Commodity recommendation method and device, storage medium and computer equipment

By analyzing users' purchase history to determine the preferred prices of product categories, calculating price deviations and adjusting product weights, this technology solves the problem of failing to provide personalized recommendations in existing technologies, achieves a balance between rationality and diversity, and improves recommendation effectiveness.

CN118917927APending Publication Date: 2024-11-08VIPSHOP (GUANGZHOU) SOFTWARE CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411191073.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing product recommendation methods fail to personalize recommendations based on users' price preferences, leading to frequent price reductions and promotions that increase operating costs and weaken brand value.

Method used

By analyzing users' target purchase history, we determine the users' preferred prices for target product categories, calculate the price deviation, and determine the initial weighting weight of each recalled product based on the price deviation. We then adjust the target weighting weight according to preset upper and lower limits, and finally sort and recommend products based on the target weighting weight.

Benefits of technology

It enables personalized recommendations based on user price preferences, avoiding excessively high or low pricing, improving the rationality of recommendations and user satisfaction, and balancing the diversity of recommendation results with the satisfaction of user preferences.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917927A_ABST
    Figure CN118917927A_ABST
Patent Text Reader

Abstract

According to the commodity recommendation method and device, the storage medium and the computer equipment provided by the invention, in response to a commodity access request of a user, a target commodity category corresponding to the commodity access request is determined, and a target purchase history of the user under the target commodity category is acquired; according to the target purchase history, determining a preference price of the user for the target commodity category, and according to the preference price, calculating a price deviation of each recalled commodity in the target commodity category; determining an initial weighted weight of each recalled commodity according to the price deviation, and determining a target weighted weight of each recalled commodity based on each initial weighted weight according to a preset upper limit score and a preset lower limit score; and sorting each recalled commodity according to each target weighted weight, and pushing the sorted recalled commodities to the user. Therefore, personalized recommendation can be carried out based on the preference of the user to the commodity price.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a product recommendation method, device, storage medium and computer equipment. Background Art

[0002] In today's increasingly competitive e-commerce industry, many platforms have adopted price cuts and other means to attract consumers. However, this short-term price competition strategy is not sustainable. On the one hand, frequent price cuts and promotions will not only increase the operating costs of the platform, but may also weaken the brand value and profit margins of the product. On the other hand, consumers have gradually become numb to conventional price cuts and promotions, and they have begun to pay more attention to product quality, thoughtfulness and personalization of services. Therefore, although the existing products attract users based on price, they do not make personalized recommendations based on users' preferences for product prices. Summary of the invention

[0003] The purpose of this application is to solve at least one of the above-mentioned technical defects, especially the technical defect that the existing products in the prior art attract users based on price but do not make personalized recommendations based on users' preferences for product prices.

[0004] In a first aspect, the present application provides a product recommendation method, the method comprising:

[0005] In response to a user's product access request, determine a target product category corresponding to the product access request, and obtain the user's target purchase history under the target product category;

[0006] Determine the user's preferred price for the target product category based on the target purchase history, and calculate the price deviation of each recalled product in the target product category based on the preferred price;

[0007] Determine an initial weighted weight for each recalled product according to the price deviation, and determine a target weighted weight for each recalled product based on each initial weighted weight and according to a preset upper limit score and a preset lower limit score;

[0008] According to the weighted weights of each target, each recalled product is sorted and the sorted recalled products are pushed to the user.

[0009] In one embodiment, the step of determining the user's preferred price for the target product category based on the target purchase history includes:

[0010] When it is determined based on the target purchase history that the user has purchased products in the target product category, the prices of each product purchased by the user in the target product category are obtained from the target purchase history;

[0011] The average price of each purchased product under the target product category is used as the preferred price.

[0012] In one embodiment, the step of determining the user's preferred price for the target product category based on the target purchase history further includes:

[0013] When it is determined based on the target purchase history that the user has not purchased a product in the target product category, the user's purchase history in each product category of the same level is obtained, where the product category of the same level is the product category of the same level as the target product category;

[0014] From the same-level purchase history, obtain the prices of each purchased product of the user under the same-level product category, and use the average price corresponding to each purchased product under the same-level product category as the preferred price.

