Product Recommendation Calculation Method, Device and Equipment

Abstract

A product recommendation calculation method, device, and equipment provided by the present application relate to the technical field of product recommendation systems. By obtaining at least one attribute information of each product, determining an initial popularity value of each product according to at least one attribute information of each product; obtaining a statistical popularity value of each product within a preset time period; wherein the statistical popularity value is calculated according to user behavior within the preset time period; inputting at least one attribute information of each product into a preset model to determine a predicted popularity value of each product; determining a comprehensive popularity value of each product according to the initial popularity value, the statistical popularity value, and the predicted popularity value of each product; and recommending products to users according to the comprehensive popularity values of each product. By adopting the technical solution, the product popularity values of all launched products can be accurately calculated, the conversion rate of products on the popular list can be improved, and thus it is convenient to recommend products to users.

Description

Technical Field

[0001] This application relates to the technical field of product recommendation systems, and particularly to a product recommendation calculation method, device, and equipment. Background Art

[0002] Traditional financial product popularity lists rely on statistical thinking, obtaining various statistical product popularity lists based on product trading volume, product view volume, and product attention volume, and recommending them to users.

[0003] However, the current method is very unfriendly to newly launched products or products with short product cycles. Because the product trading volume, product view volume, and product attention volume of newly launched products or products with short product cycles are, to some extent, inferior to those of products with long product cycles, this will result in inaccuracies in the product list results.

[0004] Therefore, there is an urgent need for a heat value calculation method that can accurately calculate the product heat values of all launched products, improve the conversion rate of products on the popular list, and thus facilitate product recommendation to users. Summary of the Invention

[0005] This application provides a product recommendation calculation method, device, and equipment that can accurately calculate the product heat values of all launched products, improve the conversion rate of products on the popular list, and thus facilitate product recommendation to users.

[0006] On the one hand, this application provides a product recommendation method, which includes:

[0007] Obtain at least one attribute information of each product, and determine the initial heat value of each product according to the at least one attribute information of each product;

[0008] Obtain the statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated based on user behavior within the preset time period;

[0009] Input the at least one attribute information of each product into a preset model to determine the predicted heat value of each product;

[0010] Determine the comprehensive heat value of each product according to the initial heat value, the statistical heat value, and the predicted heat value of each product;

[0011] Recommend products to users according to the comprehensive heat values of the products.

[0012] Optionally, obtaining at least one attribute information of each product and determining the initial heat value of each product according to the at least one attribute information of each product includes:

[0013] Obtain at least one attribute information of each product, and determine the type of each attribute information of each product;

[0014] According to the type of each attribute information of each product, determine the value of the heat value weight corresponding to the type;

[0015] According to the value of the heat value weight corresponding to the type, determine the initial heat value of each product.

[0016] Optionally, determining the initial heat value of each product according to the value of the heat value weight corresponding to the type includes:

[0017] Sum up the values of the heat value weights corresponding to the type to obtain a first processing result;

[0018] Perform normalization processing on the first processing result to obtain the initial heat value of each product.

[0019] Optionally, before inputting at least one attribute information of each product into a preset model to determine the predicted heat value of each product, it further includes:

[0020] Obtain the preset model according to the statistical heat value of each product within a preset time period.

[0021] Optionally, obtaining the preset model according to the statistical heat value of each product within a preset time period includes:

[0022] Set the products with high statistical heat value ranking of each product within a preset time period as positive samples, and set the products with low statistical heat value ranking of each product within a preset time period as negative samples;

[0023] Classify at least one attribute information of each product within a preset time period and perform numerical processing on the classified values to obtain a second processing result;

[0024] Train the second processing result according to a preset algorithm to obtain the preset model.

[0025] Optionally, the number of products with high statistical heat value ranking within a preset time period is the same as the number of products with low statistical heat value ranking within a preset time period.

