Data filtering method and device

Abstract

The present application discloses a data filtering method and device, which relates to the field of information processing technology. The specific implementation scheme is: in response to receiving a filtering request for a user's first product set, obtaining the user's first product pair set and user behavior information; based on the user repurchase selection model and the user behavior judgment model, combining and filtering the first product pair set to generate a filtered first product pair set and an unfiltered first product pair set; according to the product filtering strategy, selecting the unfiltered first product pair set to obtain a second product pair set; generating the user's second product set based on the filtered first product pair set and the second product pair set. This scheme fully considers user behavior and product attributes, analyzes the user's personalized repurchase habits, and judges the user's behavior information, making the filtering more refined and accurate, and achieving a better user experience effect.

Description

Technical Field

[0001] The present application relates to the field of computer technology, specifically to the field of information processing technology, and in particular to a data filtering method and device. Background Art

[0002] With the rapid development of Internet technology, online shopping has become a very important shopping mode for people. The recommendation system in the e-commerce scenario can help users quickly find the products they are really interested in from the massive information, saving users time and improving shopping experience. The recommendation system in the product information flow scenario pays more attention to personalization and user experience. It needs to help users "browse" in the product information flow and recommend products that may be of interest to users in a timely manner.

[0003] The user's purchase history can reflect their interests to a large extent. Common recommendation algorithms can easily capture such interests and recommend products that are very similar to the products they have purchased (for example, if you have purchased a mobile phone, you may be recommended another mobile phone). However, since products have a certain usage cycle, untimely repeated recommendations will not only waste traffic resources, but also cause trouble to users, making the products they see more and more concentrated. Especially in the information flow scenario, as the user's pull-down depth deepens, repeated recommendations of purchased products will seriously affect the user experience. Therefore, personalized repurchase recommendations for users play a very important role in improving the user experience in the information flow scenario.

[0004] At present, the main solutions for optimizing the repurchase experience are divided into two categories: one is purchased filtering, that is, after recalling the product set, filter out the products that the user has purchased through a combination of strategies. The other is to consider the modeling of purchased information, that is, consider the user's purchased or repurchase characteristics when modeling the recommendation algorithm to avoid recalling a large number of users' purchased products. Summary of the invention

[0005] The present application provides a data filtering method, apparatus, device and storage medium.

[0006] According to a first aspect of the present application, a data filtering method is provided, the method comprising: in response to receiving a filtering request for a user's first commodity set, obtaining the user's first product pair set and user behavior information, wherein the product pair comprises: a product name and a purchase time corresponding to the product name; based on a user repurchase selection model and a user behavior judgment model, performing combined filtering on the first product pair set to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior judgment model is used to characterize the judgment of the first product pair set based on user behavior information; according to the product filtering strategy, selecting the unfiltered first product pair set to obtain a second product pair set; generating the user's second commodity set based on the filtered first product pair set and the second product pair set.

[0007] In some embodiments, the first product pair set is filtered based on the user repurchase selection model, including: after clustering all product pairs in the first product pair set, calculating the repurchase period of each product in the first product pair set; judging whether each product in the first product pair set exceeds the repurchase time limit according to the repurchase period of each product and the purchase time of the corresponding product; if the products in the first product pair set exceed the repurchase time limit, storing the product pairs of the products in the filtered first product pair set; if the products in the first product pair set do not exceed the repurchase time limit, storing the product pairs of the products in the unfiltered first product pair set.

[0008] In some embodiments, the first product pair set is filtered based on the user behavior determination model, including: the user behavior determination model determines whether each product in the first product pair set meets the exemption condition according to the time when the user behavior occurs and / or the number of times the user behavior occurs, wherein the exemption condition is used to characterize the threshold judgment of the time when the user behavior occurs and / or the number of times the user behavior occurs, and the judgment is completed based on the standard product unit SPU of the product; if the products in the first product pair set meet the exemption condition, the product pair corresponding to the product is stored in the filtered first product pair set, and if the products in the first product pair set do not meet the exemption condition, the product pair corresponding to the product is stored in the unfiltered first product pair set.

[0009] In some embodiments, the user behavior information includes: at least one of the user's search behavior, the user's add-to-cart behavior, the user's attention behavior, and the user's click behavior.

[0010] In some embodiments, after clustering all product pairs in the first product pair set and calculating the repurchase cycle of each product in the first product pair set, it also includes: obtaining a third product pair set from the first product pair set, wherein the third product pair set is used to characterize the set of product pairs that have not obtained a repurchase cycle; based on a product repurchase selection model, filtering the third product pair set to generate a repurchase cycle for each product in the third product pair set, wherein the product repurchase selection model is used to characterize the selection of the third product pair set based on the repurchase cycle of the product.