[0015] In one embodiment, the step of calculating the price deviation of each recalled product in the target product category according to the preferred price includes:

[0016] For each recalled product, the corresponding price deviation is calculated according to the following expression:

[0017]

[0018] in, is the price deviation, The price of the recalled product. For preferred price.

[0019] In one embodiment, the step of determining an initial weight of each recalled product according to the price deviation includes:

[0020] For each recalled product, the corresponding initial weight is calculated according to the following expression:

[0021]

[0022] in, is the initial weight, is the price deviation, , and All are preset constants.

[0023] In one embodiment, the step of determining a target weight for each recalled product based on each initial weighted weight according to a preset upper limit score and a preset lower limit score includes:

[0024] When the number of purchased items corresponding to the preferred price determined based on the target purchase history is less than the preset threshold, each initial weighted weight is corrected according to the following expression:

[0025]

[0026] in, is the modified initial weight, is the uncorrected initial weight, m ​​is the number of purchased goods, and n is the preset threshold;

[0027] According to the preset upper limit score and the preset lower limit score, the revised initial weighted weights are constrained to obtain the target weighted weight of each recalled product.

[0028] In one embodiment, the step of constraining each modified initial weighted weight according to a preset upper limit score and a preset lower limit score to obtain a target weighted weight for each recalled product includes:

[0029] For each recalled product, the corresponding target weight is calculated according to the following expression:

[0030]

[0031] in, Weight the target, is the preset upper limit score, is the preset lower limit score, is the modified initial weight.

[0032] In a second aspect, the present application provides a product recommendation device, the device comprising:

[0033] A product access request response module, used to respond to a user's product access request, determine the target product category corresponding to the product access request, and obtain the user's target purchase history under the target product category;

[0034] a price deviation calculation module, for determining the user's preferred price for a target product category based on the target purchase history, and calculating the price deviation of each recalled product in the target product category based on the preferred price;

[0035] A target weighted weight calculation module, used to determine the initial weighted weight of each recalled product according to the price deviation, and based on each initial weighted weight, determine the target weighted weight of each recalled product according to a preset upper limit score and a preset lower limit score;

[0036] The product push module is used to sort each recalled product according to the weighted weight of each target, and push the sorted recalled products to the user.

[0037] In a third aspect, the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the product recommendation method in any of the above embodiments.

[0038] In a fourth aspect, the present application provides a computer device, comprising: one or more processors, and a memory;

[0039] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the steps of the product recommendation method in any of the above embodiments are performed.

[0040] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0041] In the product recommendation method, device, storage medium and computer device provided in the present application, in response to a user's product access request, the target product category corresponding to the product access request is determined, and the user's target purchase history under the target product category is obtained; based on the target purchase history, the user's preferred price for the target product category is determined, so that the price range that the user is more inclined to purchase can be analyzed, and the price deviation of each recalled product in the target product category is calculated based on the preferred price, which helps to quantify the price rationality of the product in the user's mind and avoid overly or underpriced recommendations; the initial weighted weight of each recalled product is determined based on the price deviation to ensure that products with prices that meet user expectations can obtain a higher recommendation weight, and based on each initial weighted weight, the target weighted weight of each recalled product is determined according to a preset upper limit score and a preset lower limit score, which helps to balance the diversity of recommendation results and the satisfaction of user preferences; according to each target weighted weight, each recalled product is sorted, and the sorted recalled products are pushed to the user, and the sorted products will be pushed to the user in the order of recommendation weight to personalize the recommended products. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0043] Figure 1 A flowchart of a product recommendation method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a commodity recommendation device provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0047] In one embodiment, the present application provides a product recommendation method. The following embodiments are described by taking the method applied to a computer device as an example. It can be understood that the computer device can be any device with data processing functions, including but not limited to a single server, a server cluster, a personal laptop computer, a desktop computer, etc. Figure 1 As shown, the method includes:

[0048] S101: In response to a user's product access request, determine a target product category corresponding to the product access request, and obtain a target purchase history of the user under the target product category.

