WO2020073524A1 - 一种线下商品推荐方法、装置和电子设备 - Google Patents

一种线下商品推荐方法、装置和电子设备 Download PDF

Info

Publication number
WO2020073524A1
WO2020073524A1 PCT/CN2018/124836 CN2018124836W WO2020073524A1 WO 2020073524 A1 WO2020073524 A1 WO 2020073524A1 CN 2018124836 W CN2018124836 W CN 2018124836W WO 2020073524 A1 WO2020073524 A1 WO 2020073524A1
Authority
WO
WIPO (PCT)
Prior art keywords
commodities
visiting customer
identity information
information
visiting
Prior art date
Application number
PCT/CN2018/124836
Other languages
English (en)
French (fr)
Inventor
王逸峰
张兆丰
黄轩
汤先锋
邱念
王孝宇
Original Assignee
深圳云天励飞技术有限公司
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 深圳云天励飞技术有限公司 filed Critical 深圳云天励飞技术有限公司
Publication of WO2020073524A1 publication Critical patent/WO2020073524A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • 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
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • 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
    • 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

Definitions

  • the present invention relates to the field of Internet technology, and in particular to an offline product recommendation method, device, and electronic equipment.
  • the offline product recommendation method is basically through the shopping guide's observation of the consumer and recommending products to the consumer based on his own experience. If a customer comes to the store, the store's shopping guide asks the customer what product to buy, and then The customer recommends the same type of product, or recommends to the customer the product that is currently active in the customer's browsing product based on the customer's behavior of browsing the product after entering the store.
  • the recommendation of shopping guides is more subjective, and the effect of recommending to customers is often poor. It can be seen that the current offline product recommendation method has a problem of poor recommendation effect.
  • Embodiments of the present invention provide an offline product recommendation method, device, and electronic device, which can improve the product recommendation effect.
  • an embodiment of the present invention provides an offline product recommendation method, including:
  • the step of determining the recommended commodities of the visiting customer from the multiple commodities based on the visual feature information and the scores of the multiple commodities obtained in advance includes:
  • the target identity information exists in the identity information database, obtain scores of multiple commodities from the target identity information in a pre-established scoring matrix;
  • the recommended commodities of the visiting customer are determined from the multiple commodities according to the scores of the multiple commodities according to the target identity information.
  • the step of determining the recommended commodities of the visiting customer from the multiple commodities based on the visual feature information and the scores of the multiple commodities obtained in advance further includes:
  • the N commodities recorded in the scoring matrix are used as recommended commodities for the visiting customer, where the N commodities are scored for the multiple commodities in accordance with Commodities ranked in the top N positions from high to low, where N is an integer greater than or equal to 1.
  • the method further includes:
  • the visiting customer If the visiting customer successfully purchases goods, collect the identity information of the visiting customer and establish a correspondence between the identity information of the visiting customer and the goods purchased by the visiting customer, wherein the correspondence includes the visiting The transaction relationship of the goods purchased by the customer.
  • an offline product recommendation device including:
  • Acquisition module used to obtain the visual information of visiting customers
  • An extraction module for extracting visual feature information from the visual information
  • a determining module configured to determine the recommended commodities of the visiting customer from the multiple commodities based on the visual feature information and the scores of the multiple commodities obtained in advance;
  • a display module is used to display the recommended products to the visiting customer.
  • an embodiment of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, when the processor executes the computer program
  • the steps in the offline product recommendation method provided by the embodiments of the present invention are implemented.
  • an embodiment of the present invention provides a computer-readable storage medium that stores a computer program on the computer-readable storage medium, and when the computer program is executed by a processor, implements the offline product recommendation method provided by the embodiment of the invention Steps.
  • the visual information of the visiting customer is obtained; the visual characteristic information is extracted from the visual information; the visiting is determined from the multiple commodities based on the visual characteristic information and the scores of the multiple commodities obtained in advance Recommended commodities of customers; showing the recommended commodities to the visiting customers.
  • the visual information of the visiting customer is obtained; the visual characteristic information is extracted from the visual information; the visiting is determined from the multiple commodities based on the visual characteristic information and the scores of the multiple commodities obtained in advance Recommended commodities of customers; showing the recommended commodities to the visiting customers.
  • FIG. 3 is a schematic structural diagram of an offline product recommendation device provided by an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of another offline product recommendation device provided by an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of another offline product recommendation device provided by an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of another offline product recommendation device provided by an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
  • FIG. 1 is a flowchart of an offline product recommendation method provided by an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:
  • the visual information of the visiting customer may include the face information of the visiting customer, and may also include physical information (such as height, measurements, etc.), wearing information (such as hats, clothes, pants, shoes, hats, clothes , Color information of pants, shoes, etc.), age information, gender information, etc .; step 101 can obtain the above visual information through a camera, the camera can be installed at the entrance of the store, cargo area, counter or smart terminal, etc. , The installation position of the camera is not limited here. As long as the visiting customer enters the visual range of the camera, the visual information of the visiting customer can be obtained. For example, the camera is installed at the entrance of the store. When there is a visiting customer, the camera can take pictures of the visiting customer and analyze the photos of the visiting customer to obtain the visual information of the visiting customer.
  • the visiting client may also be called a visitor or a consumer.
  • the extracting the visual feature information from the visual information may be recognizing the photo information of the visiting customer photographed by a camera through face recognition technology, and then extracting the visual feature information of the visiting customer from this photo information .
  • the face information of the visiting customer is extracted from the photo information captured by the camera.
  • it can also extract other visual characteristic information, such as height, clothing and other characteristic information.
  • the multiple commodities mentioned above may be multiple pre-recorded commodities applied to the commodity recommendation system (or device) in the embodiment of the present invention, for example, all or part of the commodities in a store.
  • the pre-acquired scores of multiple commodities may be based on the historical purchase records of the visiting customers, combined with the corresponding recommendation algorithm, and will generate a rating row in a matrix containing the ratings of each visiting customer's preference for different commodities at regular intervals Before the multiple commodities.