[0011] In some embodiments, the product filtering strategy performs combined screening on the unfiltered first product pair set based on multiple dimensions; according to the product filtering strategy, the unfiltered first product pair set is selected to obtain a second product pair set, including: according to the product filtering model, judging whether each product in the unfiltered first product pair set is a purchased commodity for the user, wherein the product filtering model is used to characterize at least two items of the product word based on the product, the standard product unit SPU of the product and the inventory holding unit SKU of the product, and performing combined filtering on each product in the unfiltered first product pair set; if a product in the unfiltered first product pair set is a purchased commodity for the user, then deleting the product pair corresponding to the product from the unfiltered first product pair set to obtain a second product pair set.

[0012] In some embodiments, after selecting the unfiltered first product pair set according to the product filtering strategy to obtain the second product pair set, it also includes: selecting the second product pair set based on the product category to obtain the selected second product pair set.

[0013] In some embodiments, the method further includes: sorting the user's second product set according to the product display strategy to generate a product list for the user.

[0014] According to a second aspect of the present application, a data filtering device is provided, the device comprising: an acquisition unit, configured to acquire a first product pair set and user behavior information of the user in response to receiving a filtering request for a first product set of the user, wherein the product pair comprises: a product name and a purchase time corresponding to the product name; a filtering unit, configured to perform combination filtering on the first product pair set based on a user repurchase selection model and a user behavior determination model, to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior determination model is used to characterize the determination of the first product pair set based on user behavior information; a first selection unit, configured to select the unfiltered first product pair set according to a product filtering strategy to obtain a second product pair set; a generation unit, configured to generate a second product set of the user based on the filtered first product pair set and the second product pair set.

[0015] In some embodiments, the filtering unit includes: a clustering module, configured to cluster all product pairs in the first product pair set, and then calculate the repurchase period of each product in the first product pair set; a first judgment module, configured to judge whether each product in the first product pair set exceeds the repurchase time limit based on the repurchase period of each product and the purchase time of the corresponding product; a first storage module, configured to store the product pair of the product in the filtered first product pair set if the product in the first product pair set exceeds the repurchase time limit, and store the product pair of the product in the unfiltered first product pair set if the product in the first product pair set does not exceed the repurchase time limit.

[0016] In some embodiments, the filtering unit also includes: a second judgment module, configured to judge whether each product in the first product pair set meets the exemption condition based on the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, wherein the exemption condition is used to characterize the threshold judgment of the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, and the judgment is based on the standard product unit SPU of the product; a second storage module, configured to store the product pair corresponding to the product in the filtered first product pair set if the product in the first product pair set meets the exemption condition, and store the product pair corresponding to the product in the unfiltered first product pair set if the product in the first product pair set does not meet the exemption condition.

[0017] In some embodiments, the filtering unit also includes: an acquisition module, configured to acquire a third product pair set from the first product pair set, wherein the third product pair set is used to characterize a set of product pairs that have not obtained a repurchase cycle; a filtering module, configured to filter the third product pair set based on a product repurchase selection model, and generate a repurchase cycle for each product in the third product pair set, wherein the product repurchase selection model is used to characterize the selection of the third product pair set based on the repurchase cycle of the product.

[0018] In some embodiments, the product filtering strategy in the first selection unit performs combined screening on the unfiltered first product pair set based on multiple dimensions; the first selection unit includes: a judgment module, configured to judge whether each product in the unfiltered first product pair set is a purchased commodity for the user based on a product filtering model, wherein the product filtering model is used to characterize at least two items based on the product word, the product's standard product unit SPU and the product's inventory holding unit SKU, and perform combined filtering on each product in the unfiltered first product pair set; a deletion module, configured to delete the product pair corresponding to the product from the unfiltered first product pair set if the product in the unfiltered first product pair set is a purchased commodity for the user, so as to obtain a second product pair set.

[0019] In some embodiments, the device further includes: a second selection unit configured to select a second product pair set based on the category of the product to obtain a selected second product pair set.

[0020] In some embodiments, the device further includes: a generating unit configured to sort the second commodity set of the user according to the commodity display strategy and generate a commodity list for the user.

[0021] According to the third aspect of the present application, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any implementation manner in the first aspect.

