Object recommendation method, object recommendation device, electronic device and storage medium

Abstract

The present disclosure provides an object recommendation method, an object recommendation device, an electronic device and a computer-readable storage medium, which belong to the field of computer technology. The method comprises: obtaining first-category feature data of multiple candidate objects; determining second-category feature data of each candidate object based on the user category to which the target user belongs; determining the degree of match between each candidate object and the user category through the first-category feature data and the second-category feature data of each candidate object; and selecting one or more candidate objects according to the degree of match to recommend to the target user. The present disclosure can recommend more targeted objects to users of different user categories, thereby improving the effectiveness of object recommendation.

Description

Background Art

[0002] With the increasing popularity and application of the Internet in various industries, companies in e-commerce, Internet finance, life services, games and other fields are committed to collecting and analyzing data on products, users and other objects through the Internet to explore needs and make effective object recommendations, including: recommending products or stores to users, recommending users to merchants, or recommending news, games and other service information to users.

[0003] Most existing object recommendation methods collect massive amounts of data on various objects, build object features based on these data, and identify similar objects as recommended objects. However, this method of recommending similar objects based on object features may recommend objects with similar or identical content to users multiple times during the recommendation process, making users tired of the recommended content and making it difficult to maintain user stickiness. In addition, if recommendations are made based only on object features, users with different needs may be recommended the same object, resulting in inaccurate recommendation results and making it difficult to meet user needs.

[0004] Therefore, how to adopt a reasonable approach to perform accurate and effective object recommendation is an urgent problem to be solved in existing technologies.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0006] The present disclosure provides an object recommendation method, an object recommendation device, an electronic device, and a computer-readable storage medium, thereby overcoming the problem of inaccurate recommendations in existing object recommendation methods at least to a certain extent.

[0007] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0008] According to one aspect of the present disclosure, there is provided an object recommendation method, comprising: obtaining first-category feature data of a plurality of candidate objects; determining second-category feature data of each of the candidate objects based on the user category to which a target user belongs; determining a degree of match between each of the candidate objects and the user category through the first-category feature data and the second-category feature data of each of the candidate objects; and selecting one or more of the candidate objects based on the degree of match to recommend to the target user.

[0009] In an exemplary embodiment of the present disclosure, the degree of match between each candidate object and the user category is determined through the first-category feature data and the second-category feature data of each candidate object, including: standardizing the first-category feature data and the second-category feature data of each candidate object; weighting the data of each preset indicator in the first-category feature data and the second-category feature data of each candidate object through the weights of each preset indicator, to obtain the degree of match between the candidate object and the user category.

[0010] In an exemplary embodiment of the present disclosure, the weight of each preset indicator is determined in the following manner: obtaining historical data and the corresponding click-through rate of each preset indicator of at least one of the candidate objects within a historical time period; and determining the weight of each preset indicator based on the regression relationship between the historical data and the click-through rate.

[0011] In an exemplary embodiment of the present disclosure, if the weight of the preset indicator is negative, the weight of the preset indicator is corrected to 1 / 2 of the minimum positive weight.

[0012] In an exemplary embodiment of the present disclosure, obtaining the first category feature data of multiple candidate objects includes: obtaining multiple preset indicators; based on global user data, under each of the candidate objects, counting the data of each of the preset indicators to obtain the first category feature data of each of the candidate objects.

[0013] In an exemplary embodiment of the present disclosure, determining the second type of feature data of each of the candidate objects based on the user category to which the target user belongs includes: determining the user category to which the target user belongs; and based on the user data in the user category, under each of the candidate objects, counting the data of each of the preset indicators to obtain the second type of feature data of each of the candidate objects.

[0014] In an exemplary embodiment of the present disclosure, the user category to which the target user belongs is determined in the following manner: clustering the target user with other users based on the target user's latest basic information and behavior data; and obtaining the user category to which the target user belongs after clustering.

[0015] According to one aspect of the present disclosure, there is provided an object recommendation device, comprising: a first data acquisition module, used to acquire first-category feature data of multiple candidate objects; a second data determination module, used to determine second-category feature data of each candidate object based on the user category to which a target user belongs; a matching degree determination module, used to determine the matching degree of each candidate object with the user category through the first-category feature data and the second-category feature data of each candidate object; and an object recommendation module, used to select one or more candidate objects according to the matching degree to recommend them to the target user.