  • the scores of the plurality of commodities determined in advance by the consumer's purchase records for the plurality of commodities. For example, the more a consumer buys a product, the higher the rating of the product, or the more times a consumer buys a product, the higher the rating of the product. If there are 100 consumers, score 500 commodities separately, and sum the scores of each similar commodity in 500 commodities, and then line up the scores of each commodity in a certain order, which can be high To low.
  • the 103 can be based on the above-mentioned visual feature information to determine the visiting customer's gender, age and other characteristics, and then can recommend to the customer to match these features and high-scoring products, or can be based on the visual feature information to determine the identity of the visiting customer Information, so as to obtain the ratings of each product corresponding to the identity information, and then recommend products with higher ratings to the visiting customer based on these ratings.
  • the recommended products can be displayed to the visiting customer, which can be displayed through a display device, and the display device can be a device with a display function such as a mobile phone, a computer, a tablet computer, etc. , There are no restrictions on this display device.
  • the recommended commodities displayed can be recommended to the visiting customer, and because the recommendation is based on the visual feature information and the score, the effect of product recommendation can be achieved.
  • offline product recommendation method provided by the embodiments of the present invention may be applied to devices such as smart terminals, mobile phones, and tablet computers of offline commodity stores.
  • the visual information of the visiting customer is obtained; the visual characteristic information is extracted from the visual information; the visual characteristic information and the scores of a plurality of commodities obtained in advance are used to determine the Recommended commodities from visiting customers; showing the recommended commodities to the visiting customers.
  • the visual characteristic information is extracted from the visual information; the visual characteristic information and the scores of a plurality of commodities obtained in advance are used to determine the Recommended commodities from visiting customers; showing the recommended commodities to the visiting customers.
  • FIG. 2 is a flowchart of another offline product recommendation method provided in an embodiment of the present invention. As shown in FIG. 2, it includes the following steps:
  • the identity information database records the identity information and visual feature information of multiple customers who have purchase records, so that the identity information database can query whether there is target identity information matching the visual feature information. If there is target information matching the visual feature information in the identity information database, it means that the above-mentioned visiting customer has previously purchased a commodity, otherwise, it means that the above-mentioned visiting customer has not previously purchased a commodity.
  • the extracted visual feature information of the visiting customer can be compared with the visual feature information stored in the identity information database, and the identity information corresponding to the stored visual feature information is the visiting customer Target identity information of visual feature information.
  • the target identity information exists in the identity information database, obtain scores of multiple commodities from the target identity information in a pre-established scoring matrix.
  • the aforementioned pre-established scoring matrix may be a scoring matrix composed of historical consumer ratings of commodities.
  • the high ratings of the commodities indicate that consumers are more fond of the commodities.
  • the scoring method may be set in advance according to consumer purchase records For example, the more a consumer buys a product, the higher the rating of the product, or the more times a consumer buys a product, the higher the rating of the product. Further, according to the purchase records of all consumers, combined with the corresponding recommendation algorithm, a matrix containing scores of various consumers' preferences for different commodities will be generated at regular intervals. When the above target identity information is determined, a row in the scoring matrix that belongs to the target identity information will return a preference score for each product.
  • multiple commodities with higher scores can be selected as recommended commodities for visiting customers among the multiple commodities, for example: selecting the top N in the ranking from high to low Products, the N is an integer greater than or equal to 1.
  • the products recommended to the visiting customer can be implemented according to the rating of the visiting customer on multiple commodities, for example, recommending products with a higher rating to the visiting customer, thereby further improving the product recommendation effect.
  • the method further includes:
  • the N commodities recorded in the scoring matrix are used as recommended commodities for the visiting customer, where the N commodities are scored for the multiple commodities in accordance with Commodities ranked in the top N positions from high to low, where N is an integer greater than or equal to 1.
  • the target identity information does not exist in the identity information database
  • the identity information of the visiting customer has not been recorded in the identity database
  • the visiting customer is the first time to visit a store. Since the identity information of the visiting customer has not been recorded in the above-mentioned identity database, it can be directly recommended based on the product ratings recorded by the rating matrix, for example: recommending products based on a large number of customers, such as the ratings recorded by the recommendation rating matrix (these The rating may be the rating of the product by other customers, or the highest rating of the same product by multiple customers).
  • the value of N can be pre-configured, for example, 5 or 10.
  • the above steps may be to recommend products with higher scores to customers who do not record identity information, thereby further improving the recommendation effect.
  • the use of the N commodities recorded in the scoring matrix as recommended commodities of the visiting customer includes:
  • N commodities in the commodity recommendation type are the plurality of commodities belonging to the commodity recommendation type.
  • the products with the lowest N ranking in the top N are the products with the lowest N ranking in the top N.
  • the above product recommendation type may be determined according to the age, gender, voice or dress of the visiting user, for example: a sports store, the product type may include: sports shoes, sports pants, sportswear, etc.
  • the product recommendation type may be determined according to the age, gender, voice or dress of the visiting user, for example: a sports store
  • the product type may include: sports shoes, sports pants, sportswear, etc.
  • the determining the type of product recommendation of the visiting customer includes:
  • the aforementioned voice information may be the voice information of the consumer or the voice information of the shopping guide.
  • the voice information includes the type information of the commodity.
  • the consumer When the shopping guide asks the consumer's purchase intention, the consumer generally tends to say a commodity category. For example, the consumer may indicate that he wants to buy sports shoes.
  • the voice device worn by the shopping guide can be used to collect the consumer's voice information, or the shopping guide can repeat the consumer's voice information, and then the voice recognition system can be used to identify the type of goods the consumer desires to purchase, and the consumption
  • the buyer displays the type of goods that the consumer desires to purchase, or the buyer guides the buyer to display the type of goods the consumer desires to purchase, and provides the shopping guide with a recommended reference range.
  • the shopping guide can touch the display terminal to select a category of sports shoes to narrow the recommendation range.
  • the display terminal may be a computer, mobile phone or tablet computer.
  • the recommendation range can be narrowed, and the types of goods that the consumer expects to purchase can be recommended to the consumer more quickly, which improves the recommendation effect.
  • the method further includes:
