品类关联计算模块,配置为响应用户输入的品类,对所述品类进行分析,挖掘所述品类可能存在的潜在交易场景,并确定所述潜在交易场景可能涉及的物品;
    其中,所述品牌相似计算模块扩展得到的物品和所述品类关联计算模块确定的物品组成所述物品集合。
  3. 根据权利要求2所述的系统,其中,所述品牌相似计算模块包括品牌图谱构建单元,配置为:
    计算任意两个品牌之间的品牌距离,其中,所述任意两个品牌之间的品牌距离被限定为所述两个品牌所涉及相同分类的个数和所述两个品牌所涉及的所有分类的个数的比值;
    将所述任意两个品牌之间的品牌距离与第一预设阈值进行比较,并将品牌距离大于所述第一预设阈值的两个品牌确定为关联品牌;
    在品牌图谱中构建属于关联品牌的两个品牌之间的关联关系;以及
    可视化展示所述品牌图谱。
  4. 根据权利要求2或3所述的系统,其中,所述品牌相似计算模块还包括品牌档次划分单元,配置为:
    计算每个品牌的平均价格溢出比,其中,所述每个品牌的平均价格溢出比被限定为按照如下公式计算:
    Figure PCTCN2021077669-appb-100001
    通过模糊数学确定所述每个品牌的平均价格溢出比属于不同档次的概率;
    根据所述每个品牌的平均价格溢出比属于不同档次的概率和为每个档次分配的权重计算所述每个品牌的最终概率;以及
    根据所述每个品牌的最终概率确定所述每个品牌所属的档次。
  5. 根据权利要求4所述的系统,其中,所述品牌图谱构建单元,还配置为:
    在所述品牌档次划分单元确定所述每个品牌所属的档次后,在所述品牌图谱中根据所述每个品牌所属的档次构建关于不同档次品牌的品牌图谱。
  6. 根据权利要求5所述的系统,其中,所述品牌相似计算模块还包括品牌连接判定单元,配置为:
    将任意两个品牌的最终概率相乘,计算得到所述任意两个品牌之间关联的概率;
    将所述任意两个品牌之间关联的概率与第二预设阈值进行比较,确定所述任意两个品牌之间是否关联;
    将所述任意两个品牌之间关联的概率大于或等于所述第二预设阈值的两个品牌确定为关联品牌;以及
    在所述品牌图谱中为所述确定为关联品牌的品牌建立关联关系。
  7. 根据权利要求3至6中任一项所述的系统,其中,所述品牌相似计算模块还包括相似物品选择单元,配置为:
    基于所述品牌图谱对所述用户输入的内容进行扩展,从与所述用户输入的内容相关联的品牌中扩展物品。
  8. 根据权利要求2所述的系统,其中,所述品类关联计算模块包括:
    品类场景挖掘单元,配置为通过分析预设时长范围内的历史订单,构建共同购买品类关联图谱,根据所述共同购买品类关联图谱挖掘一个或多个消费场景;以及
    物品场景查询单元,配置为根据所述用户输入的内容从所述共同购买品类关联图谱中查询获得与所述用户输入的内容关联的品类。
  9. 根据权利要求1所述的系统,其中,所述价格监控模块根据计算得到的物品实际价格和所述价格变动记录确定是否向所述用户推送提示信息包括:
    将所述计算得到的物品实际价格与所述价格变动记录中的历史价格进行比较;以及
    在所述计算得到的物品实际价格小于所述价格变动记录中的历史价格的情况下,向所述用户推送提示信息,或者通知所述价格雷达模块向所述用户推送优惠信息。
  10. 一种通过物品推荐系统进行物品推荐的方法,所述物品推荐系统包括物品扩展模块、价格雷达模块和价格监控模块,所述方法包括:
    所述物品扩展模块响应用户输入的内容,对所述用户输入的内容进行扩展,以生成用户感兴趣的物品集合,其中所述物品集合中包括一个或多个物品;
    所述价格雷达模块监控所述物品集合中的物品的优惠信息;以及
    所述价格监控模块基于所述价格雷达模块监控的物品的优惠信息计算物品实际价格,以及维护所述物品集合中的物品的价格变动记录,并根据计算得到的物品实际价格和所述价格变动记录确定是否向所述用户推送提示信息。
  11. 一种计算机系统,包括:
    一个或多个处理器;
    存储器,用于存储一个或多个程序,
    其中,当所述一个或多个程序被所述一个或多个处理器执行时,使得所述一个或多个处理器实现权利要求10所述的方法。
  12. 一种计算机可读存储介质,其上存储有可执行指令,该指令被处理器执行时使处理器实现权利要求10所述的方法。
PCT/CN2021/077669 2020-03-02 2021-02-24 物品推荐系统、物品推荐的方法、计算机系统和介质 WO2021175137A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/904,642 US20230099386A1 (en) 2020-03-02 2021-02-24 System of recommending item, method of recommending item, computer system, and medium
EP21764540.7A EP4116912A4 (en) 2020-03-02 2021-02-24 ARTICLE RECOMMENDATION SYSTEM, METHOD FOR ARTICLE RECOMMENDATION, COMPUTER SYSTEM AND MEDIUM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010137487.3A CN113362123A (zh) 2020-03-02 2020-03-02 物品推荐系统、物品推荐的方法、计算机系统和介质
CN202010137487.3 2020-03-02

Publications (1)

Publication Number Publication Date
WO2021175137A1 true WO2021175137A1 (zh) 2021-09-10