[0026] Optionally, obtaining the statistical heat value of each product within a preset time period includes:

[0027] According to the type of user behavior obtained, assign different weight values;

[0028] Determine the current statistical popularity value according to the number of times of the user behavior and the weight value;

[0029] Determine the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the statistical popularity value in the same historical period.

[0030] Optionally, determining the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the statistical popularity value in the same historical period includes:

[0031] Calculate the time decay factor of the current statistical popularity value and the time decay factor of the statistical popularity value in the same historical period;

[0032] Determine the statistical popularity value of each product within a preset time period according to the value of the current statistical popularity value, the time decay factor of the current statistical popularity value, the value of the statistical popularity value in the same historical period, and the time decay factor of the statistical popularity value in the same historical period.

[0033] On the other hand, the present application provides a product recommendation device, including:

[0034] An attribute information acquisition module, configured to acquire at least one attribute information of each product, and determine the initial popularity value of each product according to the at least one attribute information of each product;

[0035] A statistical popularity value acquisition module, configured to acquire the statistical popularity value of each product within a preset time period; wherein, the statistical popularity value is calculated according to user behaviors within a preset time period;

[0036] A predicted popularity value determination module, configured to input at least one attribute information of each product into a preset model to determine the predicted popularity value of each product;

[0037] A comprehensive popularity value determination module, configured to determine the comprehensive popularity value of each product according to the initial popularity value, the statistical popularity value, and the predicted popularity value of each product;

[0038] A recommendation module, configured to recommend products to a user according to the comprehensive popularity values of the products.

[0039] Optionally, the attribute information acquisition module includes:

[0040] A type determination unit, configured to acquire at least one attribute information of each product and determine the type of each attribute information of each product;

[0041] A popularity value weight value determination unit, configured to determine the value of the popularity value weight corresponding to the type according to the type of each attribute information of each product;

[0042] An initial heat value determination unit, configured to determine the initial heat value of each product according to the numerical value of the heat value weight corresponding to the type.

[0043] Optionally, the initial heat value determination unit includes:

[0044] A first processing result determination subunit, configured to sum up the numerical values of the heat value weights corresponding to the type to obtain a first processing result;

[0045] An initial heat value determination subunit, configured to perform normalization processing on the first processing result to obtain the initial heat value of each product.

[0046] Optionally, the device further includes:

[0047] A preset model determination module, configured to obtain the preset model according to the statistical heat values of each product within a preset time period.

[0048] The preset model determination module includes:

[0049] A setting unit, configured to set the products with high statistical heat values within a preset time period as positive samples, and set the products with low statistical heat values within a preset time period as negative samples;

[0050] A second processing result unit, configured to classify at least one attribute information of each product within a preset time period and perform numerical processing on the classified numerical values to obtain a second processing result;

[0051] A preset model determination unit, configured to train the second processing result according to a preset algorithm to obtain the preset model.

[0052] Optionally, the number of products with high statistical heat values within a preset time period is the same as the number of products with low statistical heat values within a preset time period.

[0053] The statistical heat value acquisition module includes:

[0054] A weight numerical value determination unit, configured to allocate different weight numerical values according to the obtained types of user behaviors;

[0055] A current statistical heat value determination unit, configured to determine the current statistical heat value according to the number of times of the user behavior and the weight numerical value;

[0056] A statistical heat value determination unit, configured to determine the statistical heat value of each product within a preset time period according to the current statistical heat value and the historical statistical heat value in the same period.

[0057] Optionally, the statistical heat value determination unit includes:

[0058] A calculation subunit, configured to calculate the time decay factor of the current statistical heat value and the time decay factor of the historical same - period statistical heat value;

[0059] A statistical heat value determination subunit, configured to determine the statistical heat value of each product within a preset time period according to the value of the current statistical heat value, the time decay factor of the current statistical heat value, the value of the historical same - period statistical heat value, and the time decay factor of the historical same - period statistical heat value.