  • the visiting customer If the visiting customer successfully purchases goods, collect the identity information of the visiting customer and establish a correspondence between the identity information of the visiting customer and the goods purchased by the visiting customer, wherein the correspondence includes the visiting The transaction relationship of the goods purchased by the customer.
  • the purchased product purchased by the visiting customer may be a product recommended by the recommendation system, or may be another product that the visiting customer prefers.
  • the original recommendation is clothes, but consumers are fancy shoes.
  • the transaction information can be recorded under the identity information corresponding to the visiting customer.
  • the correspondence relationship between the identity information of the visiting customer and the goods purchased by the visiting customer can be registered in the identity information database, and the identity information of the visiting customer can be established with the visiting customer
  • the correspondence relationship of the purchased products can be further added to the above-mentioned scoring matrix based on the products purchased by the visiting customer.
  • the correspondence between the identity information of the visiting customer and the goods purchased by the visiting customer can be established, so that the next time the visiting customer visits the store, the corresponding relationship can be made according to the correspondence Product recommendation to further improve the product recommendation effect.
  • FIG. 3 is a schematic structural diagram of an offline product recommendation device provided by an embodiment of the present invention. As shown in FIG. 3, it includes:
  • the obtaining module 301 is used to obtain the visual information of the visiting client
  • the determining module 303 is configured to determine the recommended commodities of the visiting customer from the multiple commodities according to the visual feature information and the scores of the multiple commodities obtained in advance;
  • the display module 304 is used to display the recommended products to the visiting customer.
  • the determination module 303 includes:
  • the judging unit 3031 is used to query whether there is target identity information matching the visual characteristic information in the identity information database;
  • the recommendation unit 3032 is configured to obtain, if the target identity information exists in the identity information database, scores of multiple commodities by the target identity information in a pre-established scoring matrix; , Determine the recommended products of the visiting customer from the plurality of products.
  • the device further includes:
  • the recommendation module 305 is configured to, if the target identity information does not exist in the identity information database, use N commodities recorded in the scoring matrix as recommended commodities of the visiting customer, wherein the N commodities are the multiple Commodities are ranked in the top N products in order of high to low, where N is an integer greater than or equal to 1.
  • the recommendation module 305 is used to determine the product recommendation type of the visiting customer if the target identity information does not exist in the identity information database; and record the scoring matrix to record N products in the product recommendation type As the recommended products of the visiting customer, the N products are the top N products in which the plurality of products belong to the product recommendation type and the ranking is from high to low.
  • the recommendation module 305 determines the product recommendation type of the visiting customer in the following ways:
  • the device further includes:
  • the transaction recording module 306 is used to collect the identity information of the visiting customer and establish a correspondence between the identity information of the visiting customer and the commodities purchased by the visiting customer if the visiting customer purchases the commodity successfully.
  • the correspondence relationship includes the transaction relationship of the goods purchased by the visiting customer.
  • the scores of the multiple pre-acquired commodities include:
  • the scores of the plurality of commodities determined in advance according to the purchase records of the consumers for the plurality of commodities.
  • the offline product recommendation system provided by the embodiment of the present invention can implement various processes implemented by the offline product recommendation method in the foregoing method embodiments, and to avoid repetition, details are not described herein again. And can achieve the same beneficial effect.
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. As shown in FIG. 7, it includes: a memory 702, a processor 701, and stored on the memory 702 and can be stored in the processor A computer program running on 701, of which:
  • the processor 701 is used to call the computer program stored in the memory 702 and perform the following steps:
  • the step of determining the recommended commodities of the visiting customer from the multiple commodities according to the visual feature information and the pre-acquired scores of the multiple commodities performed by the processor 701 includes:
  • the target identity information exists in the identity information database, obtain scores of multiple commodities from the target identity information in a pre-established scoring matrix;
  • the recommended commodities of the visiting customer are determined from the multiple commodities according to the scores of the multiple commodities according to the target identity information.
  • the processor 701 is further configured to:
  • the N commodities recorded in the scoring matrix are used as recommended commodities for the visiting customer, where the N commodities are scored for the multiple commodities in accordance with Commodities ranked in the top N positions from high to low, where N is an integer greater than or equal to 1.
  • the processor 701 executing the N commodities recorded in the scoring matrix as recommended commodities of the visiting customer includes:
  • N commodities in the commodity recommendation type are the plurality of commodities belonging to the commodity recommendation type.
  • the products with the lowest N ranking in the top N are the products with the lowest N ranking in the top N.
  • the determining of the commodity recommendation type of the visiting customer performed by the processor 701 includes:
  • the processor 701 is further used to determine whether the target identity information exists in the identity information database. If the target identity information does not exist in the identity information database, after using the N commodities recorded in the scoring matrix as recommended commodities of the visiting customer, the processor 701 is further used to determine whether the target identity information does not exist in the identity information database.
  • the visiting customer If the visiting customer successfully purchases goods, collect the identity information of the visiting customer and establish a correspondence between the identity information of the visiting customer and the goods purchased by the visiting customer, wherein the correspondence includes the visiting The transaction relationship of the goods purchased by the customer.
  • the scores of the multiple pre-acquired commodities include:
  • the scores of the plurality of commodities determined in advance according to the purchase records of the consumers for the plurality of commodities.
  • the above-mentioned electronic device may be a smart terminal, a mobile phone, a tablet computer and other devices of an offline commodity store.
  • the electronic device provided by the embodiment of the present invention can implement various implementation manners in the method embodiments of FIG. 1 and FIG. 2 and corresponding beneficial effects. To avoid repetition, details are not described herein again.
  • An embodiment of the present invention also provides a computer-readable storage medium that stores a computer program on the computer-readable storage medium.
  • the computer program is executed by the processor 701
  • each process of the offline product recommendation method provided by the embodiment of the present invention is implemented. And can achieve the same technical effect, in order to avoid repetition, no more details here.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM for short), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Mathematical Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Algebra (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明实施例提供一种线下商品推荐方法、装置和电子设备，所述方法包括：获取来访客户的视觉信息；从所述视觉信息中提取视觉特征信息；根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；向所述来访客户展示所述推荐商品。本发明实施例能够提高商品推荐效果。