[0022] According to a fourth aspect of the present application, the present application provides a non-transitory computer-readable storage medium storing computer instructions, characterized in that the computer instructions are used to enable a computer to execute a method as described in any implementation manner in the first aspect.

[0023] The technology adopted in the present application is to respond to receiving a filtering request for the user's first product set, obtain the user's first product pair set and user behavior information, and perform combined filtering on the first product pair set based on the user repurchase selection model and the user behavior judgment model to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior judgment model is used to characterize the judgment of the first product pair set based on the user behavior information, and according to the product filtering strategy, the unfiltered first product pair set is selected to obtain a second product pair set, and according to the filtered first product pair set and the second product pair set, a filtered first product pair set and an unfiltered first product pair set are generated. The second product set for the user is formed, which avoids the problems in the prior art of incorrect filtering results due to incorrect division of a certain product category and the inability to accurately describe the products purchased by the user through static attributes such as category or brand, resulting in a certain degree of missed filtering or wrong filtering. It also avoids the problem in the prior art that the filtering strategy cannot be personalized, resulting in the inability to fully utilize the traffic and causing a waste of traffic resources. At the same time, it solves the problems in the prior art of high modeling difficulty, high model complexity and certain uncertainty in the prediction results. The present application fully considers user behavior and product attributes, analyzes the user's personalized repurchase habits, and determines the user behavior information, so that the filtering is more refined and accurate, and a better user experience effect is achieved.

[0024] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present application.

[0026] Figure 1 is a schematic diagram of a first embodiment of a data filtering method according to the present application;

[0027] Figure 2 It is a scene graph that can implement the data filtering method of the embodiment of the present application;

[0028] Figure 3 is a schematic diagram of a second embodiment of a data filtering method according to the present application;

[0029] Figure 4 is a structural schematic diagram of an embodiment of a data filtering device according to the present application;

[0030] Figure 5 It is a block diagram of an electronic device used to implement the data filtering method of an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0032] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0033] Figure 1 A schematic diagram 100 of a first embodiment of a data filtering method according to the present application is shown. The data filtering method comprises the following steps:

[0034] Step 101, in response to receiving a filtering request for a user's first commodity set, obtaining the user's first product pair set and user behavior information.

[0035] In this embodiment, when the execution subject receives a filtering request for the user's first product set, the execution subject can obtain the user's first product pair set and user behavior information from other electronic devices or locally through a wired connection or a wireless connection, or can obtain the user's first product pair set and user behavior information by parsing the filtering request. Among them, the product pair includes: a product name and a purchase time corresponding to the product name, such as <mobile phone, 202005101030>. It should be pointed out that the above-mentioned wireless connection method may include but is not limited to 3G, 4G, 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0036] Step 102: Based on the user repurchase selection model and the user behavior determination model, the first product pair set is combined and filtered to generate a filtered first product pair set and an unfiltered first product pair set.

[0037] In this embodiment, the execution entity may perform combined filtering on the first product pair set based on the user repurchase selection model and the user behavior determination model, and use the first product pairs that simultaneously satisfy the user repurchase selection model and the user behavior determination model as filtered first product pairs to generate a filtered first product pair set, and use the first product pairs that do not simultaneously satisfy the user repurchase selection model and the user behavior determination model as unfiltered first product pairs to generate an unfiltered first product pair set. The user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior determination model is used to characterize the determination of the first product pair set based on user behavior information. User behavior information may include: at least one of: the user's search behavior, the user's add-to-purchase behavior, the user's attention behavior, and the user's click behavior.

[0038] Step 103: According to the product filtering strategy, the unfiltered first product pair set is selected to obtain a second product pair set.

[0039] In this embodiment, the execution entity may select a suitable product filtering method according to business requirements, and select the unfiltered first product pair set to obtain the second product pair set.

[0040] In some optional implementations of this embodiment, the product filtering strategy performs combined screening on the unfiltered first product pair set based on multiple dimensions. The multiple dimensions may include: different attribute characteristics of the goods, different behavioral characteristics of the users, and different product filtering methods. The different attribute characteristics of the goods may include: brand category of the goods, category of the goods, etc., and the different behavioral characteristics of the users may include: length of stay, number of visits, degree of preference, etc. Different product filtering methods may include: from coarse to fine, filtering methods based on product categories, based on product words, based on product standard product units SPU, and based on inventory holding units SKU. Avoiding missed filtering improves the accuracy of filtering.

