Item recommendation method and device, computer storage medium, and electronic device

Abstract

The present disclosure relates to the field of data processing technology, and provides an item recommendation method, an item recommendation device, a computer storage medium, and an electronic device, wherein the item recommendation method includes: determining the target category to which the target user belongs under the preset M feature dimensions; matching the score values ​​corresponding to each feature dimension according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space; determining the weighting parameters corresponding to each target category according to the correspondence between the target categories and the weighting parameters pre-stored in the storage space; determining the consumption potential evaluation value of the target user according to the score value and the weighting parameter; and recommending target items corresponding to the consumption potential evaluation value to the target user. The present disclosure can recommend items that match the consumption capacity of a newly registered user, so that the recommended items can accurately cover the user's needs and save the user's item search time.

Description

Background Art

[0002] With the rapid development of computer and Internet technology, the relevant data processing field is also developing rapidly. For related item display platforms, if it is possible to recommend items that match the consumption capacity of different types of users, it can greatly save users' time in searching for goods and improve user satisfaction.

[0003] In the related technologies, the consumption potential of consumers is generally evaluated based on the user's existing behaviors, such as purchasing behavior, browsing behavior, attributes of purchased goods, etc., in conjunction with models such as decision trees. It can be seen that the related technologies cannot evaluate the consumption potential of newly registered users and cannot make accurate recommendations for newly registered users.

[0004] In view of this, there is an urgent need in the art to develop a new method and device for recommending items.

[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. Summary of the invention

[0006] The purpose of the present disclosure is to provide an item recommendation method, an item recommendation device, a computer storage medium and an electronic device, thereby at least to a certain extent avoiding the defect of the related art that accurate recommendations cannot be made to newly registered users.

[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 a first aspect of the present disclosure, a method for recommending items is provided, comprising: determining a target category to which a target user belongs under M preset feature dimensions; wherein the target user includes a newly registered user, and M is a positive integer; matching the score values ​​corresponding to each of the feature dimensions according to a correspondence between feature dimensions and score values ​​pre-stored in a storage space; determining the weighting parameters corresponding to each of the target categories according to a correspondence between target categories and weighting parameters pre-stored in the storage space; determining a consumption potential evaluation value of the target user according to the score values ​​and the weighting parameters; and recommending a target item corresponding to the consumption potential evaluation value to the target user.

[0009] In an exemplary embodiment of the present disclosure, determining the target category to which the target user belongs under M preset feature dimensions includes: obtaining user feature values ​​of the target user under the M feature dimensions; and determining the target category to which the target user belongs under each feature dimension based on the user feature values.

[0010] In an exemplary embodiment of the present disclosure, before determining the target category to which the target user belongs under the preset M feature dimensions, the method also includes: dividing the acquired sample users into positive sample users and negative sample users; the sample users are users who have purchased items of a specified category within a historical time period; determining the target category to which each of the sample users belongs under the M feature dimensions; determining the chi-square value corresponding to each of the feature dimensions according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; determining the scoring value corresponding to each of the feature dimensions according to the chi-square value and the total number of the sample users; and storing the feature dimensions and the scoring values ​​in the storage space in correspondence with each other.

[0011] In an exemplary embodiment of the present disclosure, the chi-square value corresponding to each of the feature dimensions is determined according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users, including: determining a first statistical value according to the number of positive sample users in each target category, the total number of positive sample users, the number of sample users included in each target category, and the total number of sample users; determining a second statistical value according to the number of negative sample users in each target category, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; and determining the cumulative value of the first statistical value and the second statistical value as the chi-square value corresponding to each of the feature dimensions.

[0012] In an exemplary embodiment of the present disclosure, the acquired sample users are divided into positive sample users and negative sample users, including: dividing the designated category of items into low-priced items and high-priced items; determining the sample users who have purchased the high-priced items as the positive sample users; and, determining the sample users who have purchased the low-priced items as the negative sample users.

[0013] In an exemplary embodiment of the present disclosure, dividing the designated category of items into low-priced items and high-priced items includes: obtaining a first quantile of multiple item prices corresponding to the designated category of items; determining items whose prices are greater than the first quantile as the high-priced items; and determining items whose prices are less than or equal to the first quantile as the low-priced items.

[0014] In an exemplary embodiment of the present disclosure, the method also includes: obtaining a first ratio of the number of positive sample users in each target category to the total number of the positive sample users; obtaining a second ratio of the number of negative sample users in each target category to the total number of the negative sample users; obtaining a third ratio of the first ratio to the second ratio; determining the natural logarithm of the third ratio as a weighted parameter corresponding to each target category; and storing the target category and the weighted parameter in the storage space correspondingly.

[0015] In an exemplary embodiment of the present disclosure, determining the consumption potential evaluation value of the target user based on the scoring value and the weighting parameter includes: obtaining the product of the scoring value and the weighting parameter under each feature dimension; and determining the sum of M products as the consumption potential evaluation value of the target user.

[0016] In an exemplary embodiment of the present disclosure, recommending the target item corresponding to the consumption potential evaluation value to the target user includes: obtaining the second quantile of multiple consumption potential evaluation values ​​corresponding to multiple target users; determining the consumption potential level of the target user based on the relationship between the consumption potential evaluation value of the target user and the second quantile; and recommending a specified target item to the target user based on a pre-stored correspondence between the consumption potential level and the target item.