[0060] A product recommendation calculation method, device, and equipment provided by the present application obtain at least one attribute information of each product, determine the initial heat value of each product according to the at least one attribute information of each product; obtain the statistical heat value of each product within a preset time period; where the statistical heat value is calculated according to user behavior within the preset time period; input the at least one attribute information of each product into a preset model to determine the predicted heat value of each product; determine the comprehensive heat value of each product according to the initial heat value, the statistical heat value, and the predicted heat value of each product; and recommend products to users according to the comprehensive heat values of the products. By adopting this technical solution, the product heat values of all launched products can be accurately calculated, the conversion rate of products on the popular list can be improved, and it is thus convenient to recommend products to users. Description of the Drawings

[0061] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application and used together with the specification to explain the principles of the present application.

[0062] Figure 1 is a flowchart of a product recommendation method provided by Embodiment 1 of the present application;

[0063] Figure 2 is a flowchart of a product recommendation method provided by Embodiment 2 of the present application;

[0064] Figure 3 is a flowchart of a product recommendation method provided by Embodiment 3 of the present application;

[0065] Figure 4 is a schematic diagram of a product recommendation device provided by Embodiment 4 of the present application;

[0066] Figure 5 is a schematic diagram of a product recommendation device provided by Embodiment 5 of the present application;

[0067] Figure 6 It is a block diagram of a terminal device shown according to an exemplary embodiment.

[0068] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0069] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0070] The specific application scenario of the present application is to recommend products in the recommended list of financial products. In the recommended list, due to the different product launch times and product popularity levels, the heat of the products is different.

[0071] A product recommendation method provided by the present application aims to solve the above technical problems in the prior art.

[0072] The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0073] Figure 1 It is a schematic flowchart of a product recommendation method provided according to Embodiment 1 of the present application. Embodiment 1 includes the following steps:

[0074] S101. Obtain at least one piece of attribute information of each product, and determine the initial heat value of each product according to at least one piece of attribute information of each product.

[0075] Exemplarily, the attribute information is information used to characterize the current characteristics of the product. There can be multiple pieces of attribute information for the same product. For example, the attribute information in this embodiment may include the product cycle, product yield rate, minimum purchase amount, risk level, redemption period, and fund company. Among them, the types of the product cycle are also multiple, which can be calculated on a daily basis, on a monthly basis, or on a quarterly basis. Further, the types of the risk level are also multiple, which can be a low risk level, a medium risk level, and a high risk level.

[0076] For each attribute information of each product, there corresponds a different initial heat value, which can be pre-set by the user. It should be noted that the range of the initial heat value is between 0 and 1. If the user sets the value to 0.8, it means the higher the initial heat value.

[0077] S102. Obtain the statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated based on user behavior within the preset time period.

[0078] In this embodiment, the preset time period is calculated starting from the current moment and ending at a certain past moment. For example, if the current moment is September 30, 2021, and a certain past moment is June 30, 2021, then the preset time period is 3 months, and it is necessary to calculate the statistical heat value of each product within 3 months. Further, the statistical heat value for these 3 months is calculated based on the operations of each product by the user within these 3 months. Among them, the higher the popularity of the product among users, the higher the statistical heat value. It should be noted that the range of the statistical heat value is also between 0 and 1. The closer it is to 1, the higher the statistical heat value, which also indicates that the popularity among users within the preset time period is higher.

[0079] For example, a product with a statistical heat value of 0.8 is more popular among users within the preset time period than a product with a statistical heat value of 0.3.

[0080] S103. Input at least one attribute information of each product into a preset model to determine the predicted heat value of each product.