[0041] To further explain, in the e-commerce scenario, products are generally organized from coarse to fine, which is convenient for description and understanding from different dimensions. For example, an iPhone 11 mobile phone can be described from coarse to fine dimensions as follows: Category: Mobile Communication, Product Word: Smart Phone, SPU: Apple iPhone 11, SKU: Apple iPhone 11 (Space Gray, 256G), Apple iPhone 11 (Space Gray, 512G), Apple iPhone 11 (Gold, 256G)... The dimensional ranges of category, product word, SPU, and SKU are successively refined, and the set of products included is getting smaller and smaller. SKU is the smallest descriptive unit of a product. In actual applications, when a user has already purchased a product (such as an iPhone 11 mobile phone), we need to block products similar to the product to avoid repeated recommendations. Therefore, it is necessary to find a suitable dimension to describe the product.

[0042] For example, if a user buys an Apple iPhone 11 (space gray, 256G) phone, if the SKU dimension is filtered, the system will still recommend the same phone with other colors to the user; if the SPU dimension is filtered, although Apple phones will not be recommended, other brands of phones such as Huawei and OPPO will still be recommended; if the product category dimension is filtered, mobile communication products will not be recommended to this user, as the coverage is too broad. Product words are a more appropriate dimension choice. Smartphone products will not be recommended to this user, but other types of products can still be recommended to ensure user experience.

[0043] Step 104: Generate a second product set for the user based on the filtered first product pair set and the second product pair set.

[0044] In this embodiment, the execution entity may merge the filtered first product pair set and the second product pair set to generate a second product set for the user.

[0045] Continue to see Figure 2The data filtering method 200 of this embodiment runs in the electronic device 201. When the electronic device 201 receives the filtering request of the user's first product set, the electronic device 201 first obtains the user's first product pair set and the user behavior information 202, and then the electronic device 201 performs combined filtering on the first product pair set based on the user repurchase selection model and the user behavior judgment model to generate a filtered first product pair set and an unfiltered first product pair set 203, and then the electronic device 201 selects the unfiltered first product pair set according to the product filtering strategy to obtain a second product pair set 204, and finally the electronic device 201 generates the user's second product set 205 according to the filtered first product pair set and the second product pair set.

[0046] The data filtering method provided by the above-mentioned embodiment of the present application adopts a method of responding to a filtering request received from a user for a first product set, obtaining a first product pair set and user behavior information of the user, and combining and filtering the first product pair set based on a user repurchase selection model and a user behavior judgment model to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior judgment model is used to characterize the judgment of the first product pair set based on the user behavior information, and according to the product filtering strategy, the unfiltered first product pair set is selected to obtain a second product pair set, and according to the filtered first product pair set and the second product pair set, a filter is performed. For the set, a second product set for the user is generated, which avoids the problem in the prior art that the filtering results are wrong due to the incorrect division of a certain product category, and the user's purchased products cannot be accurately described only through static attributes such as category or brand, resulting in a certain degree of missed filtering or wrong filtering. It also avoids the problem in the prior art that the filtering strategy cannot be personalized, resulting in the inability to fully utilize the traffic and causing a waste of traffic resources. At the same time, it solves the problems in the prior art that the modeling is difficult, the model complexity is high, and the prediction results have a certain degree of uncertainty. The present application fully considers the user behavior and product attributes, analyzes the user's personalized repurchase habits, and judges the user behavior information, so that the filtering is more refined and accurate, and a better user experience effect is achieved.

[0047] Further references Figure 3 , which shows a schematic diagram 300 of a second embodiment of a data filtering method. The process of the method comprises the following steps:

[0048] Step 301, in response to receiving a filtering request for a user's first commodity set, obtaining the user's first product pair set and user behavior information.

[0049] Step 302: Based on the user repurchase selection model and the user behavior determination model, the first product pair set is combined and filtered to generate a filtered first product pair set and an unfiltered first product pair set.

[0050] In some optional implementations of this embodiment, the first product pair set is filtered based on the user repurchase selection model, including: after clustering all product pairs in the first product pair set, the repurchase period of each product in the first product pair set is calculated; according to the repurchase period of each product and the purchase time of the corresponding product, it is determined whether each product in the first product pair set exceeds the repurchase time limit; if the products in the first product pair set exceed the repurchase time limit, the product pairs of the products are stored in the filtered first product pair set; if the products in the first product pair set do not exceed the repurchase time limit, the product pairs of the products are stored in the unfiltered first product pair set. Analyze the personalized repurchase habits of users and generate personalized repurchase cycles for users, so that filtering is more targeted and a better user experience is obtained.