[0049] Among them, user product access requests refer to the behavior of users clicking or searching for a certain category of products on an e-commerce platform. The target product category is the category or type of product to which the user requests access. For example, if the user visits the sports shoes page, the target product category can be "sports shoes". The target purchase history refers to the user's purchase record or purchase preference under a specific product category. This history may include related products that the user has purchased in the past, as well as their interest and purchase potential in other products under the category.

[0050] In this step, the user's access request to a specific product category is identified by monitoring user behavior, such as clicks and searches. Once the user's access request is identified, the specific product category that the user is interested in can be determined by analyzing the pages the user clicks on, search keywords, and other information. After confirming the target product category, the user's purchase history under the category is retrieved and analyzed, which can include information such as the types of products the user has purchased in the past and brand preferences.

[0051] S102: Determine the user's preferred price for the target product category based on the target purchase history, and calculate the price deviation of each recalled product in the target product category based on the preferred price.

[0052] Among them, the preferred price refers to the user's price preference range or preference level under a certain target product category. For example, a user may prefer to buy electronic products between $200 and $300, and is not so interested in products that are outside or below this price range. Price deviation represents the difference between the actual price of each recalled product and the user's preferred price. By calculating the price deviation, the consistency of each product with the user's expectations in price can be evaluated, thereby affecting its weight and ranking in personalized recommendations.

[0053] In this step, purchase records related to the target product category are extracted from the user's target purchase history. For example, analysis and extraction can be performed through the database of the e-commerce platform or the user's shopping history data. Analyze the user's purchase history under the target product category, especially focusing on the price range of the purchase. For example, the user's average purchase price or the most frequent price segment under the category can be calculated, or the user's purchase frequency and amount under different price ranges can be analyzed to find the user's preferred price range. For each recalled product, such as the candidate product selected in the recommendation system, calculate the difference between its actual price and the user's preferred price. For example, the absolute value of the difference between the product price and the user's preferred price can be directly calculated, or the percentage deviation of the product price relative to the user's preferred price can be calculated to better reflect the size of the relative deviation.

[0054] S103: determining an initial weight of each recalled product according to the price deviation, and determining a target weight of each recalled product based on each initial weight, according to a preset upper limit score and a preset lower limit score.

[0055] The preset upper limit score and the preset lower limit score are used to determine the range of the target weighting weight for each recalled product. These scores are usually pre-set parameters to ensure the balance and rationality of the recommendation results.

[0056] In this step, the price deviation is used to calculate the initial weighted weight of each recalled product. For example, the price deviation can be converted into a score or a rating, that is, the price deviation is mapped to a specific range, such as between 0 and 1, through a linear or nonlinear function. For each product, according to its initial weighted weight, it is adjusted to between a preset upper and lower score. In one example, the initial weighted weight of each product is standardized to a standard range, such as between 0 and 1; a mapping function is used to map the standardized weight to between a preset upper and lower score, which can be a linear mapping or a nonlinear mapping; finally, ensure that the target weighted weights of all products meet the preset score upper and lower limits, while considering the overall balance of the recommendation system and user experience.

[0057] S104: sorting each recalled product according to the weighted weights of each target, and pushing the sorted recalled products to the user.

[0058] In this step, for each product, its target weighted weight is combined with other relevant attributes of the product, such as refined ranking, click-through rate, add-to-cart rate, conversion rate and popularity, to calculate a comprehensive weighted score. Specifically, it can be achieved through a simple weighted summation formula, such as comprehensive weighted score = (1 + target weighted weight) Weight adjustment factor, where the weight adjustment factor can be the click rate and popularity of the product, etc., which is used to adjust the influence of the target weighted weight.

[0059] All products are sorted in descending order according to the calculated comprehensive weighted scores, which means that the products with the highest scores are at the front and the products with the lowest scores are at the back. According to the sorted list, according to the first N products that the user needs to see, the first N products in the list are selected as the recommended results and pushed to the user. The push method can be displayed on the web page, push notifications in mobile applications, and emails.