[0081] Exemplarily, the preset model can be a Naive Bayes model or a Gradient Boosting Decision Tree model. Among them,

[0082] The Naive Bayes model is a corresponding simplification based on the Bayesian algorithm, that is, it is assumed that attributes are conditionally independent of each other given the target value. That is to say, no attribute variable has a large proportion for the decision result, nor does any attribute variable have a small proportion for the decision result. Although this simplification method reduces the classification effect of the Bayesian classification algorithm to a certain extent, in actual application scenarios, it greatly simplifies the complexity of the Bayesian method. The Gradient Boosting Decision Tree model is an iterative decision tree algorithm, which consists of multiple decision trees, and the conclusions of all trees are accumulated to make the final answer.

[0083] In this embodiment, the predicted heat value is output by the preset model, and it predicts the heat value of each product based on historical situations. Specifically, the predicted heat values of each product are different.

[0084] S104. Determine the comprehensive heat value of each product based on the initial heat value, statistical heat value, and predicted heat value of each product.

[0085] Exemplarily, after obtaining the initial heat value, statistical heat value, and predicted heat value of each product, the comprehensive heat value of the product can be calculated based on these three values. If the product is a newly launched product, the value of the initial heat value can be set larger, so as to ensure that the total heat value of the product is relatively high, which is beneficial to the recommendation of new products.

[0086] S105. Recommend products to users according to the comprehensive heat values of each product.

[0087] In this embodiment, after obtaining the total heat values of each product, the products are recommended according to the ranking of the comprehensive heat values. Specifically, the higher the comprehensive heat value of the product, the more preferably it can be recommended to the user. For example, if the ranking of the comprehensive heat values is Product A, Product B, and Product C, then Product A can be preferably recommended to the user.

[0088] A product recommendation calculation method provided by this application determines the initial heat value of each product by obtaining at least one attribute information of each product; obtains the statistical heat value of each product within a preset time period; wherein the statistical heat value is calculated based on user behavior within the preset time period; inputs at least one attribute information of each product into a preset model to determine the predicted heat value of each product; determines the comprehensive heat value of each product according to the initial heat value, statistical heat value, and predicted heat value of each product; and recommends products to users according to the comprehensive heat values of each product. By adopting this technical solution, the product heat values of all launched products can be accurately calculated, the conversion rate of products on the popular list can be improved, and thus it is convenient to recommend products to users.

[0089] Figure 2 It is a schematic flowchart of a product recommendation method provided by Embodiment 2 of this application. The following steps are included in Embodiment 2:

[0090] S201. Obtain at least one attribute information of each product and determine the type of each attribute information of each product.

[0091] Exemplarily, this step can refer to the above step S101 and will not be elaborated here.

[0092] S202. Determine the value of the heat value weight corresponding to the type according to the type of each attribute information of each product.

[0093] Exemplarily, the types of each attribute information of each product are different, and the numerical values of the heat value weights corresponding to different attribute information are also different. For example, the attribute information of Product A is the product cycle. The types of the product cycle include those calculated on a daily basis. For the type calculated on a daily basis, the numerical value of the corresponding heat value weight can be 1.0. If the type of the product cycle is calculated on a monthly basis, the numerical value of the corresponding heat value weight for this type can be 0.7.

[0094] S203. Determine the initial heat value of each product according to the numerical value of the heat value weight corresponding to the type.

[0095] Exemplarily, after determining the numerical value of the heat value weight corresponding to the type, use this numerical value as the initial heat value of the product. For example, if the numerical value of the heat value weight corresponding to one type of attribute information of the current product is 0.7, and the numerical value of the heat value weight corresponding to another type of attribute information is 0.6, then the initial heat value of the product is the result obtained by combining the two numerical values.

[0096] In an optional embodiment, determining the initial heat value of each product according to the numerical value of the heat value weight corresponding to the type includes:

[0097] Sum up the numerical values of the heat value weights corresponding to the type to obtain a first processing result; perform normalization processing on the first processing result to obtain the initial heat value of each product.

