Commodity automatic matching method of e-commerce platform and e-commerce platform
By combining historical orders and personal information, e-commerce platforms recommend product specifications that best meet their needs to consumers, solving the problem of inaccurate recommendations in the existing technology, and achieving accurate matching of product specifications and optimization of stocking planning.
Patent Information
- Application Number
- CN202411285322.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-13
- Publication Date
- 2024-12-06
AI Technical Summary
The existing e-commerce platforms fail to effectively consider consumers' changing needs over time in product recommendations, resulting in inaccurate recommendations, which increases the difficulty of consumers in selecting products and the inadequate optimization of sellers' stockings.
By combining consumers' historical order information and personal information, considering the product's own attributes, using a matching algorithm to recommend product specifications that best meet their current needs to consumers, and optimizing stocking and warehousing planning.
It achieves accurate matching of product specifications, saves consumers' time to choose, improves purchasing intention, and optimizes seller stocking and warehousing management.
Smart Images

Figure CN119090595A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of this specification relate to the technical field of e-commerce data processing, and specifically to a method for automatically matching goods on an e-commerce platform. Background Art
[0002] At present, buying goods on various e-commerce platforms has become a common consumption method for people to meet their daily needs. Merchants usually set multiple attributes of one or more products for consumers to choose from. However, not all consumers are aware of the parameters of different attributes such as specifications and sizes of the products they need to buy, and it is easy to make wrong or random choices when selecting product specifications, which causes considerable trouble for consumers in purchasing goods. This not only wastes more time for consumers to purchase goods, but also reduces the probability of consumers purchasing high-quality, low-priced and expected goods to a certain extent, resulting in poor user shopping experience and even reducing the competitiveness of sellers' stores. Therefore, sellers on e-commerce platforms need to launch an automatic matching function for product specifications for each buyer, so as to save time in selection and comparison.
[0003] Similarly, for sellers, because they need to stock a large number of goods of different specifications, they need to bear the storage costs and also consider the express delivery costs. However, in terms of consumer distribution, consumers in different regions can summarize certain rules for their preferences for certain fixed specifications. For example, in terms of shoe size, skirt length, and style requirements, northern consumers prefer larger sizes, while southern consumers prefer smaller sizes. If we use an automatic matching method for each consumer and make full use of the characteristics of consumers in different regions, we can optimize the stocking warehouse, thereby reducing express delivery costs and shortening delivery time.
[0004] In the existing technical solutions, such as the product recommendation method, device and equipment of an e-commerce platform in publication number CN112232917B, ingredients are recommended based on the dishes selected by users in historical orders under different environments. However, it does not take into account that consumers will change over time. For example, when they bought clothes in the past, the consumers were still in the growth and development period or had a significant change in body shape. The new purchase demand will have different requirements for product size. It is obviously not accurate to recommend completely according to the specifications originally selected by the consumers. Summary of the invention
[0005] The embodiments of this specification provide a method for automatically matching goods on an e-commerce platform, which combines historical purchase information in historical orders with refund information, and cooperates with personal information to participate in the matching algorithm, and takes into account the attributes of the goods themselves, combines multiple factors with the goods that consumers currently need to purchase, and recommends matching items that best meet their current needs to consumers, so that: 1. The matched product specifications are accurate and practical, meet the needs of consumers, and save consumers' comparison time; 2. Product sellers can use the method of this embodiment, combined with historical orders and consumers' personal information, to preset items that consumers will match, and make stocking and warehousing plans in advance.