Description

一种线下商品推荐方法、装置和电子设备
本申请要求于2018年10月10日提交中国专利局，申请号为201811179621.5、发明名称为“一种线下商品推荐方法、装置和电子设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及互联网技术领域，尤其涉及一种线下商品推荐方法、装置和电子设备。
背景技术
在产品多元化的时代，商家为了提高购物服务，推行了多种购物服务方式，例如：商品推荐服务。线下的商品推荐方式基本是通过导购员对消费者的观察以及凭借自己的经验向消费者推荐商品，如某客户来到门店，该门店的导购员向该客户询问需要购买什么商品，进而向客户推荐相同类型商品，或者根据客户进门店后的浏览商品的行为，向客户推荐客户浏览商品中当前做活动的商品。然而，导购员推荐都是比较主观，向客户推荐的效果往往比较差。可见，目前线下的商品推荐方式存在推荐效果差的问题。
发明内容
本发明实施例提供一种线下商品推荐方法、装置和电子设备，能够提高商品推荐效果。
第一方面，本发明实施例提供一种线下商品推荐方法，包括：
获取来访客户的视觉信息；
从所述视觉信息中提取视觉特征信息；
根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
向所述来访客户展示所述推荐商品。
可选的，所述根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品的步骤，包括：
在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息；
若所述身份信息数据库存在所述目标身份信息，获取预先建立的评分矩阵中所述目标身份信息对多个商品的评分；
根据所述目标身份信息对多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
可选的，所述根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品的步骤，进一步包括：
若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品的评分按照从高到低的排序排在前N位的商品，所述N为大于或者等于1的整数。
可选的，所述若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品之后，所述方法还包括：
若所述来访客户购买商品成功，则采集所述来访客户的身份信息，并建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系，其中，所述对应关系包括所述来访客户购买的商品的交易关系。
第二方面，本发明实施例提供一种线下商品推荐装置，包括：
获取模块，用于获取来访客户的视觉信息；
提取模块，用于从所述视觉信息中提取视觉特征信息；
确定模块，用于根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
展示模块，用于向所述来访客户展示所述推荐商品。
第三方面，本发明实施例提供一种电子设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现本发明实施例提供的线下商品推荐方法中的步骤。
第四方面，本发明实施例提供一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现发明实施 例提供的线下商品推荐方法中的步骤。
本发明实施例中，获取来访客户的视觉信息；从所述视觉信息中提取视觉特征信息；根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；向所述来访客户展示所述推荐商品。这样可以实现根据视觉信息和商品评分进行商品推荐，从而相比导购员主观推荐商品，能够提高商品推荐效果。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的一种线下商品推荐方法的流程图；
图2是本发明实施例提供的另一种线下商品推荐方法的流程图；
图3是本发明实施例提供的一种线下商品推荐装置的结构示意图；
图4是本发明实施例提供的另一种线下商品推荐装置的结构示意图；
图5是本发明实施例提供的另一种线下商品推荐装置的结构示意图；
图6是本发明实施例提供的另一种线下商品推荐装置的结构示意图；
图7是本发明实施例提供的一种电子设备的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参见图1，图1是本发明实施例提供的一种线下商品推荐方法的流程图，如图1所示，包括以下步骤：
101、获取来访客户的视觉信息。
其中，上述来访客户的视觉信息可以包括来访客户的人脸信息，还可以包括形体信息(例如身高、三围等)、穿着信息(例如佩戴的帽子、穿的衣服、 裤子、鞋子，以及帽子、衣服、裤子、鞋子的颜色信息等)、年龄信息、性别信息等的一种或多种；步骤101可以通过摄像头来获取上述视觉信息，该摄像头可以安装在商店门口、货物区、柜台或者智能终端等，在这里摄像头的安装位置不做限制。只要来访客户进入摄像头的可视范围内，就可以获取所述来访客户的视觉信息。例如：摄像头安装在商店门口，当有来访客户时，通过摄像头对来访客户进行拍照并将拍到的来访客户的照片进行分析，就可以得到该来访客户的视觉信息。
需要说明的是，本发明实施例中，所述来访客户也可以称作访客，也可以称作消费者。
102、从所述视觉信息中提取视觉特征信息。
其中，所述从所述视觉信息中提取视觉特征信息可以是通过人脸识别技术识别出摄像头拍下的所述来访客户的照片信息，然后从这个照片信息中提取所述来访客户的视觉特征信息。例如：从所述摄像头拍下的照片信息中提取所述来访客户的人脸信息。当然，还可以是提取其他视觉特征信息，例如：身高、衣着等特征信息。
103、根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
其中，上述多个商品可以是本发明实施例中应用于商品推荐系统(或者设备)预先记录的多个商品，例如：某门店内的所有或者部分商品。所述预先获取的多个商品的评分可以是根据来访客户的历史购买记录,结合对应的推荐算法,每隔一定时间会生成一个包含有各个来访客户对于不同商品的喜好程度评分的矩阵中评分排前的多个商品。优选的，预先根据消费者针对所述多个商品的购买记录确定的所述多个商品的评分。例如：某消费者购买某商品的数量越多，则该商品的评分越高，或者某消费者购买某商品的次数越多，则该商品的评分越高。如有100个消费者，分别对500个商品进行评分，并分别将500个商品中的每一个同类商品的分数进行求和，然后将每一个商品的分数按一定顺序进行排队，可以是从高到低。
103可以是根据上述视觉特征信息，确定来访客户的性别、年龄等特征，进而可以向该客户推荐与这些特征相匹配，且评分较高的商品，或者可以是根据视觉特征信息确定来访客户的身份信息，从而得到该身份信息对应各商品的 评分，进而根据这些评分向该来访客户推荐评分较高的商品。