[0098] In this embodiment, if the numerical value of the heat value weight corresponding to one type of attribute information of the current product is 0.7, and the numerical value of the heat value weight corresponding to another type of attribute information is 0.6, then after adding the two for calculation, the first processing result is 1.3. Specifically, perform normalization processing on the first processing result 1.3, and the calculated value is 1.3 divided by 2, then the initial heat value of the product is 0.65, so the initial heat value of the current product is 0.65.

[0099] S204. Obtain the statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated based on user behavior within the preset time period.

[0100] Exemplarily, for this step, reference can be made to the above step S102 and will not be elaborated here.

[0101] S205. Obtain a preset model according to the statistical heat value of each product within the preset time period.

[0102] In this embodiment, the preset time period can be set by the user himself / herself. It can be a time period starting from the current moment and with a duration of 3 months as the preset time period. By obtaining the statistical heat value of each product within this time period and using the above statistical heat values as samples to train the preset model.

[0103] In an alternative embodiment, a preset model is obtained based on the statistical popularity values of each product within a preset time period, including:

[0104] Products with high statistical popularity values among each product within the preset time period are set as positive samples, and products with low statistical popularity values among each product within the preset time period are set as negative samples; at least one attribute information of each product within the preset time period is classified and numerical processing is performed on the classified values to obtain a second processing result; the second processing result is trained according to a preset algorithm to obtain a preset model.

[0105] Exemplarily, the numerical range of the statistical popularity value is between 0 and 1. The value of the product with a high statistical popularity value is closer to 1, and the product with a high statistical popularity value is used as a positive sample. The value of the product with a low statistical popularity value is closer to 0, and the product with a low statistical popularity value is used as a negative sample. In this embodiment, after obtaining the attribute information of each product, the value and type of the attribute information of the product are judged. After determining the value and type of the attribute information to which it belongs, the corresponding weight value is determined. If the current value is a continuous numerical feature, the continuous numerical feature is bucketized, and one-hot encoding is performed on the discrete value and the bucketized value. Among them, one-hot encoding is one-hot encoding, also known as one-hot encoding. The method is to use an N-bit status register to encode N states. Each state has an independent register bit, and at any time, only one of them is valid. After performing numerical processing on the classified values, a second processing result is obtained, and the second processing result is used to obtain a preset model according to the naive Bayes model algorithm and the iterative decision tree model algorithm.

[0106] In an alternative embodiment, the number of products with high statistical popularity values within the preset time period is the same as the number of products with low statistical popularity values within the preset time period.

[0107] In this embodiment, in order to ensure the accuracy of the preset model trained, when obtaining positive samples and negative samples, it is necessary to make the number of positive samples and the number of negative samples the same, so that the preset model will not be inaccurate due to improper sample selection.

[0108] S206. Input at least one attribute information of each product into the preset model to determine the predicted popularity value of each product.

[0109] Exemplarily, this step can refer to the above step S103 and will not be elaborated here.

[0110] S207. Determine the comprehensive popularity value of each product according to the initial popularity value, the statistical popularity value, and the predicted popularity value of each product.

[0111] Exemplarily, this step can refer to step S104 above and will not be elaborated here.

[0112] S208. Recommend products to users according to the comprehensive popularity values of each product.

[0113] Exemplarily, this step can refer to step S105 above and will not be elaborated here.

[0114] A product recommendation calculation method provided by the present application obtains at least one attribute information of each product, determines the value of the popularity value weight corresponding to the type according to the type of each attribute information of each product, determines the initial popularity value of each product according to the value of the popularity value weight corresponding to the type, and obtains the statistical popularity value of each product within a preset time period; wherein, the statistical popularity value is calculated according to the user behavior within the preset time period, obtains a preset model according to the statistical popularity value of each product within the preset time period, inputs at least one attribute information of each product into the preset model, and determines the predicted popularity value of each product; determines the comprehensive popularity value of each product according to the initial popularity value, the statistical popularity value and the predicted popularity value of each product; and recommends products to users according to the comprehensive popularity values of each product. By adopting this technical solution, the initial popularity value can be determined through the type of the attribute information of the product, and the comprehensive popularity value of the product is finally determined by combining the statistical popularity value and the predicted popularity value. Since the initial popularity value of the product can be set in advance, the product popularity value of newly launched products can be increased, the new products can be quickly converted, and then it is convenient to recommend products to users.