[0017] According to a second aspect of the present disclosure, there is provided an item recommendation device, comprising: a determination module, for determining a target category to which a target user belongs under M preset feature dimensions; wherein the target user includes a newly registered user, and M is a positive integer; an evaluation module, for matching the score values ​​corresponding to each feature dimension according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space; determining the weighting parameters corresponding to each target category according to the correspondence between the target categories and the weighting parameters pre-stored in the storage space; determining the consumption potential evaluation value of the target user according to the score value and the weighting parameter; and an item recommendation module, for recommending a target item corresponding to the consumption potential evaluation value to the target user.

[0018] According to a third aspect of the present disclosure, there is provided a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for recommending items described in the first aspect is implemented.

[0019] According to a fourth 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 the method for recommending items described in the first aspect above by executing the executable instructions.

[0020] It can be seen from the above technical solutions that the item recommendation method, item recommendation device, computer storage medium and electronic device in the exemplary embodiments of the present disclosure have at least the following advantages and positive effects:

[0021] In the technical solutions provided by some embodiments of the present disclosure, on the one hand, the target category to which the target user belongs under the preset M feature dimensions is determined, so that the target user can be divided into different types, thereby improving the accuracy of subsequent recommendations. Further, according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space, the score values ​​corresponding to each feature dimension are matched, and according to the correspondence between the target categories and the weighted parameters pre-stored in the storage space, the weighted parameters corresponding to each target category are determined, thereby determining the consumption potential evaluation value of the target user according to the score value and the weighted parameter, providing a new idea for evaluating the user's consumption potential based on user characteristics, so that the target user can be not only a historical user, but also a newly registered user without historical behavior, solving the problem in the related art that historical users are evaluated only based on the user's existing behavior, but newly registered users cannot be evaluated. On the other hand, recommending target items corresponding to the consumption potential evaluation value to the target user can connect the item database and the user database, saving the user's time in searching for items, improving user satisfaction, and improving the user's browsing efficiency and item transaction rate.

[0022] 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

[0023] 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.

[0024] Figure 1 A schematic diagram showing a flow chart of a method for recommending items in an exemplary embodiment of the present disclosure;

[0025] Figure 2 A schematic diagram showing a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure;

[0026] Figure 3 A schematic diagram showing a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure;

[0027] Figure 4 A schematic diagram showing a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure;

[0028] Figure 5 A schematic diagram showing a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure;

[0029] Figure 6 A schematic diagram showing the overall process of a method for recommending items in an exemplary embodiment of the present disclosure;

[0030] Figure 7 A schematic diagram showing the structure of an item recommendation device in an exemplary embodiment of the present disclosure;

[0031] Figure 8 A schematic diagram showing the structure of a computer storage medium in an exemplary embodiment of the present disclosure;

[0032] Fig. 9 A schematic structural diagram of an electronic device in an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0033] 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 being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0034] The terms "a", "an", "the" and "said" are used in this specification to indicate the presence of one or more elements / components / etc.; the terms "including" and "having" are used to express an open-ended inclusion and mean that additional elements / components / etc. may exist in addition to the listed elements / components / etc.; the terms "first" and "second" etc. are used only as labels and are not intended to limit the quantity of their objects.

[0035] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and their repeated descriptions will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0036] In the related technology, the statistical learning method is generally based on the historical purchase behavior of users (for example, order amount, order quantity, number of orders, order time), browsing behavior (for example, PV of a certain category (total number of views of all pages in the store), UV (number of visitors to each page of the store, a user who visits the store multiple times in one day is counted as a visitor)) and purchased product attributes (for example, the grade of the product, the price of the product), etc., in combination with decision trees, random forests, GBDT (Gradient Boosting Decision Tree, also called MART (Multiple Additive Regression Tree, an iterative decision tree algorithm) and other models to predict the consumption potential of related users.

[0037] However, the above method has the following problems: first, the current technology is too dependent on the behavioral characteristics of users and is powerless for newly registered users who do not have behavioral characteristics; second, the decision tree algorithm processes a relatively small amount of data and is unable to perform complex calculations and iterations on large amounts of data.

[0038] In the embodiments of the present disclosure, firstly, a method for recommending items is provided, which at least to some extent overcomes the defect in the related art that accurate recommendations cannot be made to newly registered users.

[0039] Figure 1 A flow chart of a method for recommending items in an exemplary embodiment of the present disclosure is shown. The execution subject of the method for recommending items may be a server that performs a recommendation process on the items.

[0040] refer to Figure 1 According to an embodiment of the present disclosure, a method for recommending an item includes the following steps:

[0041] Step S110, determining the target category to which the target user belongs under the preset M feature dimensions;

[0042] Step S120, matching the score value corresponding to each feature dimension according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space;

[0043] Step S130, determining the weighting parameters corresponding to each target category according to the correspondence between the target categories and the weighting parameters pre-stored in the storage space;

[0044] Step S140, determining the consumption potential evaluation value of the target user according to the scoring value and the weighting parameter;

[0045] Step S150: recommending target items corresponding to the consumption potential evaluation values ​​to the target users.