[0016] In an exemplary embodiment of the present disclosure, the matching degree determination module includes: a standardization processing unit, which is used to standardize the first category feature data and the second category feature data of each candidate object; a weighted calculation unit, which is used to weight the data of each preset indicator in the first category feature data and the second category feature data of each candidate object by the weight of each preset indicator, so as to obtain the matching degree between the candidate object and the user category.

[0017] In an exemplary embodiment of the present disclosure, the weight of each preset indicator is determined in the following manner: obtaining historical data and the corresponding click-through rate of each preset indicator of at least one of the candidate objects within a historical time period; and determining the weight of each preset indicator based on the regression relationship between the historical data and the click-through rate.

[0018] In an exemplary embodiment of the present disclosure, if the weight of the preset indicator is negative, the weight of the preset indicator is corrected to 1 / 2 of the minimum positive weight.

[0019] In an exemplary embodiment of the present disclosure, the first data acquisition module includes: an indicator acquisition unit, used to acquire multiple preset indicators; a data statistics unit, used to count the data of the preset indicators under each candidate object based on global user data to obtain the first category feature data of each candidate object.

[0020] In an exemplary embodiment of the present disclosure, the second data acquisition module includes: a category determination unit, used to determine the user category to which the target user belongs; a data statistics unit, used to count the data of the preset indicators under each candidate object based on the user data in the user category to obtain the second type of feature data of each candidate object.

[0021] In an exemplary embodiment of the present disclosure, the user category to which the target user belongs is determined in the following manner: clustering the target user with other users based on the target user's latest basic information and behavior data; and obtaining the user category to which the target user belongs after clustering.

[0022] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned methods by executing the executable instructions.

[0023] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any one of the above methods is implemented.

[0024] The exemplary embodiments of the present disclosure have the following beneficial effects:

[0025] The first type of feature data of multiple candidate objects is obtained; based on the user category to which the target user belongs, the second type of feature data of each candidate object is determined; through the first type of feature data and the second type of feature data of each candidate object, the matching degree between each candidate object and the user category is determined; based on the matching degree, one or more candidate objects are selected to be recommended to the target user. On the one hand, the matching degree between the candidate object and the user category is determined through the feature data of the candidate object, and the candidate object is recommended to the target user based on the matching degree. The recommendation algorithm is simple and has high accuracy; on the other hand, based on different user categories, feature data under different categories is obtained, and object recommendation is realized in combination with user categories, and the characteristics of the candidate object and the influence of user behavior on object recommendation are comprehensively considered, so that the recommendation result is more accurate and effective, and the recommendation based on user category is more targeted.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0028] Figure 1 A flowchart of an object recommendation method in this exemplary embodiment is schematically shown;

[0029] Figure 2 A sub-flow chart schematically illustrates an object recommendation method in this exemplary embodiment;

[0030] Figure 3 A sub-flow chart schematically illustrates another object recommendation method in this exemplary embodiment;

[0031] Figure 4 A structural block diagram of an object recommendation device in this exemplary embodiment is schematically shown;

[0032] Figure 5 An electronic device for implementing the above method in this exemplary embodiment is schematically shown;

[0033] Figure 6 A computer-readable storage medium for implementing the above method in this exemplary embodiment is schematically shown. DETAILED DESCRIPTION

[0034] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0035] The exemplary embodiment of the present disclosure first provides an object recommendation method, wherein the object can be a user, commodity, store, news report, etc. on an e-commerce, social networking, etc. platform that can be used for recommendation. The application scenario of the method of this embodiment can be: in an e-commerce platform, recommending stores or commodities that may be of interest to users of different preference categories; or in an information platform, recommending news reports or hot information related to users of different interest categories.

[0036] The following is combined with Figure 1 This exemplary embodiment is further described as follows. Figure 1 As shown, the object recommendation method may include the following steps S110 to S140:

[0037] Step S110, obtaining first-category feature data of multiple candidate objects.