[0115] Figure 3 It is a schematic flowchart of a product recommendation method provided by Embodiment 3 of the present application. The following steps are included in Embodiment 3:

[0116] S301. Obtain at least one attribute information of each product, and determine the initial popularity value of each product according to the at least one attribute information of each product.

[0117] Exemplarily, this step can refer to step S101 above and will not be elaborated here.

[0118] S302. Assign different weight values according to the type of the obtained user behavior.

[0119] In this embodiment, the types of user behavior can be that the user purchases a product, clicks on a product, and shares a product, and the weights for different types of user behavior are different. Among them, the weight for the user to purchase a product can be 1.0, the weight for the user to click on a product can be 0.6, and the weight for the user to share a product can be 0.8.

[0120] S303. Determine the current statistical popularity value according to the number of user behaviors and the weight value.

[0121] In this embodiment, the current statistical popularity value can be obtained by multiplying the number of user behaviors by the weight value corresponding to this behavior. Specifically, it can be calculated through the following formula: H(T) = C · B, where H(T) is the current statistical popularity value, C is the number of user behaviors, and B is the weight value. For example, if the user behavior is to share a product, and the number of times the user shares the product is 100 times, and the weight value of the user sharing the product is 0.8, then the current statistical popularity value is 100 * 0.8.

[0122] S304. Determine the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the statistical popularity value in the same historical period.

[0123] In this embodiment, the current statistical popularity value can illustrate the current statistical popularity value of this product, while the statistical popularity value in the same historical period refers to the statistical popularity value before this product. Based on these two data, the statistical popularity value within the preset time period can be determined.

[0124] Optionally, determining the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the statistical popularity value in the same historical period includes:

[0125] Calculate the time decay factor of the current statistical popularity value and the time decay factor of the statistical popularity value in the same historical period; determine the statistical popularity value of each product within the preset time period according to the value of the current statistical popularity value, the time decay factor of the current statistical popularity value, the value of the statistical popularity value in the same historical period, and the time decay factor of the statistical popularity value in the same historical period.

[0126] In this embodiment, the time decay factor means that time will affect the value of the statistical popularity value. The longer the time, the smaller the value of the statistical popularity value. In this embodiment, the time decay factor of the current statistical popularity value and the time decay factor of the statistical popularity value in the same historical period can both be set by the user himself, and the values of the two can be set to be the same. For example, the time decay factors of both the current statistical popularity value and the statistical popularity value in the same historical period are d(T), the value of the current statistical popularity value is H(T), and the value of the statistical popularity value in the same historical period is H(T - 1), then the calculation of the statistical popularity value of this product within the preset time period is where H is the statistical popularity value of this product within the preset time period.

[0127] S305. Input at least one attribute information of each product into a preset model to determine the predicted popularity value of each product.

[0128] Exemplarily, this step can refer to the above step S103 and will not be elaborated here.

[0129] S306. Determine the comprehensive heat value of each product according to the initial heat value, the statistical heat value, and the predicted heat value of each product.

[0130] Exemplarily, for this step, reference may be made to step S104 above, and details will not be elaborated here.

[0131] S307. Recommend products to the user according to the comprehensive heat values of each product.

[0132] Exemplarily, for this step, reference may be made to step S105 above, and details will not be elaborated here.

[0133] A product recommendation calculation method provided by the present application can assign different weight values according to the types of user behaviors obtained, determine the current statistical heat value according to the number of user behaviors and the weight values, determine the statistical heat value of each product within a preset time period according to the current statistical heat value and the statistical heat value in the same historical period. By adopting this technical solution, the accuracy of the statistical heat value can be improved, and further the accuracy of the comprehensive heat value can be improved.