[0051] In some optional implementations of this embodiment, after clustering all product pairs in the first product pair set and calculating the repurchase period of each product in the first product pair set, it also includes: obtaining a third product pair set from the first product pair set, wherein the third product pair set is used to characterize the set of each product pair that has not obtained a repurchase period; based on a product repurchase selection model, filtering the third product pair set to generate a repurchase period for each product in the third product pair set, wherein the product repurchase selection model is used to characterize the selection of the third product pair set based on the repurchase period of the product. This supplements the product pairs that do not generate a repurchase period based on the user dimension, thereby improving the precision and accuracy of filtering.

[0052] In some optional implementations of this embodiment, the first product pair set is filtered based on the user behavior determination model, including: the user behavior determination model determines whether each product in the first product pair set meets the exemption condition based on the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, wherein the exemption condition is used to characterize the threshold judgment of the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, and the judgment is completed based on the standard product unit SPU of the product; if the product in the first product pair set meets the exemption condition, the product pair corresponding to the product is stored in the filtered first product pair set, and if the product in the first product pair set does not meet the exemption condition, the product pair corresponding to the product is stored in the unfiltered first product pair set. Avoid interference caused by user misoperation, avoid repeated recommendations of purchased products, model the active behavior generated after purchase, understand user behavior to the greatest extent, and make filtering more refined and accurate.

[0053] For example, set time windows T1 and T2. When the time between the user behavior and the purchase time is greater than T1, and the time between the user behavior and the current time is less than T2, the product pair corresponding to the product is deleted from the product pair set. Since search and high-quality click behaviors may occur multiple times, the number of eligible behaviors can be counted in actual use. If a certain frequency threshold is met, the product pair corresponding to the product is deleted from the product pair set.

[0054] Step 303: judging, based on the product filtering model, whether each product in the unfiltered first product pair set is a purchased commodity of the user.

[0055] In this embodiment, the execution subject can judge each product in the unfiltered first product pair set according to the product filtering model to determine whether each product in the unfiltered first product pair set is a purchased commodity of the user. The product filtering model is used to represent the combined filtering of each product in the unfiltered first product pair set based on at least two of the product word, the standard product unit SPU of the product, and the stock keeping unit SKU of the product.

[0056] Step 304: If a product in the unfiltered first product pair set is a purchased commodity of the user, the product pair corresponding to the product is deleted from the unfiltered first product pair set to obtain a second product pair set.

[0057] In this embodiment, when the execution entity determines that a product in the unfiltered first product pair set is a purchased item for the user, the product pair corresponding to the product is deleted from the unfiltered first product pair set. When the judgment on all products is completed, the second product pair set is obtained.

[0058] Step 305: Select the second product pair set based on the product category to obtain a selected second product pair set.

[0059] In this embodiment, the execution entity may select the second product pair set based on the product category to obtain the selected second product pair set.

[0060] Step 306: Generate a second product set for the user based on the filtered first product pair set and the filtered second product pair set.

[0061] In some optional implementations of this embodiment, the method further includes: sorting the user's second product set according to the product display strategy, generating a product list for the user, and recommending various products that meet the user's needs to the user. For example, for products that meet the filtering conditions, the click rate prediction model is used to sort and score them, and the top-ranked products are selected for display.

[0062] In this embodiment, the specific operations of steps 301, 302 and 306 are the same as Figure 1 The operations of steps 101 , 102 and 104 in the illustrated embodiment are substantially the same and will not be described in detail herein.

[0063] from Figure 3 It can be seen that Figure 1 Compared with the corresponding embodiment, the schematic diagram 300 of the data filtering method in this embodiment adopts a product filtering model to determine whether each product in the unfiltered first product pair set is a purchased commodity of the user, wherein the product filtering model is used to characterize at least two of the product words based on the product, the standard product unit SPU of the product and the inventory holding unit SKU of the product, and perform combined filtering on each product in the unfiltered first product pair set. If the product in the unfiltered first product pair set is a purchased commodity of the user, the product pair corresponding to the product is deleted from the unfiltered first product pair set to obtain a second product pair set. Considering a combination of strategies with multiple granularities, screening is performed based on multiple dimensions of the product, and selection is performed according to specific business conditions. This can fully describe the user's purchased commodities from different dimensions, minimize omissions and false filtering, and obtain a better user experience effect. The second product pair set is selected based on the product category to obtain the selected second product pair set as a bottom filter to ensure the completeness of the filtering.