[0046] exist Figure 1In the technical solution provided by the illustrated embodiment, the target category to which the target user belongs under the preset M feature dimensions is determined, and the target user can be divided into different types, thereby improving the accuracy of subsequent recommendations. Further, according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space, the score values ​​corresponding to each feature dimension are matched, and according to the correspondence between the target categories and the weighting parameters pre-stored in the storage space, the weighting parameters corresponding to each target category are determined, thereby determining the consumption potential evaluation value of the target user according to the score value and the weighting parameter, providing a new idea for evaluating the user's consumption potential based on user characteristics, so that the target user can be not only a historical user, but also a newly registered user without historical behavior, solving the problem in the related art that historical users are evaluated only based on the user's existing behavior, but newly registered users cannot be evaluated. On the other hand, recommending target items corresponding to the consumption potential evaluation value to the target user can connect the item database and the user database, save the user's time in searching for items, improve user satisfaction, and improve the user's browsing efficiency and item transaction rate.

[0047] The following Figure 1 The specific implementation process of each step is described in detail:

[0048] In an exemplary embodiment of the present disclosure, sample users may be pre-screened from historical users, and the score values ​​corresponding to each feature dimension may be determined in combination with the relevant feature information of the sample users. Figure 2 , Figure 2 A schematic diagram of a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure is shown, specifically a schematic diagram of a sub-process of calculating a user feature value corresponding to each feature dimension according to relevant information of a sample user, including steps S201-S205. Figure 2 Specific implementation methods are explained.

[0049] In step S201, the acquired sample users are divided into positive sample users and negative sample users.

[0050] In this step, sample users can be obtained, and sample users can be users who have purchased designated items in a historical period of time. Designated items can be mobile phone items, home appliances, daily necessities, etc. For example, when the designated items are mobile phone items, the designated items can be mobile phones of the same brand or mobile phones of different brands. For example, multiple mobile phones of the same brand can be Xiaomi 6, Xiaomi 7, Xiaomi 8, and other Xiaomi series mobile phones. For example, multiple mobile phones of different brands can be Xiaomi 8, iPhone 11, Huawei P40, etc., which can be set according to actual conditions and fall within the scope of protection of this disclosure.

[0051] Furthermore, the sample users can be divided into positive sample users and negative sample users according to the prices of the items purchased by each sample user. Figure 3 , Figure 3 A sub-process diagram of a method for recommending items in an exemplary embodiment of the present disclosure is shown, specifically showing how to divide sample users into positive sample users and negative sample users, including steps S301-S302. Figure 3 Specific implementation methods are explained.

[0052] In step S301, designated category items are divided into low-priced items and high-priced items.

[0053] In this step, the specified category of items includes item A, item B...item Z as an example for explanation, the price of item A is priceA, the price of item B is priceB...item Z is priceZ, and then, for example, the above priceA, priceB...priceZ can be sorted in order from small to large to form a price sequence. Then, the 0.9 quantile of the price sequence is calculated (the 0.9 quantile is also called the 0.9 quantile point, which is a set of values ​​from small to large, and the number of values ​​from the minimum value to the 0.9 quantile accounts for 90% of the number of all values), and then, the items with a price less than the 0.9 quantile are determined as low-priced items, and the items with a price greater than or equal to the 0.9 quantile are determined as high-priced items.

[0054] Exemplarily, the 0.7 quantile (the 0.7 quantile is also called the 0.7 quantile point, which is a group of values ​​​​taken from small to large, and the number of values ​​​​from the minimum value to the 0.7 quantile accounts for 70% of all the number of values ​​​​) and the 0.9 quantile of the price series can also be calculated, and then, items with prices less than or equal to the 0.7 quantile are determined as low-priced items, items with prices greater than the 0.7 quantile and less than the 0.9 quantile are determined as medium-priced items, and items with prices greater than or equal to the 0.9 quantile are determined as high-priced items.

[0055] In step S302, sample users who have purchased high-priced items are determined as positive sample users, and sample users who have purchased low-priced items are determined as negative sample users.

[0056] Thus, referring to the explanation of step S301, the sample users who purchased the high-priced items can be determined as positive sample users, and the sample users who purchased the low-priced items can be determined as negative sample users. Alternatively, the sample users who purchased the high-priced items can be determined as positive sample users, and the users who purchased the low-priced items and medium-priced items can be determined as negative sample users.

[0057] Continue to refer Figure 2 , in step S202, the target category to which each sample user belongs under M feature dimensions is determined.

[0058] In this step, the user characteristic values ​​of the sample users under the above M characteristic dimensions can be determined. For example, the M characteristic dimensions may include: user income (e.g., monthly income), user age, user city level, user technology hobby preference, user occupation, user marital status, user car brand gear, possibility of user having children, user mobile phone brand preference, user snack type preference, user child age, etc., which can be set according to actual conditions and fall within the scope of protection of this disclosure. Thus, the relevant evaluation indicators are diversified, ensuring the accuracy of subsequent evaluations.

[0059] For example, when the feature dimension is user income, it can be determined that the user feature value corresponding to user User1 is 60,000, the user feature value corresponding to user User2 is 20,000, and the user feature value corresponding to user User3 is 3000. When the feature dimension is age, it can be determined that the user feature value corresponding to user User1 is 55, the user feature value corresponding to user User2 is 30, and the user feature value corresponding to user User3 is 21.

[0060] Furthermore, the sample users may be divided into different target categories of the feature dimension according to the above user feature values. For example, when the feature dimension is "user income", the target categories may include "high income, middle income, low income" and the like.