[0134] Figure 4 FIG. 40 is a schematic diagram of a product recommendation device provided in Embodiment 4 of the present application. The device 40 in Embodiment 4 includes the following steps:

[0135] An attribute information acquisition module 401, configured to acquire at least one attribute information of each product, and determine the initial heat value of each product according to the at least one attribute information of each product;

[0136] A statistical heat value acquisition module 402, configured to acquire the statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated according to user behaviors within the preset time period;

[0137] A predicted heat value determination module 403, configured to input at least one attribute information of each product into a preset model to determine the predicted heat value of each product;

[0138] A comprehensive heat value determination module 404, configured to determine the comprehensive heat value of each product according to the initial heat value, the statistical heat value, and the predicted heat value of each product;

[0139] A recommendation module, configured to recommend products to the user according to the comprehensive heat values of each product.

[0140] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the above-described device can refer to the corresponding process in the foregoing method embodiment, and details will not be elaborated here.

[0141] Figure 5It is a schematic diagram of a product recommendation device provided in Embodiment 5 of the present application. The device 50 in Embodiment 5 includes the following steps:

[0142] An attribute information acquisition module 501, configured to acquire at least one attribute information of each product, and determine an initial heat value of each product according to the at least one attribute information of each product.

[0143] A statistical heat value acquisition module 502, configured to acquire a statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated according to user behaviors within the preset time period.

[0144] A predicted heat value determination module 503, configured to input at least one attribute information of each product into a preset model, and determine a predicted heat value of each product.

[0145] A comprehensive heat value determination module 504, configured to determine a comprehensive heat value of each product according to the initial heat value of each product, the statistical heat value of each product, and the predicted heat value of each product;

[0146] A recommendation module, configured to recommend products to a user according to the comprehensive heat values of each product.

[0147] Optionally, the attribute information acquisition module 501 includes:

[0148] A type determination unit 5011, configured to acquire at least one attribute information of each product, and determine the type of each attribute information of each product.

[0149] A heat value weight numerical value determination unit 5012, configured to determine the numerical value of the heat value weight corresponding to the type according to the type of each attribute information of each product.

[0150] An initial heat value determination unit 5013, configured to determine the initial heat value of each product according to the numerical value of the heat value weight corresponding to the type.

[0151] Optionally, the initial heat value determination unit 5013 includes:

[0152] A first processing result determination subunit 50131, configured to sum up the numerical values of the heat value weights corresponding to the type, and obtain a first processing result.

[0153] An initial heat value determination subunit 50132, configured to perform normalization processing on the first processing result, and obtain the initial heat value of each product.

[0154] Optionally, the device further includes:

[0155] A preset model determination module 505, configured to obtain a preset model according to the statistical heat value of each product within a preset time period.

[0156] The preset model determination module 505 includes:

[0157] A setting unit 5051, configured to set the products with high statistical popularity values among each product within a preset time period as positive samples, and set the products with low statistical popularity values among each product within a preset time period as negative samples;

[0158] A second processing result unit 5052, configured to classify at least one attribute information of each product within a preset time period and perform numerical processing on the classified numerical values to obtain a second processing result.

[0159] A preset model determination unit 5053, configured to train the second processing result according to a preset algorithm to obtain a preset model.

[0160] Optionally, the number of products with high statistical popularity values within a preset time period is the same as the number of products with low statistical popularity values within a preset time period.

[0161] The statistical popularity value acquisition module 502 includes:

[0162] A weight value determination unit 5021, configured to assign different weight values according to the types of user behaviors obtained.

[0163] A current statistical popularity value determination unit 5022, configured to determine the current statistical popularity value according to the number of user behaviors and the weight values.

[0164] A statistical popularity value determination unit 5023, configured to determine the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the historical statistical popularity value in the same period.