[0064] Further references Figure 4 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a data filtering device, which is similar to Figure 1 Corresponding to the method embodiment shown, the device is in the form of an independent Lib (static link library) and can be called by any industrial-grade recommendation system. The device can be specifically applied to various electronic devices.

[0065] like Figure 4As shown, the data filtering device 400 of this embodiment includes: an acquisition unit 401, a filtering unit 402, a first selection unit 403 and a generation unit 404, wherein the acquisition unit is configured to obtain the user's first product pair set and user behavior information in response to receiving a filtering request for the user's first product set, wherein the product pair includes: a product name and a purchase time corresponding to the product name; the filtering unit is configured to perform combined filtering on the first product pair set based on a user repurchase selection model and a user behavior judgment model to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior judgment model is used to characterize the judgment of the first product pair set based on user behavior information; the first selection unit is configured to select the unfiltered first product pair set according to the product filtering strategy to obtain a second product pair set; the generation unit is configured to generate the user's second product set based on the filtered first product pair set and the second product pair set.

[0066] In this embodiment, the specific processing of the acquisition unit 401, the filtering unit 402, the first selection unit 403 and the generation unit 404 of the data filtering device 400 and the technical effects thereof can be referred to in Figure 1 The relevant descriptions of steps 101 to 104 in the corresponding embodiment are not repeated here.

[0067] In some optional implementations of the present embodiment, the filtering unit includes: a clustering module, configured to cluster all product pairs in the first product pair set, and then calculate the repurchase period of each product in the first product pair set; a first judgment module, configured to judge whether each product in the first product pair set exceeds the repurchase time limit based on the repurchase period of each product and the purchase time of the corresponding product; a first storage module, configured to store the product pair of the product in the filtered first product pair set if the product in the first product pair set exceeds the repurchase time limit, and store the product pair of the product in the unfiltered first product pair set if the product in the first product pair set does not exceed the repurchase time limit.

[0068] In some optional implementations of the present embodiment, the filtering unit further includes: a second judgment module, configured to judge whether each product in the first product pair set meets the exemption condition based on the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, wherein the exemption condition is used to characterize the threshold judgment of the time of occurrence of the user behavior and / or the number of occurrences of the user behavior, and the judgment is based on the standard product unit SPU of the product; a second storage module, configured to store the product pair corresponding to the product in the filtered first product pair set if the product in the first product pair set meets the exemption condition, and store the product pair corresponding to the product in the unfiltered first product pair set if the product in the first product pair set does not meet the exemption condition.

[0069] In some optional implementations of the present embodiment, the filtering unit further includes: an acquisition module, configured to acquire a third product pair set from the first product pair set, wherein the third product pair set is used to characterize a set of product pairs that have not obtained a repurchase cycle; a filtering module, configured to filter the third product pair set based on a product repurchase selection model, and generate a repurchase cycle for each product in the third product pair set, wherein the product repurchase selection model is used to characterize the selection of the third product pair set based on the repurchase cycle of the product.

[0070] In some optional implementations of the present embodiment, the product filtering strategy in the first selection unit performs combined screening on the unfiltered first product pair set based on multiple dimensions; the first selection unit includes: a judgment module, configured to judge whether each product in the unfiltered first product pair set is a purchased commodity for the user based on a product filtering model, wherein the product filtering model is used to characterize at least two items based on the product word, the product's standard product unit SPU and the product's inventory holding unit SKU, and perform combined filtering on each product in the unfiltered first product pair set; a deletion module, configured to delete the product pair corresponding to the product from the unfiltered first product pair set if the product in the unfiltered first product pair set is a purchased commodity for the user, so as to obtain a second product pair set.

[0071] In some optional implementations of this embodiment, the device also includes: a second selection unit, configured to select a second product pair set based on the category of the product to obtain a selected second product pair set.

[0072] In some optional implementations of this embodiment, the device further includes: a generating unit configured to sort the second commodity set of the user according to the commodity display strategy and generate a commodity list for the user.

[0073] According to an embodiment of the present application, the present application also provides an electronic device and a readable storage medium.

[0074] like Figure 5 , is a block diagram of an electronic device according to a data filtering method according to an embodiment of the present application. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0075] like Figure 5 As shown, the electronic device includes: one or more processors 501, a memory 502, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In other embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 501 is taken as an example.

[0076] The memory 502 is a non-transient computer-readable storage medium provided in the present application. The memory stores instructions executable by at least one processor to enable at least one processor to perform the data filtering method provided in the present application. The non-transient computer-readable storage medium of the present application stores computer instructions, which are used to enable a computer to perform the data filtering method provided in the present application.