[0061] For example, it can be set that when the user's income is greater than or equal to 50,000, the target category to which it belongs is "high income"; when the user's income is greater than 10,000 and less than 50,000, the target category to which it belongs is "middle income"; when the user's income is less than or equal to 10,000, the target category to which it belongs is "low income". Referring to the relevant explanation of the above step S203, it can be known that the target category to which user User1 belongs under the feature dimension "user income" is "high income", the target category to which user User2 belongs under the feature dimension "user income" is "middle income", and the target category to which user User3 belongs under the feature dimension "user income" is "low income".

[0062] For example, it can also be set that when the user age is greater than or equal to 60 years old, the target category to which he belongs is "elderly"; when the user age is greater than 45 years old and less than 59 years old, the target category to which he belongs is "middle-aged"; when the user age is less than or equal to 45 years old, the target category to which he belongs is "youth". Referring to the relevant explanation of the above step S203, it can be known that the target category to which user User1 belongs under the feature dimension "user age" is "middle-aged", and the target categories to which users User2 and User3 belong under the feature dimension "user age" are "youth".

[0063] In step S203, the chi-square value corresponding to each feature dimension is determined according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users.

[0064] In this step, for example, reference can be made to Figure 4 , Figure 4 A schematic diagram of a sub-process of a consumption potential evaluation method in an exemplary embodiment of the present disclosure is shown, specifically a schematic diagram of a sub-process of determining a chi-square value corresponding to a feature dimension, including steps S401-S403. Figure 4 Specific implementation methods are explained.

[0065] In step S401, a first statistical value is determined according to the number of positive sample users in each target category, the total number of positive sample users, the number of sample users included in each target category, and the total number of sample users.

[0066] For example, you can refer to the following Table 1. Taking the target category of "high income" as an example, the number of positive sample users it contains is a11, and the number of sample users it contains is P1; taking the target category of "middle income" as an example, the number of positive sample users it contains is a12, and the number of sample users it contains is P2; taking the target category of "low income" as an example, the number of positive sample users it contains is a13, and the number of sample users it contains is P3; the total number of positive sample users is R1, and the total number of sample users is N.

[0067] Table 1

[0068]

[0069] Thus, for example, for a positive sample user whose target category is "high income", the number of positive sample users belonging to "high income" can be taken as the observed frequency f0, then f0 = a11. Furthermore, the expected frequency f can be calculated based on the following formula 1: e :

[0070]

[0071] Among them, R1 is the number of positive sample users, P1 is the number of sample users belonging to "high income", and N is the total number of sample users.

[0072] After determining f0 and f e Afterwards, the statistical value T11 can be calculated based on the following formula 2:

[0073]

[0074] Similarly, for users whose target category is "middle income", a statistical value can be calculated. For users with the target category "low income" A statistical value T13 may be calculated, and further, the accumulated value of the above T11, T12, and T13 may be determined as the above first statistical value.

[0075] In step S402, a second statistical value is determined according to the number of negative sample users in each target category, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users.

[0076] Exemplarily, reference can be made to Table 1 in step S401. Taking the target category of "high income" as an example, the number of negative sample users it contains is a21, and the number of sample users it contains is P1; taking the target category of "middle income" as an example, the number of negative sample users it contains is a22, and the number of sample users it contains is P2; taking the target category of "low income" as an example, the number of positive sample users it contains is a23, and the number of sample users it contains is P3; the total number of negative sample users is R2, and the total number of sample users is N.

[0077] For example, for negative sample users whose target category is "high income", the number of negative sample users belonging to "high income" can be used as the observed frequency f0, then f0 = a21. Furthermore, the frequency f can be calculated based on the following formula 3: e :

[0078]

[0079] Among them, R2 is the number of negative sample users, P1 is the number of sample users belonging to "high income", and N is the total number of sample users.

[0080] After determining f0 and f e Afterwards, the statistical value T21 can be calculated based on the above formula 4:

[0081]

[0082] Similarly, for users whose target category is "middle income", a statistical value can be calculated. For users whose target category is "low income", a statistical value can be calculated Furthermore, the accumulated value of T21, T22, and T23 may be determined as the second statistical value.

[0083] In step S403, the cumulative value of the first statistical value and the second statistical value is determined as the chi-square value corresponding to each feature dimension.

[0084] After the first statistical value and the second statistical value are determined, the cumulative sum of the first statistical value and the second statistical value can be determined as the chi-square value X corresponding to the feature dimension “user income”. 2 .

[0085] In step S204, the score value corresponding to each feature dimension is determined according to the chi-square value and the total number of sample users.

[0086] In this step, the score C1 corresponding to the feature dimension "user income" can be calculated based on the following formula 5:

[0087]

[0088] Among them, X 2 is the chi-square value mentioned above, and N is the total number of the above sample users.

[0089] In step S205, the feature dimensions and the scoring values ​​are stored in a storage space in correspondence.

[0090] In this step, you can refer to the relevant explanations of the above steps to calculate the score values ​​corresponding to each feature dimension, and store them in the storage space accordingly. Exemplarily, the feature dimension "user income" and its corresponding score value C1 can be stored in the storage space accordingly, the feature dimension "user age" and its corresponding score value C2 can be stored in the storage space accordingly, and the feature dimension "user's city level" and its corresponding score value C3 can be stored in the storage space accordingly. It should be noted that the above storage space can be set by yourself. For example, the storage space can be a user database, an optical drive, a mobile storage device, a computer disk (for example: E disk), a specific folder, the internal storage space of a mobile phone, an external secure digital card (Secure Digital Card, SD card) of a mobile phone, etc.