[0060] In the above embodiment, in response to a user's product access request, the target product category corresponding to the product access request is determined, and the user's target purchase history under the target product category is obtained; based on the target purchase history, the user's preferred price for the target product category is determined, so that the price range that the user is more inclined to purchase can be analyzed, and the price deviation of each recalled product in the target product category is calculated based on the preferred price, which helps to quantify the price rationality of the product in the user's mind and avoid overly high or low pricing recommendations; the initial weighted weight of each recalled product is determined based on the price deviation to ensure that products with prices that meet user expectations can obtain higher recommendation weights, and based on each initial weighted weight, the target weighted weight of each recalled product is determined according to a preset upper limit score and a preset lower limit score, which helps to balance the diversity of recommendation results and the satisfaction of user preferences; according to each target weighted weight, each recalled product is sorted, and the sorted recalled products are pushed to the user, and the sorted products will be pushed to the user in the order of recommendation weights to personalize the recommended products.

[0061] In one embodiment, the step of determining the user's preferred price for the target product category based on the target purchase history includes:

[0062] When it is determined based on the target purchase history that the user has purchased products in the target product category, the prices of each product purchased by the user in the target product category are obtained from the target purchase history;

[0063] The average price of each purchased product under the target product category is used as the preferred price.

[0064] Specifically, first, it is necessary to analyze the user's purchase history data, especially the purchase records for the target product category, which usually includes information such as the purchased product ID, category and price. According to the defined target product category, filter out all the products that the user has purchased that belong to the category. It can be understood that various products on various platforms basically have similar classification methods, which are usually divided into primary categories, secondary categories and tertiary categories. For example: men's clothing, women's clothing, underwear, etc. can be used as primary categories, and each primary category can be subdivided into several secondary categories. Women's tops and bottoms can be used as secondary categories under the primary category of women's clothing. Each secondary category can be subdivided into several tertiary categories. Women's coats, women's shirts, etc. can be used as tertiary categories under women's tops. Based on this, if the user wants to purchase goods under a certain tertiary category and is an old customer of the tertiary category, the user's purchased goods under the tertiary category are filtered out.

[0065] For each purchased item belonging to the target product category, obtain its price data from the purchase history. This price is usually the actual price at the time of purchase, which can be the original price or the discounted price. Average the prices of all purchased items to obtain the user's preferred price under the target product category.

[0066] In an example, within n days of historical statistics, a user purchased m items under the third-level category a, with prices of price1, price2, … price m , take the average price of the m items purchased by the user in the third-level category as the user's preferred price in the third-level category. The specific calculation method is:

[0067]

[0068] In another example, a user purchased a women's coat for RMB 282, a women's coat for RMB 216, and a women's coat for RMB 248 within 180 days. The user's preferred price in the third-level category of women's coats is: (282+216+248) / 3=248.67 yuan.

[0069] In this embodiment, taking the user's preferred price into consideration can avoid recommending overpriced or underpriced products to the user. By understanding the user's price preference in a specific category, the recommendation system can more accurately meet the user's purchasing intention and enhance the personalized experience.

[0070] In one embodiment, the step of determining the user's preferred price for the target product category based on the target purchase history further includes:

[0071] When it is determined based on the target purchase history that the user has not purchased a product in the target product category, the user's purchase history in each product category of the same level is obtained, where the product category of the same level is the product category of the same level as the target product category;

[0072] From the same-level purchase history, obtain the prices of each purchased product of the user under the same-level product category, and use the average price corresponding to each purchased product under the same-level product category as the preferred price.

[0073] Specifically, when a user has not purchased a product in the target product category, it is necessary to use the user's purchase history in the same product category to infer his or her preferred price. First, traverse the user's purchase history in the same product category, that is, other categories at the same level as the target product category. These records should include information such as the purchased product ID, category, and price. For each product that has been purchased in the same product category, obtain its price data. Average the prices of all purchased products to obtain the user's preferred price in the same product category.

[0074] In one example, if a user has not purchased a product in a certain third-level category, the user is called a new customer in this category. The average price of all third-level categories purchased by the user is calculated as the user's preferred price in this third-level category. The specific calculation method is:

[0075]

[0076] In another example, a user's preferred price for tertiary category A is 180 yuan, the preferred price for tertiary category B is 200 yuan, and the preferred price for tertiary category C is 220 yuan. The user has never shopped in tertiary category D. In this case, the user is a new customer for tertiary category D. The user's preferred price for tertiary category D is (180+200+220) / 3=200 yuan.