104、向所述来访客户展示所述推荐商品。
其中，确定所述来访客户的推荐商品后就可以向所述来访客户展示所述推荐商品，可以是通过显示设备进行展示，所述显示设备可以是手机、电脑、平板电脑等具有显示功能的设备，在这里对这种显示设备不做限制。
这样可以将展示的所述推荐商品向来访客户推荐该推荐商品，且由于是根据视觉特征信息和评分进行推荐，从而能够商品推荐的效果。
需要说明的是，本发明实施例提供的线下商品推荐方法可以应用于线下商品门店的智能终端、手机、平板电脑等设备。
在本发明实施例中，获取来访客户的视觉信息；从所述视觉信息中提取视觉特征信息；根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；向所述来访客户展示所述推荐商品。这样可以实现根据视觉信息和商品评分进行商品推荐，从而相比导购员主观推荐商品，能够提高商品推荐效果。
参见图2，图2是本发明实施例中提供的另一种线下商品推荐方法的流程图，如图2所示，包括以下步骤：
201、获取来访客户的视觉信息。
202、从所述视觉信息中提取视觉特征信息。
203、在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息。
其中，上述身份信息数据库记录有多个存在购买记录的客户的身份信息和视觉特征信息，从而在该身份信息数据库可以查询是否存在与所述视觉特征信息相匹配的目标身份信息。若所述身份信息数据库中存在与所述视觉特征信息相匹配的目标信息，则表示上述来访客户之前购买过商品，反之，则表示上述来访客户之前未购买过商品。具体可以是，将提取到的所述来访客户视觉特征信息与所述身份信息数据库中存储的视觉特征信息进行对比，并存在于所述存储的视觉特征信息对应的身份信息即为，所述来访客户视觉特征信息的目标身份信息。
204、若所述身份信息数据库存在所述目标身份信息，获取预先建立的评分矩阵中所述目标身份信息对多个商品的评分。
上述预先建立的评分矩阵可以是历史记录的消费者对商品的评分组成的评分矩阵，所述商品的评分高说明消费者对商品更加喜爱，所述评分的方式可以预先根据消费者购买记录设置，例如：某消费者购买某商品的数量越多，则该商品的评分越高，或者某消费者购买某商品的次数越多，则该商品的评分越高。进一步，可以是根据所有消费者的购买记录,结合对应的推荐算法,每隔一定时间会生成一个包含有各个消费者对于不同商品的喜好程度评分的矩阵。当确定上述目标身份信息后，该评分矩阵中属于该目标身份信息的一行会返回对于各个商品的喜好评分。
205、根据所述目标身份信息对多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
当目标身份信息对多个商品的评分确定之后，从而可以在多个商品中选择评分较高的多个商品作为来访客户的推荐商品，例如：选择评分按照从高到低的排序中的前N个商品，该N为大于或者等于1的整数。
206、向所述来访客户展示所述推荐商品。
通过步骤206可以实现向来访客户推荐的商品为，根据该来访客户对多个商品的评分进行推荐的，例如：推荐该来访客户评分较高的商品，从而进一步提高商品推荐效果。
作为一种可选的实施方式，所述在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息之后，所述方法还包括：
若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品的评分按照从高到低的排序排在前N位的商品，所述N为大于或者等于1的整数。
其中，上述若所述身份信息数据库不存在所述目标身份信息可以理解为，上述身份数据库中没有记录过所述来访客户的身份信息，例如：上述来访客户是第一次来门店。由于上述身份数据库中没有记录过所述来访客户的身份信息，从而可以直接根据评分矩阵记录的商品评分进行推荐，例如：根据大量客户对商品的评分进行推荐，如推荐评分矩阵记录的评分(这些评分可以是其他客户对商品的评分，或者多个客户对同一商品评分的平均分)最高的一个或者多个商品。而上述N可以预先配置的数值，例如：5或者10等。
该实施方式中，通过上述步骤可以是向没有记录身份信息的客户推荐评分较高的商品，进一步提高推荐效果。
可选的，所述将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，包括：
确定所述来访客户的商品推荐类型；
将所述评分矩阵记录所述商品推荐类型中的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品属于所述商品推荐类型中评分按照从高到低的排序排在前N位的商品。
其中，上述商品推荐类型可以是根据来访用户的年龄、性别、语音或者着装确定的，例如：某运动类商店，所述商品类型就可以包括：运动鞋、运动裤、运动衣等，当上述来访用户说明需要买运动衣时，则可以确定运动衣这一商品推荐类型，或者在智能终端上提供用户选择商品推荐类型的接口，通过该接口与用户的交互确定商品推荐类型。
通过上述步骤可以实现在确定的商品推荐类型选择推荐商品，从而可以提高商品推荐的准确性，以及降低计算量。
可选的，所述确定所述来访客户的商品推荐类型，包括：
接收语音信息，将与所述语音信息对应的商品推荐类型作为所述来访客户的商品推荐类型；或者
接收触摸输入，将所述触摸输入选择的商品推荐类型作为所述来访客户的商品推荐类型。
其中，上述的语音信息可以是消费者的语音信息也可以是导购员的语音信息。所述语音信息包括商品的类型信息，当导购询问消费者的购买意向时,消费者一般倾向于说出一个商品类别，例如：消费者可能会表示自己想购买运动鞋。
这时可以通过导购员佩戴的语音装置，收录消费者的语音信息，或者导购员重复消费者的语音信息，则通过语音识别系统来识别出所述消费者期望购买的商品类型，并所述消费者展示所述消费者期望购买的商品类型，又或者向导购员展示所述消费者期望购买的商品类型，给所述导购员提供一个推荐参考范围。
这时如果所述导购员面对着的是显示终端，就可以通过导购员触摸显示终 端上触碰选择运动鞋者一类别来缩小推荐范围。其中，所述显示终端可以是电脑、手机或者平板电脑。
这样可以缩小推荐范围，更快的将所述消费者期望购买的商品类型推荐给消费者，提高了推荐效果。
可选的，所述若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品之后，所述方法还包括：
若所述来访客户购买商品成功，则采集所述来访客户的身份信息，并建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系，其中，所述对应关系包括所述来访客户购买的商品的交易关系。
其中，所述来访客户购买商品可以是通过推荐系统推荐的商品，也可以是来访客户自己中意的其他商品。例如：本来推荐的是衣服，但是消费者却看中了鞋子。只要是所述来访客户的购买商品都可以将所述交易信息记录在所述来访客户对应的身份信息下。