[0091] For example, the weighted parameters corresponding to different target categories under each feature dimension can also be calculated separately. For details, please refer to Figure 5 , Figure 5A schematic diagram of a sub-process of a method for recommending items in an exemplary embodiment of the present disclosure is shown, specifically a schematic diagram of a sub-process for calculating weighted parameters corresponding to different target categories, including steps S501-S505. Figure 5 Specific implementation methods are explained.

[0092] In step S501, a first ratio of the number of positive sample users in each target category to the total number of positive sample users is obtained.

[0093] In this step, the feature dimension of "user income" is still used as an example for explanation. When the target category is "high income", a first ratio a11 / R1 of the number of positive sample users a11 to the total number of positive sample users R1 can be obtained.

[0094] In step S502, a second ratio of the number of negative sample users in each target category to the total number of negative sample users is obtained.

[0095] In this step, when the target category is “high income”, a second ratio a12 / R2 of the number a12 of negative sample users to the total number R2 of negative sample users may be obtained.

[0096] In step S503, a third ratio of the first ratio to the second ratio is obtained.

[0097] In this step, a third ratio of the first ratio and the second ratio can be obtained, that is,

[0098] In step S504, the natural logarithm of the third ratio is determined as a weighting parameter corresponding to each target category.

[0099] In this step, the natural logarithm of the third ratio can be determined as the weighted parameter W11 corresponding to the target category “high income”, that is,

[0100] In step S505, the target categories and weighted parameters are stored in a storage space in correspondence.

[0101] For example, the weighted parameter corresponding to the target category of "middle income" can also be calculated. Weight parameter W13 for target category “low income”

[0102] Furthermore, "high income" and its corresponding weighted parameter W11 can be stored in the storage space accordingly, "middle income" and its corresponding weighted parameter W12 can be stored in the storage space accordingly, and "low income" and its corresponding weighted parameter W13 can be stored in the storage space accordingly.

[0103] Similarly, when the feature dimension is "user age", the weighted parameter W21 corresponding to "elderly" can be stored in the storage space, the weighted parameter W22 corresponding to "middle-aged" can be stored in the storage space, and the weighted parameter W23 corresponding to "youth" can be stored in the storage space.

[0104] Continue to refer Figure 1 In step S110, the target category to which the target user belongs under the preset M feature dimensions is determined.

[0105] In this step, the user feature values ​​of the target user under M feature dimensions can be obtained, and the target category to which the target user belongs under each feature dimension can be determined according to the user feature values. For details, reference can be made to the relevant explanation of the above step S202, which will not be described in detail in this disclosure. Thus, the target users can be divided into different types, thereby improving the accuracy of subsequent recommendations.

[0106] It should be noted that the target user can be a historical user or a newly registered user. Thus, a new idea of ​​evaluating the consumption potential of a user based on user characteristics is provided, which solves the problem that the related art only evaluates historical users based on the user's existing behavior, but cannot evaluate newly registered users.

[0107] When the target user is a historical user, the user feature values ​​of the target user in the preset M feature dimensions can be determined through the user portrait, and then the target category to which the target user belongs in each feature dimension can be determined according to the user feature values.

[0108] When the target user is a historical user, a questionnaire can also be sent to the historical user, and the user characteristic values ​​corresponding to each characteristic dimension can be determined based on the questionnaire results fed back by the target user, and the target category to which the target user belongs under each characteristic dimension can be determined based on the user characteristic values.

[0109] When the target user is a newly registered user, the user's registration information may be obtained, and the user feature values ​​corresponding to each feature dimension may be determined based on the registration information, and the target category to which the target user belongs under each feature dimension may be determined based on the user feature values.

[0110] In step S120, the score value corresponding to each feature dimension is matched according to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space.

[0111] In this step, the score values ​​corresponding to each feature dimension can be read from the storage space (e.g., user database) with reference to the explanation of step S205. For example, when the feature dimension is the same as the M feature dimensions, the score values ​​corresponding to the feature dimension can be read as C1, C2, C3, ... CM.

[0112] In step S130, the weighting parameter corresponding to the target category is determined according to the pre-stored correspondence between the target category and the weighting parameter.

[0113] In this step, the weighted parameters corresponding to each target category can be read from the above storage space (for example, user database) with reference to the relevant explanation of the above step S505. For example, when the target category to which the target user belongs under the feature dimension "user income" is "high income", the weighted parameter corresponding to the feature dimension "user income" can be read as W11; when the target category to which the user age belongs is "middle-aged", the weighted parameter corresponding to the feature dimension "user age" can be read as W22, and so on. Similarly, the weighted parameters corresponding to the target categories to which other feature dimensions belong can be read.

[0114] In step S140, the consumption potential evaluation value of the target user is determined according to the scoring value and the weighting parameter.

[0115] After determining the above-mentioned scoring values ​​and corresponding weighting parameters, the consumption potential evaluation value S of the target user can be calculated based on the following formula 6:

[0116] S=C1*W11+C2*W22+……+CM*WMj Formula 6

[0117] Among them, S represents the above-mentioned consumption potential evaluation value, C1 represents the scoring value corresponding to the first feature dimension, W11 represents the weighting parameter corresponding to the target category 1 to which the target user belongs under the first feature dimension; C2 represents the scoring value corresponding to the second feature dimension, W22 represents the weighting parameter corresponding to the target category 2 to which the target user belongs under the second feature dimension; CM represents the scoring value corresponding to the Mth feature dimension, and WMj represents the weighting parameter corresponding to the target category j to which the target user belongs under the Mth feature dimension.