[0006] A first aspect of an embodiment of the present specification provides a method for automatically matching products on an e-commerce platform, comprising: upon receiving a matching request for a first product from a user, obtaining one or more first selection items of the first product, wherein the first selection item includes an item subject and multiple selection item items; Acquire user information of the user, determine one of the first selection items based on the user information as a recommended item of the first selection item, and push the recommended item to the user; wherein the user information includes historical purchase information of the user, the historical purchase information includes one or more first purchase items, and the first purchase item includes an item subject and a purchase item item; The determining, based on the user information, a selection item in the first selection item as a recommendation item of the first selection item, and pushing the recommendation item to the user includes determining, based on a purchase item item corresponding to the first purchase item of the first selection item in the user's historical purchase information, a selection item in the first selection item as a first matching item of the first selection item; generating a recommendation item of the first selection item based on the first matching item, and pushing the recommendation item to the user.
[0007] Further as a preferred embodiment, the purchase item item of the first purchase item and the selection item item of the first selection item have the same description type; and determining a selection item item in the first selection item as a first matching item of the first selection item based on the purchase item item of the first purchase item corresponding to the first selection item in the historical purchase information of the user includes determining a selection item item in the first selection item that is equal to the purchase item item of the first purchase item as the first matching item.
[0008] As a further preferred solution, one of the purchase item of the first purchase item and the selection item of the first selection item is a level description item and the other is a range value description item; The method of determining an item in the first selection item as a first matching item of the first selection item according to the item item of the first purchase item of the first selection item corresponding to the item item of the first purchase item in the historical purchase information of the user comprises determining the selection item item in the first selection item corresponding to the purchase item item of the first purchase item as the first matching item.
[0009] As a further preferred solution, the user information also includes the user's personal information, the personal information includes one or more user personal items, and the user personal items include item subjects and item values; The determining, based on the user information, one of the selection items in the first selection items as a recommended item of the first selection item, and pushing the recommended item to the user, further includes determining, based on an item value of a user personal item corresponding to the first selection item in the user's personal information, one of the selection items in the first selection items as a second matching item of the first selection item; The generating a recommendation item for the first selection item based on the first matching item and pushing the recommendation item to the user includes generating a recommendation item for the first selection item according to the first matching item and the second matching item, and pushing the recommendation item to the user.
[0010] As a further preferred solution, the first purchase item also includes item time, and the user personal item also includes item time; The method comprises generating a recommendation item for the first selected item based on the first matching item and the second matching item, and pushing the recommendation item to the user, including, when the first matching item and the second matching item are different, based on the entry time of a first purchase item corresponding to the first matching item and the entry time of a user personal item corresponding to the second matching item, taking the matching item corresponding to the item with a later entry time as the recommendation item.
[0011] Further as a preferred embodiment, the historical purchase information also includes one or more second purchase items, and the second purchase items include an item subject and an item item; after pushing the recommended item to the user, if there are one or more item items of the second purchase items in the second purchase items of the user that match the recommended item, pushing the item items corresponding to the second purchase items to the user.
[0012] As a further preferred solution, the selection item in the first selection item is a range value description type item; The personal information corresponds to the maximum value of the first selection item of the first selection item, and corresponds to the minimum value of the second selection item of the first selection item; The determining, according to the item value of the user personal item corresponding to the first selection item in the user's personal information, a selection item in the first selection item as a second matching item of the first selection item, comprises: If the elasticity value of the first product is greater than or equal to 0, the first selection item is used as the second matching item of the first selection item; If the elasticity value of the displayed first product is less than 0, the second selection item is used as the second matching item of the first selection item.
[0013] Further as a preferred solution, when a user's matching request for the first product is received, the second product is associated, and one or more second selection items of the second product are obtained, the second selection items including an item subject and multiple second item items; based on the user information, one second item item in the second selection item is determined as a recommended item of the second selection item, and the recommended item is pushed to the user.
[0014] As a further preferred embodiment, the one or more first purchase items also include corresponding historical delivery locations; the number of first purchase items corresponding to various purchase item items issued by each historical delivery location in the historical purchase information of all users is counted; and based on the number of first purchase items corresponding to each historical delivery location, the storage quantity required for the selected item items corresponding to the purchase item items in the warehouse where each historical delivery location is located is preset.