而上述建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系可以在是上述身份信息数据库中注册所述来访客户的身份信息，并建立来访客户的身份信息与所述来访客户购买的商品的对应关系，进一步，还可以根据该来访客户购买的商品在上述评分矩阵中添加该客户对多个商品的评分。
该实施方式中，由于在来访客户购买商品成功，可以建立来访客户的身份信息与所述来访客户购买的商品的对应关系，从而该来访客户下次来门店时，可以根据该对应关系进行相应的商品推荐，从而进一步提高商品推荐效果。
本实施例中，在图1所示的实施例的基础上增加了多种可选的实施方式，且可以进一步提高商品推荐效果。
请参见图3，图3是本发明实施例提供的一种线下商品推荐装置的结构示意图，如图3所示，包括：
获取模块301，用于获取来访客户的视觉信息；
提取模块302，用于从所述视觉信息中提取视觉特征信息；
确定模块303，用于根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
展示模块304，用于向所述来访客户展示所述推荐商品。
可选的，如图4所示，所述确定模块303包括：
判断单元3031，用于在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息；
推荐单元3032，用于若所述身份信息数据库存在所述目标身份信息，获取预先建立的评分矩阵中所述目标身份信息对多个商品的评分；根据所述目标身份信息对多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
可选的，如图5所示，所述装置还包括：
推荐模块305，用于若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品的评分按照从高到低的排序排在前N位的商品，所述N为大于或者等于1的整数。
可选的，推荐模块305用于若所述身份信息数据库不存在所述目标身份信息，确定所述来访客户的商品推荐类型；以及将所述评分矩阵记录所述商品推荐类型中的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品属于所述商品推荐类型中评分按照从高到低的排序排在前N位的商品。
选的，推荐模块305通过以下方式确定所述来访客户的商品推荐类型：
接收语音信息，将与所述语音信息对应的商品推荐类型作为所述来访客户的商品推荐类型；或者
接收触摸输入，将所述触摸输入选择的商品推荐类型作为所述来访客户的商品推荐类型。
可选的，如图6所示，所述装置还包括：
交易记录模块306，用于若所述来访客户购买商品成功，则采集所述来访客户的身份信息，并建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系，其中，所述对应关系包括所述来访客户购买的商品的交易关系。
可选的，所述预先获取的多个商品的评分，包括：
预先根据消费者针对所述多个商品的购买记录确定的所述多个商品的评分。
本发明实施例提供的线下商品推荐系统能够实现上述方法实施例中线下商品推荐方法实现的各个过程，为避免重复，这里不再赘述。且可以达到相同的有益效果。
参见图7，图7是本发明实施例提供的一种电子设备的结构示意图，如图7所示，包括：存储器702、处理器701及存储在所述存储器702上并可在所述处理器701上运行的计算机程序，其中：
处理器701用于调用存储器702存储的计算机程序，执行如下步骤：
从所述视觉信息中提取视觉特征信息；
根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
向所述来访客户展示所述推荐商品。
可选的，所述处理器701执行的所述根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品的步骤，包括：
在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息；
若所述身份信息数据库存在所述目标身份信息，获取预先建立的评分矩阵中所述目标身份信息对多个商品的评分；
根据所述目标身份信息对多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
可选的，所述在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息之后，处理器701还用于：
若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品的评分按照从高到低的排序排在前N位的商品，所述N为大于或者等于1的整数。
可选的，处理器701执行的所述将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，包括：
确定所述来访客户的商品推荐类型；
将所述评分矩阵记录所述商品推荐类型中的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品属于所述商品推荐类型中评分按照从高到低的排序排在前N位的商品。
可选的，处理器701执行的所述确定所述来访客户的商品推荐类型，包括：
接收语音信息，将与所述语音信息对应的商品推荐类型作为所述来访客户的商品推荐类型；或者
接收触摸输入，将所述触摸输入选择的商品推荐类型作为所述来访客户的商品推荐类型。
可选的，所述若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品之后，处理器701还用于：
若所述来访客户购买商品成功，则采集所述来访客户的身份信息，并建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系，其中，所述对应关系包括所述来访客户购买的商品的交易关系。
可选的，所述预先获取的多个商品的评分，包括：
预先根据消费者针对所述多个商品的购买记录确定的所述多个商品的评分。
需要说明的是，上述电子设备可以是线下商品门店的智能终端、手机、平板电脑等设备。
本发明实施例提供的电子设备能够实现图1和图2的方法实施例中的各个实施方式，以及相应有益效果，为避免重复，这里不再赘述。
本发明实施例还提供一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，该计算机程序被处理器701执行时实现本发明实施例提供的线下商品推荐方法的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存取存储器(Random Access Memory，简称RAM)等。
以上所揭露的仅为本发明较佳实施例而已，当然不能以此来限定本发明之权利范围，因此依本发明权利要求所作的等同变化，仍属本发明所涵盖的范围。