[0118] Furthermore, the consumption potential evaluation value of the target user can be written into the above-mentioned user database to avoid data loss.

[0119] In step S150, target items corresponding to the consumption potential evaluation values ​​are recommended to the target users.

[0120] In this step, target items corresponding to the consumption potential evaluation value can be recommended to the target user. Exemplarily, the display of the specified items in the current recommendation list of the target user can be cancelled according to the above consumption potential evaluation value to avoid the interface occupation of invalid items (i.e., items that do not match the user's consumption potential, such as top luxury goods recommended for users with low consumption potential, or discounted items recommended for target users with high consumption potential) and optimize the user experience.

[0121] For example, when there are multiple target users, the consumption potential evaluation values ​​of the multiple target users can be sorted in ascending order to form a numerical sequence. Then, the 0.9 quantile corresponding to the numerical sequence can be calculated, and the target users whose consumption potential evaluation values ​​are greater than the 0.9 quantile are determined as "high consumption potential users", and the target users whose consumption potential evaluation values ​​are less than or equal to the above 0.9 quantile are determined as "low consumption potential users".

[0122] Exemplarily, the 0.7 quantile and 0.9 quantile corresponding to the numerical sequence can also be calculated, and then the target users whose consumption potential evaluation value is greater than or equal to the above 0.9 quantile are determined as "high consumption potential users", the target users whose consumption potential evaluation value is greater than the above 0.7 quantile and less than the above 0.9 quantile are determined as "medium consumption potential users", and the target users whose consumption potential evaluation value is less than or equal to the above 0.7 quantile are determined as "low consumption potential users".

[0123] Furthermore, the target item can be recommended to the target user according to the correspondence between the pre-stored consumption potential level and the target item. For example, if the target user is the above-mentioned "high consumption potential user", "a certain brand of high-end configuration mobile phone, a certain brand of high-end configuration computer" and the like can be recommended to the target user to match his consumption capacity. Thus, the user's interest in the target item can be increased, the user's time in searching for items can be saved, the user's satisfaction can be improved, and the user's browsing efficiency and item transaction rate can be improved.

[0124] Exemplarily, the information of the above-mentioned target items can also be read from the item database, and an association relationship between the target item and the target user can be established, thereby connecting the item database and the user database, and recommending different target items to the target user in different time periods, avoiding repeated determination of the target user's consumption potential assessment value within a preset time period (for example: one year), saving the system's computing resources, and improving the system's processing efficiency for other businesses.

[0125] It should be noted that in the present disclosure, the characteristic dimensions of the sample users and the user characteristic values ​​of the sample users under the above M characteristic dimensions can also be input into a machine learning model (for example, a decision tree, random forest, reinforcement learning, etc.), and the parameters can be adjusted multiple times to train the above machine learning model so that the loss function of the above machine learning model tends to converge, thereby obtaining a consumption potential prediction model. Furthermore, the user characteristic values ​​of the target user in each characteristic dimension can be input into the above trained consumption potential evaluation model, and the consumption potential evaluation value corresponding to the above target user can be obtained according to the output of the above consumption potential evaluation model. Furthermore, the target items that match the consumption ability of the target user can be recommended to the target user according to the consumption potential evaluation value.

[0126] For example, please refer to Figure 6 , Figure 6 The overall flow chart of the method for recommending items in an exemplary embodiment of the present disclosure is shown, including steps S601-S605. Figure 6 Specific implementation methods are explained.

[0127] In step S601, M feature dimensions are screened according to user attributes of sample users;

[0128] In step S602, the sample users are divided into positive sample users and negative sample users; the score values ​​(C coefficients, also known as column connection coefficients) corresponding to each feature dimension of different sample users are calculated respectively, as well as the weighting parameters (WOE codes, Weight of Evidence, abbreviated as WOE, weight of evidence) corresponding to different target categories of each feature dimension; and the data are stored in the storage space accordingly;

[0129] In step S603, according to the user feature value of the target user in each feature dimension, the score value corresponding to each feature dimension is matched, and the weighting parameter corresponding to the target category is matched;

[0130] In step S604, the consumption potential evaluation value of the target user is determined according to the scoring value and the weighting parameter;

[0131] In step S605, the consumption potential level of the target user is determined according to the consumption potential evaluation value.

[0132] Based on the above technical solutions, the present disclosure can not only solve the technical problem in the related art that items can only be recommended to users based on their existing behaviors (for example, purchasing behaviors and browsing behaviors) but cannot be recommended to newly registered users, but can also recommend items that match their consumption capabilities to users, so that the recommended items can accurately cover user needs, save users' time in searching for items, and improve users' interest in target items, user satisfaction, user browsing efficiency and item transaction rate.

[0133] The present disclosure also provides an item recommendation device, Figure 7 A schematic diagram showing the structure of an item recommendation device in an exemplary embodiment of the present disclosure; Figure 7 As shown, the item recommendation device 700 may include a determination module 701, an evaluation module 702 and an item recommendation module 703. Among them:

[0134] The determination module 701 is used to determine the target category to which the target user belongs under the preset M feature dimensions.