[0015] A second aspect of the embodiments of the present disclosure provides an e-commerce platform, which applies the automatic commodity matching method described in the first aspect.
[0016] The main beneficial effects of the above technical solution are: The matched items are accurate and practical, meeting the needs of consumers, saving consumers time in comparing different parameters of the same product. Sellers can use the method of this embodiment, combined with historical orders and consumers' personal information, to preset items that consumers will match, and prepare stock and warehouse planning in advance. When consumers demand the first product, a second product that meets their needs and has accurately matched items is pushed to increase consumers' willingness to buy.
[0017] Further or more detailed beneficial effects will be described in detail in conjunction with specific examples in the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The present invention will be further described below in conjunction with the accompanying drawings: Figure 1 A schematic diagram is shown of an exemplary system in which the various methods described herein may be implemented according to an embodiment of the present disclosure.
[0019] Figure 2 A schematic diagram of a process 100 according to an embodiment of the automatic commodity matching method disclosed in the present invention is shown.
[0020] Figure 3 A schematic diagram of a process 200 according to an embodiment of the method for automatic commodity matching of the present disclosure is shown.
[0021] Figure 4 A schematic diagram of a process 300 according to an embodiment of the method for automatic commodity matching of the present disclosure is shown.
[0022] Figure 5 A schematic diagram of a process 400 according to an embodiment of the method for automatic commodity matching of the present disclosure is shown.
[0023] Figure 6 A schematic diagram of a process 500 according to an embodiment of the method for automatic commodity matching of the present disclosure is shown.
[0024] Figure 7 A schematic diagram of a process 600 according to an embodiment of the method for automatic commodity matching of the present disclosure is shown. DETAILED DESCRIPTION
[0025] In order to enable those skilled in the art to better understand the present solution, the technical solution in the specific implementation mode will be clearly and completely described below in conjunction with the drawings in the implementation mode.
[0026] Figure 1 A schematic diagram of an exemplary system in which the methods and apparatus described herein may be implemented according to an embodiment of the present disclosure is shown. Figure 1 The system includes one or more client devices 1, a server 2, and one or more communication networks 11 coupling the one or more client devices to the server 2. The client device 1 can be configured to execute one or more applications. In an embodiment of the present disclosure, the server 2 can run to enable the automatic commodity matching method of the present embodiment to be executed. The user can use the client device 1 to perform human-computer interaction and make a matching request for the first commodity. The client device can provide an interface that enables the user of the client device to interact with the client device, and the client device can also output matching recommendation information to the user via the interface.
[0027] Specifically, in this embodiment, when the server 2 receives the user's matching request for the first product, it executes the product automatic matching method of this embodiment. The user's matching request for the first product is the user's willingness to purchase the first product. The first product represents a collection of the same product with different attributes and specifications. After the user puts forward the demand for the product collection, the user needs to select a product with specific attributes and specifications that is suitable for his or her own situation and needs from a large number of options. Therefore, it is necessary to simplify and further automate the selection step.
[0028] Among them, the first selection item is one or more stock keeping units (SKUs) of the first commodity, which jointly identify a certain commodity from one or more attributes. For example, if the first commodity is a top, the top is described from the attributes of the top, including color, size, etc., so as to specifically limit a certain type of top: a red top of size S. The information that the first selection item needs to contain includes an item subject and multiple selection item items under each item subject. The item subject is the attribute of the first commodity, such as size, color, and dimensions, and the item subject has multiple specific values to describe this attribute, that is, the selection item, such as red, blue, size S, and 10~20cm. The same item subject and different selection item items are different first selection items. In summary, for the first commodity, there may be multiple item subjects, each item subject has multiple selection item items, which are combined to form one or more first selection items of the first commodity.
[0029] In order to complete the matching and provide a corresponding reference for the server 2, the method of this embodiment needs to obtain the user information of the user through the client device 1. Based on the user information, one of the first selection items is determined as a recommended item of the first selection item, and the recommended item is pushed to the user, that is, the interface of the client device 1 displays the selected recommended item to the user.