Claims (10)

  1. 一种线下商品推荐方法，其特征在于，包括：
    获取来访客户的视觉信息；
    从所述视觉信息中提取视觉特征信息；
    根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
    向所述来访客户展示所述推荐商品。
  2. 如权利要求1所述的方法，其特征在于，所述根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品的步骤，包括：
    在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息；
    若所述身份信息数据库存在所述目标身份信息，获取预先建立的评分矩阵中所述目标身份信息对多个商品的评分；
    根据所述目标身份信息对多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品。
  3. 如权利要求2所述的方法，其特征在于，所述在身份信息数据库中查询是否存在与所述视觉特征信息相匹配的目标身份信息之后，所述方法还包括：
    若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品的评分按照从高到低的排序排在前N位的商品，所述N为大于或者等于1的整数。
  4. 如权利要求3所述的方法，其特征在于，所述将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品，包括：
    确定所述来访客户的商品推荐类型；
    将所述评分矩阵记录所述商品推荐类型中的N个商品作为所述来访客户的推荐商品，其中，所述N个商品为所述多个商品属于所述商品推荐类型中评分按照从高到低的排序排在前N位的商品。
  5. 如权利要求4所述的方法，其特征在于，所述确定所述来访客户的商 品推荐类型，包括：
    接收语音信息，将与所述语音信息对应的商品推荐类型作为所述来访客户的商品推荐类型；或者
    接收触摸输入，将所述触摸输入选择的商品推荐类型作为所述来访客户的商品推荐类型。
  6. 如权利要求3所述的方法，其特征在于，所述若所述身份信息数据库不存在所述目标身份信息，将所述评分矩阵记录的N个商品作为所述来访客户的推荐商品之后，所述方法还包括：
    若所述来访客户购买商品成功，则采集所述来访客户的身份信息，并建立所述来访客户的身份信息与所述来访客户购买的商品的对应关系，其中，所述对应关系包括所述来访客户购买的商品的交易关系。
  7. 如权利要求1至6中任一项所述的方法，其特征在于，所述预先获取的多个商品的评分，包括：
    预先根据消费者针对所述多个商品的购买记录确定的所述多个商品的评分。
  8. 一种线下商品推荐装置，其特征在于，包括：
    获取模块，用于获取来访客户的视觉信息；
    提取模块，用于从所述视觉信息中提取视觉特征信息；
    确定模块，用于根据视觉特征信息，以及预先获取的多个商品的评分，从所述多个商品中确定所述来访客户的推荐商品；
    展示模块，用于向所述来访客户展示所述推荐商品。
  9. 一种电子设备，其特征在于，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1-7中任一项所述的线下商品推荐方法中的步骤。
  10. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至7中任一项所述的线下商品推荐方法中的步骤。
PCT/CN2018/124836 2018-10-10 2018-12-28 一种线下商品推荐方法、装置和电子设备 WO2020073524A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811179621.5 2018-10-10
CN201811179621.5A CN111027351B (zh) 2018-10-10 2018-10-10 一种线下商品推荐方法、装置和电子设备