[0077] In this embodiment, by comprehensively considering the user's purchase history in the target product category and product categories of the same level, the user's price preference can be predicted more accurately, thereby providing product recommendations that better meet the user's expectations.

[0078] In one embodiment, the step of calculating the price deviation of each recalled product in the target product category according to the preferred price includes:

[0079] For each recalled product, the corresponding price deviation is calculated according to the following expression:

[0080]

[0081] in, is the price deviation, The price of the recalled product. For preferred price.

[0082] Specifically, first, the information of all recalled commodities in the target commodity category is obtained, including the price of each commodity, and then the expression in this embodiment is used to calculate the price deviation corresponding to each recalled commodity.

[0083] In this embodiment, by considering the price preference of the user, commodities can be recommended more accurately according to the budget and preference of the user.

[0084] In one embodiment, the step of determining the initial weight of each recalled product according to the price deviation includes:

[0085] For each recalled product, the corresponding initial weight is calculated according to the following expression:

[0086]

[0087] in, is the initial weight, is the price deviation, , and All are preset constants.

[0088] It is understandable that according to the data analysis of historical data, when the price increases by more than 10% or the price decreases by more than 15%, the user's purchasing desire will decrease significantly. Therefore, a piecewise function is designed for the price deviation as the weighted weight. Among them, a=-0.15, b=0.1, k=0.1. Different platforms can pull historical data for analysis. , and Adjust parameters to obtain the weighted impact of price on ranking that is most suitable for users on each platform.

[0089] In this embodiment, considering the price deviation can more accurately reflect the user's sensitivity and preference for price, and converting the price deviation into an initial weighted weight helps to improve the accuracy and effect of product recommendation.

[0090] In one embodiment, based on each initial weighted weight, according to a preset upper limit score and a preset lower limit score, the step of determining a target weighted weight for each recalled product includes:

[0091] When the number of purchased items corresponding to the preferred price determined based on the target purchase history is less than the preset threshold, each initial weighted weight is corrected according to the following expression:

[0092]

[0093] in, is the modified initial weight, is the uncorrected initial weight, m ​​is the number of purchased goods, and n is the preset threshold;

[0094] According to the preset upper limit score and the preset lower limit score, the revised initial weighted weights are constrained to obtain the target weighted weight of each recalled product.

[0095] It is understandable that when the number of purchases by users in each category is small, that is, the number of samples is small, the initial weighted weight has a certain deviation. Therefore, a weighted score correction method is designed. When the number of goods purchased by users in a certain category is m (greater than 0), and m is less than n, the initial weighted weight is reduced. Among them, n can be 5.

[0096] Specifically, for each recalled product, the revised initial weighted weight is calculated based on its uncorrected initial weighted weight, the number of purchased products corresponding to the preferred price determined according to the target purchase history, and the preset threshold, and the revised initial weighted weight is constrained to obtain the target weighted weight of each recalled product.

[0097] In this embodiment, by adjusting the initial weights according to the purchase history and preferred price, the output of the recommendation system can be better personalized. The weighted weights after constraint correction can ensure that the recommendation results are within the preset score range, avoiding the situation where the weights are too high or too low, which helps to optimize the quality of the recommendation and make the recommendation results more reasonable and credible.

[0098] In one embodiment, the step of constraining each modified initial weighted weight according to a preset upper limit score and a preset lower limit score to obtain a target weighted weight for each recalled product includes:

[0099] For each recalled product, the corresponding target weight is calculated according to the following expression:

[0100]

[0101] in, Weight the target, is the preset upper limit score, is the preset lower limit score, is the modified initial weight.

[0102] It is understandable that setting upper and lower limits on the revised initial weights can prevent unlimited weighting from promoting products that are ranked very low to the front, and products ranked very low are likely to be not very good products. In other words, we cannot only consider price matching, but also "quality".

[0103] In this embodiment, setting a preset upper limit score and a preset lower limit score can ensure that the final weighted weight of each product is within a reasonable range, which helps to avoid extreme situations where the weight is too high or too low, making the weight distribution more reasonable and stable.