[0030] like Figure 2 As shown, a process 100 of an embodiment of the commodity automatic matching method disclosed in the present invention includes the following steps: Step 101, obtain user information stored in server 2. Specifically, the user information includes the user's historical purchase information, and the historical purchase information includes one or more first purchase items, and the first purchase item includes an item subject and a purchase item item. The first purchase item is the specific information of the user's historical purchase, which is a selection record of the goods that have been purchased related to the first product. The record includes the item subject of the product when it was selected and the purchase item item selected under each item subject. For example, regarding the first product, a top, the first purchase item for purchasing the top can be found in the user information, and the first purchase item includes the item subject: color, size; and the specific purchase value corresponding to the item subject: the user once selected a red size S.
[0031] Step 102, based on the user's historical purchase information, determine the first purchase item corresponding to the first selection item. That is to say, when searching for a reference value in the historical purchase information, it is necessary to match the category and attribute of the first purchase item with the category and attribute of the first selection item that the current user is facing. For example, if the first selection item is a certain type of tops, the method of this embodiment retrieves the first purchase item that is also classified as tops in the historical purchase information. The item themes of the two items also need to correspond to each other, and find the purchase item items under each item theme of the first purchase item and use them as reference items.
[0032] Step 103, based on the purchase item item used as a reference, determine a selection item item of the first selection item as the first matching item of the first selection item. If the description type of the purchase item item of the first purchase item and the selection item item of the first selection item are the same, that is, both are level description type, or both are range value description type item items, then the selection item item is equivalent to the purchase item item. In some embodiments, the two are exactly the same, for example, if the purchase item item in the historical purchase information is size S under the item theme of size, then the selection item item also selects size S. In some embodiments, the purchase item item, that is, the selection item item, is a range value description type, and the range described by the purchase item item completely includes the range described by the selection item item. For example, if the purchase item item of the item theme size in the historical purchase information is 50~60cm, then the selection item item selects the same range or a smaller set within this range: 50~60cm, 50~55cm, 57~58cm, etc.
[0033] On the other hand, if the description types of the purchase item of the first purchase item and the selection item of the first selection item are different, that is, one of them is a level description type item and the other is a range value description type item. After determining the purchase item as the reference item, match the item corresponding to the selected purchase item in the selection item. Specifically, corresponding means that the two items correspond to each other after mathematical conversion or common sense conversion. For example, the purchase item is a level description type item under the item theme of size, specifically size S; in the description of the first purchase item, the specific range referred to by the multiple purchase item items under the item theme of size can be obtained, that is, size S refers to 50~60cm, which can be matched with the item of 50~60cm in the selection item or the item of a smaller range within this range.
[0034] Step 104: Generate a recommendation item for the first selection item based on the first matching item, and push the recommendation item to the user. The selected selection item is the first matching item, which can be presented to the user as a recommendation item, and the user can directly select the recommendation item to place an order.
[0035] It should be noted that using the first purchase item in the user's historical purchase information as a reference to achieve automatic product matching is only one implementation path of the method of this embodiment. The following shows the process of another embodiment of the automatic product matching method 200 according to the present disclosure.
[0036] The process of the image generation method includes the following steps: Step 201, obtain user information stored in the server 2. The user information also includes the user's personal information, which includes one or more user personal items, and the user personal items include item topics and item values. Specifically, the item topics of the user's personal items include height, waist circumference, chest circumference, etc., and the item values are specific values of each evaluation dimension, and the item values are exact numerical description types.