Publications (1)

Publication Number Publication Date
WO2020073524A1 true WO2020073524A1 (zh) 2020-04-16

Family

ID=70164406

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/124836 WO2020073524A1 (zh) 2018-10-10 2018-12-28 一种线下商品推荐方法、装置和电子设备

Country Status (2)

Country Link
CN (1) CN111027351B (zh)
WO (1) WO2020073524A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111861670A (zh) * 2020-07-24 2020-10-30 深圳市爱深盈通信息技术有限公司 商品推荐方法、装置、设备及存储介质
CN111985997A (zh) * 2020-08-19 2020-11-24 济南浪潮高新科技投资发展有限公司 一种基于区块链技术的线下零售商品智能推荐方法及工具

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111612576A (zh) * 2020-05-09 2020-09-01 向培红 一种商品推荐方法、装置和电子设备
CN112036987B (zh) * 2020-09-11 2024-04-02 杭州海康威视数字技术股份有限公司 确定推荐商品的方法和装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103412937A (zh) * 2013-08-22 2013-11-27 成都数之联科技有限公司 一种基于手持终端的搜索购物方法
CN104112208A (zh) * 2014-03-11 2014-10-22 百度在线网络技术（北京）有限公司 电子商务中商品内容的提供方法、系统及装置
CN104298749A (zh) * 2014-10-14 2015-01-21 杭州淘淘搜科技有限公司 一种图像视觉和文本语义融合商品检索方法
CN106469184A (zh) * 2015-08-20 2017-03-01 阿里巴巴集团控股有限公司 数据对象标签处理、显示方法及服务器和客户端

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SG10201506111SA (en) * 2015-08-04 2017-03-30 Mastercard International Inc Method and system for providing a personalised customer service from a physical provider of goods or services
CN107247759A (zh) * 2017-05-31 2017-10-13 深圳正品创想科技有限公司 一种商品推荐方法及装置
CN108022152A (zh) * 2017-11-30 2018-05-11 北京长城华冠汽车技术开发有限公司 基于图像识别的用户商品自动推荐系统及推荐方法
CN108564414A (zh) * 2018-04-23 2018-09-21 帷幄匠心科技（杭州）有限公司 基于线下行为的商品推荐方法和系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103412937A (zh) * 2013-08-22 2013-11-27 成都数之联科技有限公司 一种基于手持终端的搜索购物方法
CN104112208A (zh) * 2014-03-11 2014-10-22 百度在线网络技术（北京）有限公司 电子商务中商品内容的提供方法、系统及装置
CN104298749A (zh) * 2014-10-14 2015-01-21 杭州淘淘搜科技有限公司 一种图像视觉和文本语义融合商品检索方法
CN106469184A (zh) * 2015-08-20 2017-03-01 阿里巴巴集团控股有限公司 数据对象标签处理、显示方法及服务器和客户端

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111861670A (zh) * 2020-07-24 2020-10-30 深圳市爱深盈通信息技术有限公司 商品推荐方法、装置、设备及存储介质
CN111985997A (zh) * 2020-08-19 2020-11-24 济南浪潮高新科技投资发展有限公司 一种基于区块链技术的线下零售商品智能推荐方法及工具

Also Published As

Publication number Publication date
CN111027351A (zh) 2020-04-17
CN111027351B (zh) 2023-10-13

Similar Documents

Publication Publication Date Title
US10580057B2 (en) Photorealistic recommendation of clothing and apparel based on detected web browser input and content tag analysis
WO2020073524A1 (zh) 一种线下商品推荐方法、装置和电子设备
US20220261862A1 (en) Systems and methods for pre-communicating shoppers' communication preferences to retailers
WO2020048084A1 (zh) 资源推荐方法、装置、计算机设备及计算机可读存储介质
US20170039628A1 (en) Image processing method and apparatus
US20130275270A1 (en) Method, web server and web browser of providing information
US10956928B2 (en) Cognitive fashion product advertisement system and method
US10672055B2 (en) Method and system for presenting personalized products based on digital signage for electronic commerce
US10083463B2 (en) Automating visual literacy to present data in a visually organized fashion
CN110766456A (zh) 商品推荐方法及装置
US9449025B1 (en) Determining similarity using human generated data
US20230274280A1 (en) Dynamically populated user interface feature
CN113689259A (zh) 基于用户行为的商品个性化推荐方法及系统
CN110503457A (zh) 用户满意度的分析方法及装置、存储介质、计算机设备
KR20200118615A (ko) 사용자에 의해 선택된 의류 이미지를 기반으로 사용자의 스타일 확인 기능을 제공하는 전자 단말 장치
CN107967637B (zh) 一种商品对象型号的推荐方法、装置及电子设备
JP2020013447A (ja) 決定装置、決定方法および決定プログラム
US11170428B2 (en) Method for generating priority data for products
AU2017258949A1 (en) Secondary market integration within existing data framework
CN116228342B (zh) 一种商品推荐方法、装置及计算机可读存储介质
WO2019192455A1 (zh) 门店系统、物品搭配方法、装置及电子设备
US10740815B2 (en) Searching device, searching method, recording medium, and program
US20230016483A1 (en) Browser plug-in for identification and display of alternative products for purchase
CN111127128B (zh) 商品推荐方法、装置及存储介质
CN112035624A (zh) 文本推荐方法和装置及存储介质

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: 18936321

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18936321

Country of ref document: EP

Kind code of ref document: A1