[0077] The memory 502 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the data filtering method in the embodiment of the present application (for example, the attached Figure 4 The processor 501 executes various functional applications and data processing of the server by running the non-transient software programs, instructions and modules stored in the memory 502, that is, the data filtering method in the above method embodiment is implemented.

[0078] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created according to the use of the data filtering electronic device, etc. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 502 may optionally include a memory remotely arranged relative to the processor 501, and these remote memories may be connected to the data filtering electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0079] The electronic device of the data filtering method may further include: an input device 503 and an output device 504. The processor 501, the memory 502, the input device 503 and the output device 504 may be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0080] The input device 503 can receive input digital or character information, and generate key signal input related to user settings and function control of the data filtering electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 504 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device may be a touch screen.

[0081] Various implementations of the systems and techniques described herein can be realized in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0082] These computer programs (also referred to as programs, software, software applications, or code) include machine instructions for programmable processors and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or means (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0083] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0084] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0085] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0086] According to the technical solution of the embodiment of the present application, in response to receiving a filtering request for the user's first commodity set, the user's first product pair set and user behavior information are obtained, and based on the user repurchase selection model and the user behavior judgment model, the first product pair set is combined and filtered to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and purchase time, and the user behavior judgment model is used to characterize the judgment of the first product pair set based on the user behavior information, and according to the product filtering strategy, the unfiltered first product pair set is selected to obtain a second product pair set, and based on the filtered first product pair set and the second product pair set The invention can generate a second product set for the user, thereby avoiding the problems in the prior art of incorrect filtering results due to incorrect division of a certain product category and the inability to accurately describe the products purchased by the user through static attributes such as category or brand, resulting in a certain degree of missed filtering or wrong filtering. It also avoids the problem in the prior art that the filtering strategy cannot be personalized, resulting in the inability to fully utilize the traffic and causing a waste of traffic resources. At the same time, it solves the problems in the prior art of high modeling difficulty, high model complexity and certain uncertainty in the prediction results. The present application fully considers user behavior and product attributes, analyzes the personalized repurchase habits of users, and judges user behavior information, so that the filtering is more refined and accurate, and a better user experience effect is achieved.

[0087] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0088] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A data filtering method, the method comprising: In response to receiving a filtering request for a user's first commodity set, obtaining a first product pair set of the user and the user behavior information, wherein the product pair includes: a product name and a purchase time corresponding to the product name; Based on the user repurchase selection model and the user behavior determination model, the first product pair set is combined and filtered to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and the purchase time, and the user behavior determination model is used to characterize the determination of the first product pair set based on the user behavior information, the filtered first product pair set includes first product pairs that simultaneously meet the user repurchase selection model and the user behavior determination model, and the unfiltered first product pair set includes first product pairs that do not simultaneously meet the user repurchase selection model and the user behavior determination model; According to the strategies of multiple dimensions, the unfiltered first product pair set is combined and screened to obtain a second product pair set; generating a second product set for the user according to the filtered first product pair set and the second product pair set; The method also includes: obtaining a third product pair set from the first product pair set, the third product pair set being used to characterize a set of product pairs that have not obtained the repurchase cycle; filtering the third product pair set based on a product repurchase selection model to generate a repurchase cycle for each product in the third product pair set, the product repurchase selection model being used to characterize the selection of the third product pair set based on the product repurchase cycle.

2. The method according to claim 1, wherein: The filtering of the first product pair set based on the user repurchase selection model includes: After clustering all product pairs in the first product pair set, the repurchase period of each product in the first product pair set is calculated; According to the repurchase period of each product and the purchase time of the corresponding product, determining whether each product in the first product pair set exceeds the repurchase time limit; If the products in the first product pair set exceed the repurchase time limit, the product pairs of the products are stored in the filtered first product pair set; if the products in the first product pair set do not exceed the repurchase time limit, the product pairs of the products are stored in the unfiltered first product pair set.

3. The method according to claim 1, wherein: The filtering the first product pair set based on the user behavior determination model includes: According to the time when the user behavior occurs and / or the number of times the user behavior occurs, judging whether each product in the first product pair set meets the exemption condition, wherein the exemption condition is used to represent a threshold judgment on the time when the user behavior occurs and / or the number of times the user behavior occurs, and the judgment is completed based on the standard product unit SPU of the product; If the products in the first product pair set meet the exemption conditions, the product pairs corresponding to the products are stored in the filtered first product pair set; if the products in the first product pair set do not meet the exemption conditions, the product pairs corresponding to the products are stored in the unfiltered first product pair set.