[0037] Step 202, according to the item value of the user's personal item corresponding to the first selection item in the user's personal information, determine a selection item in the first selection item as the second matching item of the first selection item. Specifically, when searching for the reference value in the historical purchase information, match the item subject of the user's personal item with the item subject of the first selection item that the current user is facing, for example, match the waist circumference in the personal information with multiple selection items under the item subject of size in the first selection item; during the matching process, the method of this embodiment considers common sense conversion and the detailed description of the crawled page, and performs formal conversion. In some embodiments, if the waist circumference in the personal information is 65cm, then according to common sense conversion or formal conversion, select the selection item of different description types such as M code (applicable to waist circumference of 60~70cm in the detailed description), 60~70cm, etc., and use the selection item as the second matching item.
[0038] Step 203: Generate a recommendation item for the first selection item based on the second matching item, and push the recommendation item to the user.
[0039] The process 200 in the above embodiment matches the first selection item with the first purchase item and the user's personal item respectively, and the two matching methods are performed in parallel to obtain the first matching item and the second matching item respectively. When the two are consistent, that is, the first matching item and the second matching item are the same selection item, then the selection item is a recommended item pushed to the customer. In contrast, when the first matching item and the second matching item are different selection items, the method of this embodiment needs to make further selections. Specifically, the first purchase item also includes the entry time, and the user's personal item also includes the entry time. The method of this embodiment obtains the entry time when obtaining user information, that is, the time when the first purchase item corresponding to the first matching item is generated and the time when the user's personal item corresponding to the second matching item is updated. The two entry times are compared, and the matching item corresponding to the entry with a later entry time is used as a recommended item. That is to say, the later the entry time, the more timely the information update in the item represented, and the closer it is to the user's current state, and the higher the accuracy of the generated recommended item.
[0040] Furthermore, when the recommended items matched by the above two processes are clearly refunded by the user and are highly likely not suitable for recommendation again, it is necessary to determine whether to continue to push them to the user. Therefore, process 300 of another embodiment of the automatic product matching method disclosed in the present invention includes the following steps: Step 301, obtaining user information stored in server 2.
[0041] Step 302: Generate a recommendation item for the first selection item according to process 100 or process 200. In this embodiment, the automatic product matching method can only execute process 100 or only execute other processes to obtain the recommended items of the product. Alternatively, the two processes can be executed simultaneously to obtain the final recommended items after comparison.
[0042] Step 303: determine whether to push the recommended item to the user.
[0043] Specifically, the historical purchase information also includes one or more second purchase items, and the second purchase item includes an item subject and an item item. The second purchase item is a purchase item that the user has returned, and the information contained in the second purchase item also includes the item subject and its corresponding item item that has been selected but returned. If one or more item items of the second purchase item in the second purchase item of the user match the recommended item, the recommended item will not be pushed, but the item item corresponding to the second purchase item will be pushed to the user. In other words, when the matched recommended item is an item item in the second purchase item that has been returned, the recommended item will not be pushed to the user, but the item item corresponding to the second purchase item will be pushed to remind the user that the option has been returned. In some embodiments, the information of the entire second purchase item can be pushed to the user at the same time, including the entry time and the detailed description of the corresponding item item, so as to better assist the user in making a judgment.
[0044] Furthermore, a process 400 of another embodiment of the commodity automatic matching method disclosed herein includes the following steps: Step 401, obtaining user information stored in server 2.
[0045] Step 402: Determine, based on the user's personal information, the item subject of the first selected item corresponding to the item subject of the user's personal item.
[0046] Step 403: Based on the personal information and the properties of the first commodity, determine one of the first selection items as a second matching item of the first selection item.
[0047] Specifically, in this embodiment, the selection item in the first selection item is a range value description type item. If the item value of the user's personal item is the minimum value of the range described by one selection item in the first selection item, and is also the maximum value of the range described by another selection item. According to the properties of the first commodity, one of the two selection items is determined to be the second matching item. Specifically, if the item value of the user's personal item is the minimum value of the description range of the first selection item in the first selection item, according to the properties of the first commodity, either the first selection item is used as the second matching item of the first selection item, or the second selection item is used as the second matching item of the first selection item. The second selection item is a selection item whose range value is smaller than or larger than the range value of the first selection item and is adjacent to the first selection item.