[0135] In an exemplary embodiment of the present disclosure, the determination module is used to obtain user feature values ​​of the target user in M ​​feature dimensions; and determine the target category to which the target user belongs in each feature dimension according to the user feature values.

[0136] Evaluation module 702 is used to match the scoring value corresponding to each feature dimension according to the correspondence between the feature dimension and the scoring value pre-stored in the storage space; determine the weighting parameter corresponding to each target category according to the correspondence between the target category and the weighting parameter pre-stored in the storage space; and determine the consumption potential evaluation value of the target user according to the scoring value and the weighting parameter.

[0137] In an exemplary embodiment of the present disclosure, an evaluation module is used to divide the acquired sample users into positive sample users and negative sample users; the sample users are users who have purchased items of a specified type within a historical time period; determine the target category to which each sample user belongs under M feature dimensions; determine the chi-square value corresponding to each feature dimension according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; determine the score value corresponding to each feature dimension according to the chi-square value and the total number of sample users; and store the feature dimensions and the score values ​​in a storage space in correspondence with each other.

[0138] In an exemplary embodiment of the present disclosure, an evaluation module is used to determine a first statistical value based on the number of positive sample users in each target category, the total number of positive sample users, the number of sample users included in each target category, and the total number of sample users; determine a second statistical value based on the number of negative sample users in each target category, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; and determine the cumulative value of the first statistical value and the second statistical value as the chi-square value corresponding to each feature dimension.

[0139] In an exemplary embodiment of the present disclosure, the evaluation module is used to classify items of a specified category into low-priced items and high-priced items; determine sample users who have purchased high-priced items as positive sample users; and determine sample users who have purchased low-priced items as negative sample users.

[0140] In an exemplary embodiment of the present disclosure, the evaluation module is used to obtain the first quantile of multiple item prices corresponding to a specified category of items; determine items whose prices are greater than the first quantile as high-priced items; and determine items whose prices are less than or equal to the first quantile as low-priced items.

[0141] In an exemplary embodiment of the present disclosure, the evaluation module is used to obtain a first ratio of the number of positive sample users in each target category to the total number of positive sample users; obtain a second ratio of the number of negative sample users in each target category to the total number of negative sample users; obtain a third ratio of the first ratio to the second ratio; determine the natural logarithm of the third ratio as a weighted parameter corresponding to each target category; and store the target category and the weighted parameter in a storage space correspondingly.

[0142] The item recommendation module 703 is used to recommend target items corresponding to the consumption potential evaluation values ​​to the target users.

[0143] In an exemplary embodiment of the present disclosure, the item recommendation module is used to obtain the product of the score value and the weighting parameter under each feature dimension; and the sum of M products is determined as the consumption potential evaluation value of the target user.

[0144] In an exemplary embodiment of the present disclosure, the item recommendation module is used to obtain the second quantiles of multiple consumption potential evaluation values ​​corresponding to multiple target users; determine the consumption potential level of the target user based on the relationship between the consumption potential evaluation value of the target user and the second quantile; and recommend specified target items to the target user based on the pre-stored correspondence between the consumption potential level and the target item.

[0145] The specific details of each module in the above-mentioned item recommendation device have been described in detail in the corresponding item recommendation method, so they will not be repeated here.

[0146] 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 the 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.

[0147] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0148] 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 mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0149] In the exemplary embodiments of the present disclosure, a computer storage medium capable of implementing the above method is also provided. A program product capable of implementing the above method of this specification is stored thereon. In some possible embodiments, 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 cause the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification.

[0150] refer to Figure 8 As shown, a program product 800 for implementing the above method according to an 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. 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.

[0151] 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 (non-exhaustive list) of readable storage media 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.

[0152] 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.

[0153] 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.

[0154] 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).

[0155] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0156] It will be appreciated by those skilled in the art 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 implementations, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0157] Refer to the following Fig. 9 The electronic device 900 according to this embodiment of the present disclosure is described. Fig. 9 The electronic device 900 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.

[0158] like Fig. 9As shown, the electronic device 900 is in the form of a general computing device. The components of the electronic device 900 may include, but are not limited to: the at least one processing unit 910, the at least one storage unit 920, a bus 930 connecting different system components (including the storage unit 920 and the processing unit 910), and a display unit 940.

[0159] The storage unit stores a program code, which can be executed by the processing unit 910, so that the processing unit 910 performs 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 910 can perform the following steps: Figure 1 As shown in: Step S110, determine the target category to which the target user belongs under the preset M feature dimensions; Step S120, match the score value corresponding to each feature dimension according to the correspondence between the feature dimensions and the score value pre-stored in the storage space; Step S130, determine the weighting parameters corresponding to each target category according to the correspondence between the target category and the weighting parameters pre-stored in the storage space; Step S140, determine the consumption potential evaluation value of the target user according to the score value and the weighting parameter; Step S150, recommend the target item corresponding to the consumption potential evaluation value to the target user.

[0160] The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 9201 and / or a cache storage unit 9202 , and may further include a read-only storage unit (ROM) 9203 .

[0161] The storage unit 920 may also include a program / utility 9204 having a set (at least one) of program modules 9205, such program modules 9205 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.

[0162] Bus 930 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.

[0163] The electronic device 900 may also communicate with one or more external devices 1000 (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 900, and / or communicate with any device that enables the electronic device 900 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 950. Furthermore, the electronic device 900 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 960. As shown, the network adapter 960 communicates with other modules of the electronic device 900 via a bus 930. 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 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0164] 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 implementation of the present disclosure.