4. The method according to claim 1, wherein: The combining and screening the unfiltered first product pair set according to the strategies of multiple dimensions includes: Determine, according to a product filtering model, whether each product in the unfiltered first product pair set is a purchased commodity of the user, wherein the product filtering model is used to characterize at least two items of a product word, a standard product unit SPU of a product, and a stock keeping unit SKU of a product, and perform combined filtering on each product in the unfiltered first product pair set; If a product in the unfiltered first product pair set is a purchased commodity of the user, the product pair corresponding to the product is deleted from the unfiltered first product pair set to obtain a second product pair set.

5. According to the method of claim 1, after combining and screening the unfiltered first product pair set according to the strategies of multiple dimensions to obtain the second product pair set, the method further comprises: Based on the category of the product, the second product pair set is selected to obtain a selected second product pair set.

6. The method according to claim 1, further comprising: The second product set of the user is sorted according to the product display strategy to generate a product list for the user.

7. A data filtering device, comprising: an acquisition unit, configured to acquire, in response to receiving a filtering request for a user's first commodity set, a first set of product pairs of the user and the user behavior information, wherein the product pair includes: a product name and a purchase time corresponding to the product name; a filtering unit configured to perform combined filtering on the first product pair set based on a user repurchase selection model and a user behavior determination model to generate a filtered first product pair set and an unfiltered first product pair set, wherein the user repurchase selection model is used to characterize the selection of the first product pair set based on the user's repurchase cycle and the purchase time, and the user behavior determination model is used to characterize the determination of the first product pair set based on the user behavior information, the filtered first product pair set includes first product pairs that simultaneously satisfy the user repurchase selection model and the user behavior determination model, and the unfiltered first product pair set includes first product pairs that do not simultaneously satisfy the user repurchase selection model and the user behavior determination model; A first selection unit is configured to perform combined screening on the unfiltered first product pair set according to strategies of multiple dimensions to obtain a second product pair set; a generating unit configured to generate a second commodity set for the user according to the filtered first product pair set and the second product pair set; The filtering unit is further configured to obtain a third product pair set from the first product pair set, the third product pair set being used to characterize a set of product pairs that have not obtained the repurchase cycle; filter the third product pair set based on a product repurchase selection model to generate a repurchase cycle for each product in the third product pair set, the product repurchase selection model being used to characterize the selection of the third product pair set based on the repurchase cycle of the product.

8. The device according to claim 7, wherein: The filtering unit comprises: A clustering module is configured to cluster all product pairs in the first product pair set and calculate a repurchase period of each product in the first product pair set; A first judgment module is configured to judge whether each product in the first product pair set exceeds the repurchase time limit according to the repurchase period of each product and the purchase time of the corresponding product; The first storage module is configured to store the product pairs of the products in the first product pair set into the filtered first product pair set if the products in the first product pair set exceed the repurchase time limit, and to store the product pairs of the products in the unfiltered first product pair set if the products in the first product pair set do not exceed the repurchase time limit.

9. The device according to claim 7, wherein: The filtering unit further comprises: a second judgment module, configured to judge whether each product in the first product pair set meets an exemption condition according to the time when the user behavior occurs and / or the number of times the user behavior occurs, wherein the exemption condition is used to represent a threshold judgment on the time when the user behavior occurs and / or the number of times the user behavior occurs, and the judgment is completed based on the standard product unit SPU of the product; The second storage module is configured to store the product pair corresponding to the product in the filtered first product pair set if the product in the first product pair set meets the exemption condition, and store the product pair corresponding to the product in the unfiltered first product pair set if the product in the first product pair set does not meet the exemption condition.

10. The device according to claim 7, wherein: The first selection unit includes: a judgment module configured to judge whether each product in the unfiltered first product pair set is a purchased commodity of the user according to a product filtering model, wherein the product filtering model is used to characterize the combination filtering of each product in the unfiltered first product pair set based on at least two of a product word, a standard product unit SPU of a product, and a stock keeping unit SKU of a product; The deleting module is configured to delete the product pair corresponding to the product from the unfiltered first product pair set if the product in the unfiltered first product pair set is a purchased commodity of the user, so as to obtain a second product pair set.

11. The device according to claim 7, further comprising: The second selection unit is configured to select the second product pair set based on the product category to obtain the selected second product pair set.

12. The device according to claim 7, further comprising: The generating unit is configured to sort the second commodity set of the user according to the commodity display strategy and generate a commodity list for the user.

13. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

Images