[0048] Specifically, the selection item of the first selection item is an option described by multiple range values, and the user needs to select one of the ranges that suits his or her own situation. When selecting the second matching item, by matching the entry value in the personal information with the various range values of the selection item, if the entry value falls within the range of one of the selection item, the selection item is the second matching item. In this process, the entry value of the personal information may be exactly at the extreme value of the first selection item range and the second selection item adjacent to the first selection item. For example, the entry value is the maximum value of the first selection item and the minimum value of the second selection item, or the entry value is the minimum value of the first selection item and the maximum value of the second selection item. It can be determined based on the material characteristics of the first product to select one of the two selection items.
[0049] The above step 403 is illustrated as follows: For example, the first product is a top, which includes sizes S (weight 50-55KG), M (weight 55-60KG), L (weight 60-65KG), and XL (weight 65-70KG). The entry value of the personal information is weight 60KG, which is both the maximum value of the M size and the minimum value of the L size. If the properties of the first product include an elasticity value indicating the elasticity of the fabric of the product (an elasticity value of 1 indicates maximum elasticity, an elasticity value of 0 indicates low elasticity, and an elasticity value of -1 indicates easy shrinkage), if the elasticity value of the top is 1, it indicates that the fabric of the top is a synthetic material with relatively large elasticity, and the first selected item M (weight 55-60KG) is used as the second matching item (at this time, the user's weight of 60KG is the maximum value of the range of this item); if the elasticity value of the top is -1, it indicates that the fabric of the top is cotton material that is easy to shrink, and the second selected item L (weight 60-65KG) with a larger range value is used as the second matching item (at this time, the user's weight of 60KG is the minimum value of the range of this item); if the elasticity value of the top is 0, it indicates that the fabric of the top is a fabric with relatively small elasticity, and the first selected item M (weight 55-60KG) is used as the second matching item (at this time, the user's weight of 60KG is the maximum value of the range of this item.
[0050] For another example, the first product is a dress, which includes sizes S (waist 65-68cm), M (waist 68-72cm), and L (waist 72-75cm). The entry value of personal information is waist 68cm, which is both the maximum value of size S and the minimum value of size M. If the elasticity value of the dress is greater than or equal to 0, it means that the fabric of the dress is elastic, and the first selection item S (waist 65-68cm) is used as the second matching item; if the elasticity value of the dress is less than 0, it means that the fabric of the dress is cotton material that is easy to shrink, and the second selection item M (waist 68-72cm) with a larger range value is selected as the second matching item.
[0051] Step 404: Generate a recommendation item for the first selection item based on the second matching item, and push the recommendation item to the user. The recommendation item is a personalized recommendation item that fully considers the properties of each product and is more in line with the user's experience.
[0052] Furthermore, a process 500 of another embodiment of the commodity automatic matching method disclosed herein includes the following steps: Step 501, obtaining user information stored in server 2.
[0053] Step 502: determine a recommended item for the first selection item according to all the above processes, and push the recommended item to the user.
[0054] Step 503: associate a second product with the first product and its recommended items.
[0055] Specifically, in this embodiment, upon receiving a matching request for the first product from the user, the method further associates the second product. The second product is an object that is associated with the first product and can be sold in combination. For example, if the first product is casual pants, the second product can be associated with pantyhose for the user. Obtain one or more second selection items for the second product, the second selection item includes an item theme and multiple second item items; based on the user information, determine a second item item in the second selection item as a recommended item for the second selection item, and push the recommended item of the second product to the user. In other words, based on the existing user information, the second product can be pre-matched with items suitable for the user, and the second product item items that have been matched to the user's historical purchase information and personal information can be directly pushed to the user, saving selection time and increasing the user's willingness to buy.