[0038] Among them, candidate objects refer to objects that can be used for object recommendation, which vary according to the recommendation scenario. For example, in an e-commerce platform, candidate objects can be stores, commodities or marketing activities, etc. In an information platform, candidate objects can be news reports, hot information or media advertisements, etc. Candidate objects can be all recommendable objects in the corresponding application scenario, or they can be part of the recommendable objects. For example, in an e-commerce platform, in the scenario of recommending stores to users, all stores in the e-commerce platform can be determined as candidate objects, or a part of the stores can be determined as candidate objects. The present disclosure does not make specific limitations on this. The first type of feature data refers to data that can reflect the attribute characteristics of candidate objects, which are inherent feature data possessed by all candidate objects, that is, no matter what kind of object the candidate object is, it has this attribute characteristic. For example, when the candidate object is a store, the first type of feature data can be data under indicators such as the store's order volume, logistics evaluation, and after-sales evaluation. In this exemplary embodiment, in order to effectively analyze the data, it is also possible to determine the first type of feature data within a certain time period based on the type of candidate object. For example, in a certain e-commerce platform, the top 50 stores are selected as candidate objects to obtain data such as order volume, logistics evaluation, and after-sales evaluation in the most recent quarter.

[0039] In an exemplary embodiment, the above step S110 may include the following steps:

[0040] Get multiple preset indicators;

[0041] Based on the global user data, under each candidate object, the data of each preset indicator is counted to obtain the first type of feature data of each candidate object.

[0042] Among them, the preset index refers to the attribute index of the first type of feature data. To obtain the first type of feature data is to obtain the data under the preset index. The preset index varies according to different candidate objects. For example, when the candidate object is a store, the preset index can be an attribute index related to the store, such as the average daily UV (User View), the average daily order volume, the average daily GMV (Gross Merchandise Volume), after-sales evaluation, logistics evaluation, etc. Global user data refers to all users used for data statistics. Considering that the first type of feature data is data under the inherent characteristics of the candidate object, it will not have a large difference according to the different user types. Therefore, when obtaining the first type of feature data, data statistics of multiple preset indicators can be performed based on the global user data and each candidate object as a unit to obtain the first type of feature data of multiple candidate objects.

[0043] Step S120: determining the second type of feature data of each candidate object based on the user category to which the target user belongs.

[0044] Among them, the target user refers to the main user who recommends the object, that is, the recommended object will recommend the user. In this exemplary embodiment, users can be classified according to basic user attributes (such as age, gender) or other attributes (such as preferences, attention), etc., to obtain different categories. The user category can be specifically embodied in the form of user tags or user portraits, etc., and the present disclosure does not make specific restrictions on this. The second type of feature data is similar to the first type of feature data, and both are data that can reflect the attribute characteristics of the candidate object. The difference is that the second type of feature data has a certain correlation with the user itself, and users under different categories may affect the second type of feature data of the candidate object. For example, users under category A are interested in clothing, and users under category B are interested in electronic products. Then the behavior data of users under category A in clothing stores will be more, and the behavior data in electronic product stores will be less. It can be seen that the data of clothing stores and electronic product stores under indicators such as the number of fans and fan attributes will have great differences according to different audience groups. Therefore, in order to combine the relationship between the user behavior characteristics and the content of the candidate objects, this exemplary embodiment determines the second type of feature data of the candidate objects under different user categories, so that the acquired data is more targeted and analyzable.

[0045] In an exemplary embodiment, the above step S120 may include:

[0046] Determine the user category to which the target user belongs;

[0047] Based on the user data in the user category, under each candidate object, data of preset indicators are counted to obtain the second type of feature data of each candidate object.

[0048] In this exemplary embodiment, obtaining the second type of feature data of the candidate object under the user category may be to determine a user group portrait under a category, and determine the second feature data of the candidate object under the category of the user group portrait. The preset indicators of the second type of feature data refer to attribute indicators related to the user category. Under different user categories, the data under the preset indicators may be quite different. For example, when the candidate object is a store, the preset indicators may be the number of fans, the number of valid fans, the number of users who have pinned the store, and so on.

[0049] Step S130 : determining the matching degree between each candidate object and the user category through the first type feature data and the second type feature data of each candidate object.