[0104] The following is a description of the commodity recommendation device provided in the embodiment of the present application. The commodity recommendation device described below and the commodity recommendation method described above can be referred to in correspondence with each other. Figure 2 As shown, the present application provides a commodity recommendation device, which includes:

[0105] The product access request response module 201 is used to respond to the user's product access request, determine the target product category corresponding to the product access request, and obtain the user's target purchase history under the target product category;

[0106] The price deviation calculation module 202 is used to determine the user's preferred price for the target product category based on the target purchase history, and calculate the price deviation of each recalled product in the target product category based on the preferred price;

[0107] The target weighted weight calculation module 203 is used to determine the initial weighted weight of each recalled product according to the price deviation, and determine the target weighted weight of each recalled product based on each initial weighted weight, according to a preset upper limit score and a preset lower limit score;

[0108] The product push module 204 is used to sort each recalled product according to the weighted weights of each target, and push the sorted recalled products to the user.

[0109] In one embodiment, the price deviation calculation module 202 includes:

[0110] A price acquisition unit, configured to acquire the price of each purchased product of the user in the target product category from the target purchase history when it is determined that the user has purchased products in the target product category according to the target purchase history;

[0111] The first preferred price determination unit is used to use the average price corresponding to each purchased commodity under the target commodity category as the preferred price.

[0112] In one embodiment, the price deviation calculation module 202 further includes:

[0113] A same-level purchase history acquisition unit, used to acquire the same-level purchase history of the user in each same-level product category when it is determined according to the target purchase history that the user has not purchased a product in the target product category, where the same-level product category is a product category of the same level as the target product category;

[0114] The second preferred price determination unit is used to obtain the prices of each purchased commodity of the user under the same level commodity category from the same level purchase history, and use the average price corresponding to each purchased commodity under the same level commodity category as the preferred price.

[0115] In one embodiment, the price deviation calculation module 202 includes:

[0116] The price deviation calculation unit is used to calculate the corresponding price deviation for each recalled product according to the following expression:

[0117]

[0118] in, is the price deviation, The price of the recalled product. For preferred price.

[0119] In one embodiment, the target weight calculation module 203 includes:

[0120] The initial weight calculation unit is used to calculate the corresponding initial weight for each recalled product according to the following expression:

[0121] .

[0122] in, is the initial weight, is the price deviation, , and All are preset constants.

[0123] In one embodiment, the target weight calculation module 203 includes:

[0124] The initial weighted weight correction unit is used to correct each initial weighted weight according to the following expression when the number of purchased goods corresponding to the preferred price determined according to the target purchase history is less than a preset threshold:

[0125]

[0126] in, is the modified initial weight, is the uncorrected initial weight, m ​​is the number of purchased goods, and n is the preset threshold;

[0127] The target weighted weight calculation unit is used to constrain each revised initial weighted weight according to a preset upper limit score and a preset lower limit score to obtain a target weighted weight for each recalled product.

[0128] In one embodiment, the target weight calculation unit includes:

[0129] The target weight calculation subunit is used to calculate the corresponding target weight for each recalled product according to the following expression:

[0130]

[0131] in, Weight the target, is the preset upper limit score, is the preset lower limit score, is the modified initial weight.

[0132] In one embodiment, the present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the product recommendation method as described in any of the above embodiments.

[0133] In one embodiment, the present application also provides a computer device having computer-readable instructions stored therein. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the product recommendation method as described in any one of the above embodiments.

[0134] Indicatively, if Figure 3 As shown, Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. The computer device 300 may be provided as a server. Figure 3 The computer device 300 includes a processing component 302, which further includes one or more processors, and a memory resource represented by a memory 301, for storing instructions executable by the processing component 302, such as an application. The application stored in the memory 301 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 302 is configured to execute instructions to perform the product recommendation method of any of the above embodiments.

[0135] The computer device 300 may further include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate based on an operating system stored in the memory 301, such as Windows Server TM, Mac OS X TM, Unix TM, Linux TM, Free BSD TM, or the like.