[0165] 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.

[0166] 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.

Claims

1. A method for recommending an item, characterized in that: include: Divide the obtained sample users into positive sample users and negative sample users; The sample users are users who have purchased items of a specified type in a historical period; The step of dividing the acquired sample users into positive sample users and negative sample users includes: Classify the designated category of items into low-priced items and high-priced items; Determine the sample users who have purchased the high-priced items as the positive sample users; and, Determine the sample users who have purchased the low-priced items as the negative sample users; Determine the target category to which each of the sample users belongs under the M feature dimensions; Obtaining a first ratio of the number of positive sample users in each of the target categories to the total number of the positive sample users; Obtaining a second ratio of the number of negative sample users in each of the target categories to the total number of negative sample users; Obtaining a third ratio of the first ratio to the second ratio; Determine the natural logarithm of the third ratio as a weighting parameter corresponding to each of the target categories; Storing the target category and the weighted parameter in a storage space in correspondence; Determine the target category to which the target user belongs under the preset M feature dimensions; wherein the target user includes a newly registered user, and M is a positive integer; According to the correspondence between the feature dimensions and the score values ​​pre-stored in the storage space, matching the score values ​​corresponding to the feature dimensions; Determining the weighted parameters corresponding to each target category according to the correspondence between the target categories and the weighted parameters pre-stored in the storage space; Determining the consumption potential evaluation value of the target user according to the scoring value and the weighting parameter; A target item corresponding to the consumption potential evaluation value is recommended to the target user.

2. The method according to claim 1, characterized in that: The determining the target category to which the target user belongs under the preset M feature dimensions includes: Obtain user feature values ​​of the target user in the M feature dimensions; The target category to which the target user belongs under each feature dimension is determined according to the user feature value.

3. The method according to claim 1 or 2, characterized in that: Before determining the target category to which the target user belongs under the preset M feature dimensions, the method further includes: Determine the chi-square value corresponding to each of the feature dimensions according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; Determine the score value corresponding to each of the feature dimensions according to the chi-square value and the total number of the sample users; The feature dimensions and the scoring values ​​are stored in the storage space in correspondence.

4. The method according to claim 3, characterized in that Determining the chi-square value corresponding to each of the feature dimensions according to the number of positive sample users in each target category, the number of negative sample users in each target category, the total number of positive sample users, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users includes: Determine a first statistical value according to the number of positive sample users in each target category, the total number of positive sample users, the number of sample users included in each target category, and the total number of sample users; Determine a second statistical value according to the number of negative sample users in each target category, the total number of negative sample users, the number of sample users included in each target category, and the total number of sample users; The cumulative value of the first statistical value and the second statistical value is determined as the chi-square value corresponding to each of the feature dimensions.

5. The method according to claim 1, characterized in that The classifying the designated items into low-priced items and high-priced items includes: Obtain the first quantile of the prices of multiple items corresponding to the specified category of items; Determine the items whose prices are greater than the first quantile as the high-priced items; Items whose prices are less than or equal to the first quantile are determined as the low-priced items.

6. The method according to claim 1, characterized in that Determining the consumption potential evaluation value of the target user according to the scoring value and the weighting parameter includes: Obtaining the product of the score value and the weighting parameter under each feature dimension; The sum of the M products is determined as the consumption potential evaluation value of the target user.

7. The method according to claim 1, characterized in that The recommending the target item corresponding to the consumption potential evaluation value to the target user includes: Obtain the second quantiles of multiple consumption potential evaluation values ​​corresponding to multiple target users; Determining the consumption potential level of the target user according to the relationship between the consumption potential evaluation value of the target user and the second quantile; According to the pre-stored correspondence between the consumption potential level and the target item, a designated target item is recommended to the target user.

8. An item recommendation device, characterized in that: include: A determination module is used to divide the obtained sample users into positive sample users and negative sample users; The sample users are users who have purchased items of a specified type in a historical period; The determination module is further used to classify the designated category of items into low-priced items and high-priced items; Determine the sample users who have purchased the high-priced items as the positive sample users; and, Determine the sample users who have purchased the low-priced items as the negative sample users; The determination module is further used to determine the target category to which each of the sample users belongs under the M feature dimensions; Obtaining a first ratio of the number of positive sample users in each of the target categories to the total number of the positive sample users; Obtaining a second ratio of the number of negative sample users in each of the target categories to the total number of negative sample users; Obtaining a third ratio of the first ratio to the second ratio; Determine the natural logarithm of the third ratio as a weighting parameter corresponding to each of the target categories; Storing the target category and the weighted parameter in a storage space in correspondence; and determining the target category to which the target user belongs under the preset M feature dimensions; wherein the target user includes a newly registered user, and M is a positive integer; An evaluation module, used to match the score value corresponding to each feature dimension according to the correspondence between the feature dimension and the score value pre-stored in the storage space; determine the weighting parameter corresponding to each target category according to the correspondence between the target category and the weighting parameter pre-stored in the storage space; determine the consumption potential evaluation value of the target user according to the score value and the weighting parameter; The item recommendation module is used to recommend target items corresponding to the consumption potential evaluation value to the target user.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for recommending an item according to any one of claims 1 to 7 is implemented.

10. 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 execute the method for recommending an item according to any one of claims 1 to 7 by executing the executable instructions.

Images