[0050] In this exemplary embodiment, processing can be performed based on the acquired feature data to evaluate each candidate object. The evaluation result can be used to measure whether the candidate object is suitable for recommendation, whether it meets the requirements, etc. The evaluation result is the matching degree between the candidate object and the user category. The higher the matching degree, the better the indicators of the candidate object in all aspects, and the more suitable it is to be recommended to the user. This exemplary embodiment can also establish a candidate object set based on the matching degree of each candidate object and the user category, and in the set, sort them in order of matching degree. Candidate objects with matching degrees lower than a preset threshold can also be directly discarded, and objects with matching degrees within a preset range are preferentially recommended, etc.

[0051] In an exemplary embodiment, if Figure 2 As shown, the above step S130 may include the following steps:

[0052] Step S210, standardizing the first-category feature data and the second-category feature data of each candidate object;

[0053] Step S220 , weighting the data of each preset indicator in the first type feature data and the second type feature data of each candidate object by the weight of each preset indicator, to obtain the matching degree between the candidate object and the user category.

[0054] Considering that the acquired characteristic data is data under different preset indicators, its dimension, dimension unit, and value range may be different, or even at different magnitudes. In order to facilitate subsequent data calculation and analysis and make the characteristic data comparable, this exemplary embodiment may perform standardization processing on the first type of characteristic data and the second type of characteristic data. Specifically, the data under each preset indicator may be normalized so that the data is at the same magnitude, which is convenient for comprehensive comparison.

[0055] In this exemplary embodiment, the normalization process may be performed in any of the following ways:

[0056] (1) Minmax normalization processing:

[0057] Minmax standardization is a linear transformation method, using the formula: The data under each preset indicator is processed to limit the value of each data to between [0, 1].

[0058] (2) Log standardization processing:

[0059] Log normalization is a nonlinear transformation method, which can be expressed as: By processing the data under each preset indicator, the value of the data can also be limited to between [0, 1].

[0060] In addition, this exemplary embodiment may also use other methods for standardization, such as Z-score standard deviation, etc., which is not specifically limited in this disclosure. In particular, in this exemplary embodiment, Minmax standardization may be preferably used, and if the data distribution is biased, the 99th percentile may be used as the maximum value.

[0061] Taking into account that the data under different preset indicators can reflect different aspects of the attributes of the candidate object, they have different degrees of influence on determining the matching degree between the candidate object and the user category. Therefore, this exemplary embodiment can perform weighted calculation on the data of each preset indicator according to the weight of each preset indicator to obtain the matching degree between the candidate object and the user category.

[0062] In an exemplary embodiment, if Figure 3 As shown, the weights of the above preset indicators can be determined in the following ways:

[0063] Step S310, obtaining historical data of various preset indicators of at least one candidate object and the corresponding click-through rate within a historical time period;

[0064] Step S320, determining the weight of each preset indicator based on the regression relationship between historical data and click-through rate.

[0065] Take the example of e-commerce platforms recommending stores to users. CTR (Click-Through-Rate) is an indicator parameter that can be used to reflect the popularity of store operations. It is the ratio of actual clicks to store display volume. The higher the click-through rate of a store, the more popular the store is, the more it is liked by users, or the more it can meet users' needs.

[0066] In this exemplary embodiment, the relationship between the click-through rate and the characteristic data under the above preset indicators can be quantified by using the ridge regression algorithm, and the weight of each preset indicator can be calculated by the following formula:

[0067]

[0068] Among them, y represents the click-through rate, x i Represents the characteristic data under each preset indicator, w i It represents the weight value corresponding to each preset indicator. α is the penalty degree, which is usually set to 0.5. The larger the value, the greater the penalty for more complex models.

[0069] Considering that the ridge regression algorithm is a nonlinear programming problem, this exemplary embodiment can be solved by a gradient descent algorithm and the feature data can be modeled using Python.

[0070] Take the example of recommending stores to users on an e-commerce platform. When analyzing store data, the user's attention time window can be determined by analyzing the number of fans of the store and the time distribution of fans' attention; the number of people who have paid attention in the past X days can effectively monitor the recent popularity of the store; by analyzing the revisit behavior of store fans, the concept of effective attention can be introduced to effectively distinguish between active fans and "zombie" fans among store fans; in addition, during the data analysis process, it was found that the core fans of the store usually pin their favorite stores to the top, and the number of pinned stores for a store effectively illustrates the number of core fans of the store; for the store operation level, the core indicators of the store's average daily UV, average daily number of orders and average daily GMV in the past X days can effectively reflect the store's recent business popularity; finally, through the order completion of the user's logistics evaluation, service evaluation and after-sales evaluation indicators of the store, it can effectively reflect the store's performance on the service side. Therefore, multiple preset indicators can be determined: fan score, effective attention score, top score, attention score in the past 30 days, evaluation score, logistics score, after-sales score, UV score, order score and sales score, etc., and the first feature data and the second feature data can be determined based on these preset indicators. Table 1 shows an example of the first feature data and the second feature data of the candidate stores 1 to 5 obtained:

[0071] Table 1

[0072]

[0073] Minmax normalization is used to process the feature data in Table 1. Then, sklearn in Python can be used for modeling and the weights of each preset indicator can be output to obtain the following weights:

[0074] In[5]: model.fit(x,y)

[0075] …:model.coef_

[0076] Out[5]:

[0077] Array([-0.03385396,0.3287327,0.14313122,0.06682226,-0.07337746,0.04527901,0.08966719,0.21070297,0.40227266])

[0078] That is, the preset indicators of fan score, effective attention score, top score, attention score in the past 30 days, evaluation score, logistics score, after-sales score, UV score, order score and sales score are obtained, and their weights are: -0.03385396, 0.3287327, 0.14313122, 0.06682226, -0.07337746, 0.04527901, 0.08966719, 0.21070297, 0.40227266.

[0079] In practical applications, due to some errors or other special reasons, the obtained weights may be negative. For example, in the above embodiment, the weights of two preset indicators are less than 0. Therefore, this exemplary embodiment adjusts the weights based on this situation. Specifically, if the weight of a preset indicator is negative, the weight of the preset indicator can be corrected to 1 / 2 of the minimum positive weight. For example, among the preset indicators shown in Table 2, there are two preset indicators with negative weights before correction, namely the evaluation score (-0.0339) and the after-sales score (-0.0734). The two negative weights are corrected to half of the minimum value of the non-negative weights. The corrected weight values ​​are as follows:

[0080] Table 2

[0081] Preset indicators Weight Correction weight Review Rating -0.0339 0.0226 Sales 0.3287 0.3287 Sticky 0.1431 0.1431 fan 0.0668 0.0668 After-sales rating -0.0734 0.0226 Logistics Rating 0.0453 0.0453 Order 0.0897 0.0897 Effective attention 0.2107 0.2107 UV 0.4023 0.4023

[0082] Furthermore, a scoring model is established based on the corrected weights and the feature data under each preset indicator to obtain the degree of matching. The specific scoring model can be a weighted sum of the acquired feature data. For example, the degree of matching can be determined by the formula: S=w1*score1+w2*score2+w3*score3+…+wn*scoren, where score1,…,scoren represent the feature data under the preset indicators, and w1,…,wn represent the weight values ​​corresponding to each preset indicator.

[0083] Step S140: selecting one or more candidate objects based on the matching degree to recommend to the target user.

[0084] In this exemplary embodiment, step S130 can determine the degree of match between each candidate object and the user category. Based on the degree of match, the candidate objects can be ranked. When recommending, one or more candidate objects with a preset number of rankings can be recommended to the target user. For example, the store with the highest score obtained according to the above-mentioned matching degree calculation formula is recommended to the target user. This exemplary embodiment can evaluate the candidate objects based on the user category and the characteristic data of the candidate objects, and can recommend the candidate objects with the highest degree of match under the category according to different categories of users. It combines the user behavior and the content of the recommended object to jointly determine the recommendation result, which is more effective than recommending only based on user behavior or only based on the object.

[0085] Based on the above description, in this exemplary embodiment, the first type of feature data of multiple candidate objects is obtained; based on the user category to which the target user belongs, the second type of feature data of each candidate object is determined; through the first type of feature data and the second type of feature data of each candidate object, the matching degree between each candidate object and the user category is determined; according to the matching degree, one or more candidate objects are selected to be recommended to the target user. On the one hand, the matching degree between the candidate object and the user category is determined through the feature data of the candidate object, and the candidate object is recommended to the target user based on the matching degree. The recommendation algorithm is simple and has high accuracy; on the other hand, based on different user categories, feature data under different categories are obtained, and object recommendation is realized in combination with user categories, and the characteristics of the candidate object and the influence of user behavior on object recommendation are comprehensively considered, so that the recommendation result is more accurate and effective, and the recommendation based on user category is more targeted.