[0056] Furthermore, a process 600 of another embodiment of the commodity automatic matching method disclosed herein includes the following steps: Step 601, obtaining user information stored in server 2.
[0057] Step 602 , counting the delivery location information of each historical purchase information in all the information stored in the server 2 .
[0058] Specifically, the one or more first purchase items also include corresponding historical distribution locations: for example, the eastern coastal area, the northwestern plateau area, etc. The number of first purchase items corresponding to various purchase items issued by each historical distribution location is counted; by counting and integrating the number of each selection item under an item theme of each historical distribution location, it can be known in which regions the sales volume of a specific selection item of the first commodity is the best. According to the statistical results of the sales volume, the required storage quantity of the selection item corresponding to the purchase item in the warehouse where each historical distribution location is located is preset. In other words, under the premise of knowing which selection item has the highest sales volume under the same item theme for each historical distribution location, a larger number of such selection items are pre-equipped in the warehouse at the corresponding location. After applying the method of this embodiment, users in the area have a higher probability of being pushed and placing orders for such selection items, thereby cooperating with the warehousing planning.
[0059] Similarly, in step 603, statistics are collected on the historical purchase information of each time period in the information stored in the server 2.
[0060] Specifically, the entry time is classified into multiple time periods: for example, spring, autumn, summer, etc. The number of first purchase items corresponding to various purchase items issued in each time period is counted in the information stored in the entire server 2, thereby obtaining the sales volume change and trend of each selection item for a certain item theme in each time period; the required storage quantity of the selection item corresponding to each purchase item in each time period is preset according to the quantity, and the planned storage quantity can meet the matching needs of the user in the season, so that the storage of each selection item is more flexible under the premise of meeting the demand trend.
[0061] The beneficial effects that this embodiment may produce include: 1. The matched items are accurate and practical, meeting the needs of consumers and saving consumers time in comparing different parameters of the same product.
[0062] 2. The seller can use the method of this embodiment, combined with historical orders and consumers' personal information, to preset items that consumers will match, and prepare stock and storage plans in advance.
[0063] 3. When consumers demand the first product, push the second product that meets their needs and accurately matches the items, thereby increasing consumers' willingness to buy.
[0064] An embodiment of the present disclosure further provides an e-commerce platform, which adopts the automatic product matching method described in any one of the above embodiments to recommend products to users.
[0065] The above description is only a preferred embodiment of the present invention and does not limit the scope of the present invention. In addition, the terms "vertical", "lateral", "front", "rear" and the like mentioned in the embodiments of the present invention indicate the orientation or position relationship based on the orientation or position relationship shown in the drawings, or the orientation or position relationship in which the product is usually placed when in use, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be further explained that, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", "fix" and the like in the description should be understood in a broad sense, for example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected, or indirectly connected through an intermediate medium, or the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.
[0066] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.
Claims
1. A method for automatically matching goods on an e-commerce platform, characterized in that: include: When receiving a matching request for a first product from a user, obtaining one or more first selection items of the first product, wherein the first selection items include an item subject and a plurality of selection item items; Acquire user information of the user, determine one of the first selection items based on the user information as a recommended item of the first selection item, and push the recommended item to the user; wherein the user information includes historical purchase information of the user, the historical purchase information includes one or more first purchase items, and the first purchase item includes an item subject and a purchase item item; The determining, based on the user information, one of the first selection items as a recommended item of the first selection item, and pushing the recommended item to the user includes: According to the purchase item item corresponding to the first purchase item of the first selection item in the historical purchase information of the user, determining a selection item item in the first selection item as a first matching item of the first selection item; Based on the first matching item, a recommendation item of the first selection item is generated, and the recommendation item is pushed to the user.
2. The method for automatically matching goods on an e-commerce platform according to claim 1, characterized in that: The purchase item item of the first purchase item and the selection item item of the first selection item have the same description type; The step of determining, according to the purchase item item corresponding to the first purchase item of the first selection item in the historical purchase information of the user, a selection item item in the first selection item as a first matching item of the first selection item, comprises: A selection item item in the first selection item that is equal to a purchase item item in the first purchase item is determined as a first matching item.