Family

ID=77523310

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/077669 WO2021175137A1 (zh) 2020-03-02 2021-02-24 物品推荐系统、物品推荐的方法、计算机系统和介质

Country Status (4)

Country Link
US (1) US20230099386A1 (zh)
EP (1) EP4116912A4 (zh)
CN (1) CN113362123A (zh)
WO (1) WO2021175137A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116308683B (zh) * 2023-05-17 2023-08-04 武汉纺织大学 基于知识图谱的服装品牌定位推荐方法、设备及存储介质
CN117217865B (zh) * 2023-09-12 2024-06-11 深圳市思维无限网络科技有限公司 基于大数据分析的个性化推荐系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8719110B1 (en) * 2011-07-06 2014-05-06 Intuit Inc. Discount offers based on a social graph
CN107341694A (zh) * 2017-07-07 2017-11-10 厦门趣品网络科技有限公司 一种基于商品大数据的智能导购方法
CN108335159A (zh) * 2017-01-20 2018-07-27 广州市动景计算机科技有限公司 项目组合推荐方法及系统
CN110009459A (zh) * 2019-04-15 2019-07-12 江苏易汇聚软件科技有限公司 电子商务平台的买家需求定制系统及方法
CN110163705A (zh) * 2018-02-13 2019-08-23 北京京东尚科信息技术有限公司 用于推送信息的方法和装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9443209B2 (en) * 2009-04-30 2016-09-13 Paypal, Inc. Recommendations based on branding
US20130325653A1 (en) * 2012-06-05 2013-12-05 Myworld, Inc. Commerce System and Method of Providing Personalized Marketing Presentation with Products of Interest for Consumer
US10803472B2 (en) * 2012-10-30 2020-10-13 Transform Sr Brands Llc Method and system for emergent data processing
CN102968506A (zh) * 2012-12-14 2013-03-13 北京理工大学 一种基于扩展特征向量的个性化协同过滤推荐方法
CN106447429A (zh) * 2016-09-09 2017-02-22 宇龙计算机通信科技(深圳)有限公司 一种消息推送方法、装置及终端
CN106897899A (zh) * 2017-01-24 2017-06-27 武汉奇米网络科技有限公司 一种将客户分群后个性化推荐商品的方法及系统
WO2020041335A1 (en) * 2018-08-21 2020-02-27 Wal-Mart Stores, Inc. Item line assignment systems and methods
CN110443649A (zh) * 2019-08-02 2019-11-12 杭州讯酷科技有限公司 基于匹配算法的智能推荐方法、系统及售货机

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8719110B1 (en) * 2011-07-06 2014-05-06 Intuit Inc. Discount offers based on a social graph
CN108335159A (zh) * 2017-01-20 2018-07-27 广州市动景计算机科技有限公司 项目组合推荐方法及系统
CN107341694A (zh) * 2017-07-07 2017-11-10 厦门趣品网络科技有限公司 一种基于商品大数据的智能导购方法
CN110163705A (zh) * 2018-02-13 2019-08-23 北京京东尚科信息技术有限公司 用于推送信息的方法和装置
CN110009459A (zh) * 2019-04-15 2019-07-12 江苏易汇聚软件科技有限公司 电子商务平台的买家需求定制系统及方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4116912A4

Also Published As

Publication number Publication date
CN113362123A (zh) 2021-09-07
EP4116912A1 (en) 2023-01-11
US20230099386A1 (en) 2023-03-30
EP4116912A4 (en) 2023-12-27

Similar Documents

Publication Publication Date Title
Saura Using data sciences in digital marketing: Framework, methods, and performance metrics
JP6913241B2 (ja) 信用力があると判定された消費者にローンを発行するシステムおよび方法
Mishra et al. Big Data and supply chain management: a review and bibliometric analysis
US11354584B2 (en) Systems and methods for trend aware self-correcting entity relationship extraction
KR101939554B1 (ko) 일시적 거래 한도 결정
Fu et al. Fused latent models for assessing product return propensity in online commerce
Yu et al. Shop-type recommendation leveraging the data from social media and location-based services
US20180240158A1 (en) Computer implemented system and method for customer profiling using micro-conversions via machine learning
WO2021175137A1 (zh) 物品推荐系统、物品推荐的方法、计算机系统和介质
WO2022100518A1 (zh) 一种基于用户画像的物品推荐方法和装置
US10402840B2 (en) Systems and methods for setting product prices
Kumar et al. A computational framework for ranking prediction of cloud services under fuzzy environment
Clausen et al. Big data driven order-up-to level model: Application of machine learning
JP2019091355A (ja) 判定装置、判定方法及び判定プログラム
CN110880082A (zh) 服务评价方法、装置、系统、电子设备和可读存储介质
US11232470B2 (en) Customer flow line and customer flow hot zone determining method and apparatus
Zenkina About current trends in global e-commerce
CN110720099A (zh) 基于种子监督学习提供推荐的系统和方法
Dutta et al. Big data analytics for real time systems
JP2022011154A (ja) 評価装置、評価方法及び評価プログラム
CN110796520A (zh) 商品推荐的方法、装置、计算设备以及介质
Kamala Development of an effective method of data collection for advertising and marketing on the internet
US11244329B2 (en) Metric forecasting in a digital medium environment
CN113159877A (zh) 数据处理方法、装置、系统、计算机可读存储介质
Zhu Research on multi‐source mobile commerce service recommendation model of data fusion based on tree network

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21764540

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021764540

Country of ref document: EP

Effective date: 20221004