[0086] In an exemplary embodiment, the user category to which the target user belongs in step S120 can be determined in the following manner:

[0087] Cluster the target user with other users based on the target user’s latest basic information and behavior data;

[0088] Get the user category to which the target user belongs after clustering.

[0089] In this exemplary embodiment, user categories can be used to label users, that is, multiple user group portraits can be generated, and each user group portrait is a user category. User group portraits are similar to user portraits, both of which label user information and abstract the overall picture of users by collecting and analyzing various information data of users. This exemplary embodiment clusters users by obtaining basic information of target users, such as age, address, occupation, payment method or consumption level, and behavioral data, such as browsing, purchasing, and collection. The clustering result is the user classification result. Each clustering result can reflect similar or identical features of users of this type. For example, in user category 1, users are aged 18 to 25 years old, buy more clothes, and have a monthly purchasing power of 3,000. Through the clustering results, user portraits of this type of users can be abstracted, which can also be regarded as user labels.

[0090] This exemplary embodiment can jointly realize object recommendation from two aspects. First, through the accumulated user data, a user group portrait is established. Then, through steps S110 to S130, a set of candidate objects under each user group portrait is determined. The candidate objects in the set can be sorted according to a certain priority. In actual application, the user portrait of the target user is determined, and object recommendation is performed in the candidate object set corresponding to the user portrait. The recommendation process is more targeted and effective.

[0091] The exemplary embodiments of the present disclosure also provide an object recommendation device. Figure 4 The device 400 may include a first data acquisition module 410 for acquiring first-category feature data of multiple candidate objects; a second data determination module 420 for determining second-category feature data of each candidate object based on the user category to which the target user belongs; a matching degree determination module 430 for determining the matching degree between each candidate object and the user category through the first-category feature data and the second-category feature data of each candidate object; and an object recommendation module 440 for selecting one or more candidate objects according to the matching degree to recommend to the target user.

[0092] In an exemplary embodiment, the matching degree determination module includes: a standardization processing unit, which is used to standardize the first-category feature data and the second-category feature data of each candidate object; a weighted calculation unit, which is used to weight the data of each preset indicator in the first-category feature data and the second-category feature data of each candidate object by using the weights of each preset indicator to obtain the matching degree between the candidate object and the user category.

[0093] In an exemplary embodiment, the weight of each preset indicator is determined in the following manner: obtaining historical data of each preset indicator of at least one candidate object and the corresponding click-through rate within a historical time period; and determining the weight of each preset indicator based on the regression relationship between the historical data and the click-through rate.

[0094] In an exemplary embodiment, if the weight of the preset indicator is negative, the weight of the preset indicator is corrected to 1 / 2 of the minimum positive weight.

[0095] In an exemplary embodiment, the first data acquisition module includes: an indicator acquisition unit for acquiring multiple preset indicators; a data statistics unit for counting the data of each preset indicator under each candidate object based on global user data to obtain the first type of feature data of each candidate object.

[0096] In an exemplary embodiment, the second data acquisition module includes: a category determination unit, used to determine the user category to which the target user belongs; a data statistics unit, used to count the data of each preset indicator under each candidate object based on the user data in the user category to obtain the second type of feature data of each candidate object.

[0097] In an exemplary embodiment, the user category to which the target user belongs is determined in the following manner: clustering the target user with other users according to the target user's latest basic information and behavior data; and obtaining the user category to which the target user belongs after clustering.

[0098] The specific details of each module / unit in the above-mentioned device have been described in detail in the embodiments of the method part. The undisclosed details can be found in the embodiments of the method part, so they will not be repeated here.

[0099] Exemplary embodiments of the present disclosure also provide an electronic device capable of implementing the above method.

[0100] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0101] Refer to the following Figure 5 An electronic device 500 according to such an exemplary embodiment of the present disclosure is described below. Figure 5 The electronic device 500 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0102] like Figure 5As shown, the electronic device 500 is in the form of a general computing device. The components of the electronic device 500 may include but are not limited to: at least one processing unit 510, at least one storage unit 520, a bus 530 connecting different system components (including the storage unit 520 and the processing unit 510), and a display unit 540.