3. The method for automatically matching goods on an e-commerce platform according to claim 1, characterized in that: One of the purchase item item of the first purchase item and the selection item item of the first selection item is a level description item and the other is a range value description item; The determining, according to the item item of the first purchase item of the first selection item corresponding to the item item of the first selection item in the historical purchase information of the user, an item item in the first selection item as a first matching item of the first selection item comprises: A selection item item in the first selection item corresponding to a purchase item item of the first purchase item is determined as a first matching item.
4. The method for automatically matching goods on an e-commerce platform according to claim 1, characterized in that: The user information also includes the user's personal information, which includes one or more user personal entries, and the user personal entries include entry subjects and entry values; The determining, based on the user information, one of the first selection items as a recommended item of the first selection item, and pushing the recommended item to the user, further includes: determining, according to an item value of a user personal item corresponding to the first selection item in the user's personal information, a selection item item in the first selection item as a second matching item of the first selection item; The step of generating a recommendation item for the first selection item based on the first matching item, and pushing the recommendation item to the user includes: A recommendation item of the first selection item is generated according to the first matching item and the second matching item, and the recommendation item is pushed to the user.
5. The method for automatically matching goods on an e-commerce platform according to claim 4, characterized in that: The first purchase item also includes the item time, and the user personal item also includes the item time; The step of generating a recommendation item for the first selection item according to the first matching item and the second matching item, and pushing the recommendation item to the user includes: When the first matching item and the second matching item are different, based on the entry time of the first purchase item corresponding to the first matching item and the entry time of the user's personal item corresponding to the second matching item, the matching item corresponding to the entry with the later entry time is used as the recommended item.
6. The method for automatically matching goods on an e-commerce platform according to claim 1, characterized in that: The historical purchase information further includes one or more second purchase items, wherein the second purchase items include an item subject and an item item; After pushing the recommendation item to the user, the method further includes: If one or more items of the second purchase items in the second purchase items of the user match the recommended item, the item corresponding to the second purchase item is pushed to the user.
7. The method for automatically matching goods on an e-commerce platform according to claim 4, characterized in that: The selection item in the first selection item is a range value description type item; The personal information corresponds to the maximum value of the first selection item of the first selection item, and corresponds to the minimum value of the second selection item of the first selection item; The determining, according to the item value of the user personal item corresponding to the first selection item in the user's personal information, a selection item in the first selection item as a second matching item of the first selection item, comprises: If the elasticity value of the first product is greater than or equal to 0, the first selection item is used as the second matching item of the first selection item; If the elasticity value of the displayed first product is less than 0, the second selection item is used as the second matching item of the first selection item.
8. The method for automatically matching goods on an e-commerce platform according to claim 1, characterized in that: When receiving a matching request for a first product from a user, associating the second product with the second product, and obtaining one or more second selection items of the second product, wherein the second selection items include an item subject and a plurality of second item items; Based on the user information, a second item in the second selection items is determined as a recommended item of the second selection item, and the recommended item is pushed to the user.
9. The method for automatic product matching of an e-commerce platform according to claim 1, characterized in that: Also includes: The one or more first purchase items also include corresponding historical delivery locations; Counting the number of first purchase items corresponding to various purchase items issued by each historical delivery location in the historical purchase information of all users; According to the quantity of the first purchase items corresponding to each historical delivery location, the storage quantity required for the selection item corresponding to the purchase item in the warehouse where each historical delivery location is located is preset.
10. An e-commerce platform, characterized in that: The method for automatic commodity matching according to any one of claims 1 to 9 is applied.
Citation Information
Patent Citations
Product recommendation methods, devices and equipment for e-commerce platforms
CN112232917B