[0136] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not clearly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or equipment including the elements. Herein, "one", "one", "said", "the" and "it" may also include plural forms, unless the context clearly indicates another way. A plurality refers to at least two cases, such as 2, 3, 5 or 8, etc. "And / or" includes any and all combinations of the relevant listed items.

[0138] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can refer to each other.

[0139] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A product recommendation method, characterized in that: The method comprises: In response to a user's product access request, determining a target product category corresponding to the product access request, and obtaining a target purchase history of the user under the target product category; Determining the user's preferred price for the target product category based on the target purchase history, and calculating the price deviation of each recalled product in the target product category based on the preferred price; Determining an initial weighting weight of each of the recalled products according to the price deviation, and determining a target weighting weight of each of the recalled products based on each of the initial weighting weights according to a preset upper limit score and a preset lower limit score; Each of the recalled products is sorted according to each of the target weights, and the sorted recalled products are pushed to the user.

2. The commodity recommendation method according to claim 1, characterized in that: The step of determining the user's preferred price for the target commodity category based on the target purchase history includes: When it is determined according to the target purchase history that the user has purchased goods in the target product category, obtaining the price of each purchased product in the target product category by the user from the target purchase history; The average price corresponding to each purchased commodity under the target commodity category is used as the preferred price.

3. The commodity recommendation method according to claim 1 or 2, characterized in that: The step of determining the user's preferred price for the target commodity category based on the target purchase history further includes: When it is determined according to the target purchase history that the user has not purchased a product in the target product category, obtaining the user's same-level purchase history in each same-level product category, where the same-level product category is a product category of the same level as the target product category; From the same-level purchase history, the prices of each purchased commodity by the user under the same-level commodity category are obtained, and the average price corresponding to each purchased commodity under the same-level commodity category is used as the preferred price.

4. The commodity recommendation method according to claim 1, characterized in that: The step of calculating the price deviation of each recalled product in the target product category according to the preferred price includes: For each of the recalled products, the corresponding price deviation is calculated according to the following expression: in, is the price deviation, The price of the recalled product. is the preferred price.

5. The commodity recommendation method according to claim 1, characterized in that: The step of determining the initial weight of each of the recalled products according to the price deviation comprises: For each of the recalled products, the corresponding initial weight is calculated according to the following expression: in, is the initial weight, is the price deviation, , and All are preset constants.

6. The commodity recommendation method according to claim 1, characterized in that: The step of determining the target weight of each recalled product based on each of the initial weighted weights according to a preset upper limit score and a preset lower limit score comprises: When it is determined according to the target purchase history that the number of purchased goods corresponding to the preferred price is less than a preset threshold, each of the initial weighted weights is modified according to the following expression: in, is the modified initial weight, is the uncorrected initial weighted weight, m ​​is the number of purchased goods, and n is the preset threshold; According to the preset upper limit score and the preset lower limit score, the revised initial weights are constrained to obtain the target weight of each recalled product.

7. The commodity recommendation method according to claim 6, characterized in that: The step of constraining each of the revised initial weighted weights according to the preset upper limit score and the preset lower limit score to obtain the target weighted weight of each of the recalled products includes: For each of the recalled products, the corresponding target weight is calculated according to the following expression: in, Weight the target, is the preset upper limit score, is the preset lower limit score, is the modified initial weight.

8. A commodity recommendation device, characterized in that: The device comprises: A product access request response module, configured to respond to a user product access request, determine a target product category corresponding to the product access request, and obtain a target purchase history of the user under the target product category; a price deviation calculation module, configured to determine the user's preferred price for the target commodity category based on the target purchase history, and calculate the price deviation of each recalled commodity in the target commodity category based on the preferred price; A target weighted weight calculation module, used to determine the initial weighted weight of each of the recalled products according to the price deviation, and determine the target weighted weight of each of the recalled products based on each of the initial weighted weights according to a preset upper limit score and a preset lower limit score; The product push module is used to sort each of the recalled products according to each of the target weights, and push the sorted recalled products to the user.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the product recommendation method as described in any one of claims 1 to 7.

10. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the product recommendation method according to any one of claims 1 to 7 are performed.

Citation Information

Cited By

  • Wish list purchase prompting method and device based on price reduction law prediction, and medium

    CN122865630A