[0103] The storage unit stores program codes, which can be executed by the processing unit 510, so that the processing unit 510 executes the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 510 can execute Figure 1 Steps S110 to S140 shown in FIG. 1 may also be performed Figure 2 Steps S210 to S220 etc. are shown.

[0104] The storage unit 520 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 521 and / or a cache memory unit 522 , and may further include a read-only memory unit (ROM) 523 .

[0105] The storage unit 520 may also include a program / utility 524 having a set (at least one) of program modules 525, such program modules 525 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0106] Bus 530 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0107] The electronic device 500 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 500, and / or communicate with any device that enables the electronic device 500 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 550. Furthermore, the electronic device 500 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 560. As shown, the network adapter 560 communicates with other modules of the electronic device 500 via a bus 530. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0108] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the exemplary embodiment of the present disclosure.

[0109] The exemplary embodiments of the present disclosure also provide a computer-readable storage medium on which a program product capable of implementing the above-mentioned method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above-mentioned "Exemplary Method" section of the present specification.

[0110] refer to Figure 6 As shown, a program product 600 for implementing the above method according to an exemplary embodiment of the present disclosure is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0111] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0112] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0113] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0114] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0115] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0116] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to exemplary embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0117] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0118] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An object recommendation method, characterized in that: include: Obtaining first-category feature data of multiple candidate objects; The obtaining of the first type of feature data of the plurality of candidate objects comprises: obtaining a plurality of preset indicators; based on the global user data, under each of the candidate objects, counting the data of each of the preset indicators to obtain the first type of feature data of each of the candidate objects; Determine the second type of feature data of each candidate object based on the user category to which the target user belongs; the determining the second type of feature data of each candidate object based on the user category to which the target user belongs includes: determining the user category to which the target user belongs; based on the user data in the user category, under each candidate object, counting the data of each preset indicator to obtain the second type of feature data of each candidate object; Determining the matching degree between each of the candidate objects and the user category through the first type of feature data and the second type of feature data of each of the candidate objects; According to the matching degree, one or more candidate objects are selected to be recommended to the target user; The determining the matching degree between each candidate object and the user category by using the first type feature data and the second type feature data of each candidate object includes: Performing standardization processing on the first-category feature data and the second-category feature data of each candidate object; By weighting the preset indicators, the data of the preset indicators in the first category feature data and the second category feature data of each candidate object are weighted to obtain the matching degree between the candidate object and the user category.

2. The method according to claim 1, characterized in that The weights of the preset indicators are determined in the following way: Acquire historical data of each of the preset indicators of at least one of the candidate objects and the corresponding click-through rate within a historical time period; Based on the regression relationship between the historical data and the click-through rate, the weight of each of the preset indicators is determined.

3. The method according to claim 2, characterized in that If the weight of the preset indicator is negative, the weight of the preset indicator is corrected to 1 / 2 of the minimum positive weight.

4. The method according to claim 1, characterized in that: The user category to which the target user belongs is determined in the following manner: Clustering the target user with other users according to the latest basic information and behavior data of the target user; The user category to which the target user belongs after clustering is obtained.

5. An object recommendation device, characterized in that: include: A first data acquisition module, used to acquire first-category feature data of multiple candidate objects; The first data acquisition module is configured to: acquire a plurality of preset indicators; Based on the global user data, under each candidate object, the data of each preset indicator is counted to obtain the first type of feature data of each candidate object; A second data determination module, used to determine the second type of feature data of each candidate object based on the user category to which the target user belongs; The second data determination module is configured to: determine the user category to which the target user belongs; Based on the user data in the user category, under each of the candidate objects, statistics are collected on the data of each of the preset indicators to obtain the second type of feature data of each of the candidate objects; a matching degree determination module, configured to determine a matching degree between each candidate object and the user category through the first type feature data and the second type feature data of each candidate object; An object recommendation module, configured to select one or more candidate objects according to the matching degree, and recommend them to the target user; The matching degree determination module is configured to: Performing standardization processing on the first-category feature data and the second-category feature data of each candidate object; By weighting the preset indicators, the data of the preset indicators in the first category feature data and the second category feature data of each candidate object are weighted to obtain the matching degree between the candidate object and the user category.

6. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to perform the method according to any one of claims 1 to 4 by executing the executable instructions.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images