[0165] Optionally, the statistical popularity value determination unit 5023 includes:

[0166] A calculation subunit 50231, configured to calculate the time decay factor of the current statistical popularity value and the time decay factor of the historical statistical popularity value in the same period.

[0167] A statistical popularity value determination subunit 50232, configured to determine the statistical popularity value of each product within a preset time period according to the value of the current statistical popularity value, the time decay factor of the current statistical popularity value, the value of the historical statistical popularity value in the same period, and the time decay factor of the historical statistical popularity value in the same period.

[0168] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0169] Figure 6FIG. 0 is a block diagram of a terminal device shown according to an exemplary embodiment. The device may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0170] Device 600 may include one or more of the following components: a processing component 602, a memory 604, a power component 606, a multimedia component 608, an audio component 610, an input / output (I / O) interface 612, a sensor component 614, and a communication component 616.

[0171] The processing component 602 generally controls the overall operation of the device 600, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 602 may include one or more processors 620 to execute instructions to complete all or part of the steps of the above-described methods. In addition, the processing component 602 may include one or more modules to facilitate the interaction between the processing component 602 and other components. For example, the processing component 602 may include a multimedia module to facilitate the interaction between the multimedia component 608 and the processing component 602.

[0172] The memory 604 is configured to store various types of data to support the operation of the device 600. Examples of such data include instructions for any application or method operating on the device 600, contact data, phone book data, messages, pictures, videos, etc. The memory 604 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a disk, or an optical disk.

[0173] The power component 606 provides power to the various components of the device 600. The power component 606 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 600.

[0174] The multimedia component 608 includes a screen that provides an output interface between the device 600 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of the touch or swipe actions, but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 608 includes a front camera and / or a rear camera. When the device 600 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0175] The audio component 610 is configured to output and / or input audio signals. For example, the audio component 610 includes a microphone (MIC) that is configured to receive external audio signals when the device 600 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 604 or transmitted via the communication component 616. In some embodiments, the audio component 610 further includes a speaker for outputting audio signals.

[0176] The I / O interface 612 provides an interface between the processing component 602 and a peripheral interface module, which can be a keyboard, a click wheel, buttons, etc. These buttons can include, but are not limited to: a home button, a volume button, a power button, and a lock button.

[0177] The sensor component 614 includes one or more sensors for providing an assessment of various aspects of the state of the device 600. For example, the sensor component 614 can detect the on / off state of the device 600, the relative positioning of components, such as the display and keypad of the device 600. The sensor component 614 can also detect a change in the position of the device 600 or a component of the device 600, the presence or absence of user contact with the device 600, the orientation or acceleration / deceleration of the device 600, and the temperature change of the device 600. The sensor component 614 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 614 can also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 614 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0178] The communication component 616 is configured to facilitate communication between the device 600 and other devices in a wired or wireless manner. The device 600 can access a communication standard-based wireless network, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 616 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 616 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0179] In an exemplary embodiment, the device 600 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above method.

[0180] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 604 including instructions, and the above instructions can be executed by a processor 620 of the device 600 to complete the above method. For example, the non-transitory computer-readable storage medium can be a ROM, Random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0181] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a terminal device, enables the terminal device to execute the product recommendation method of the terminal device described above.

[0182] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include known common knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and embodiments are only to be regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0183] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A product recommendation method, characterized in that, the method includes: obtaining at least one attribute information of each product, determining an initial popularity value of each product according to the at least one attribute information of each product, where the attribute information is information used to characterize the current characteristics of the product, and the initial popularity value of each attribute information of each product is preset by the user; allocating different weight values according to the types of user behaviors obtained, where the types of user behaviors include user purchasing products, user clicking on products, and user sharing products; determining a current statistical popularity value according to the number of times of the user behavior and the weight value; determining the statistical popularity value of each product within a preset time period according to the current statistical popularity value and the historical same-period statistical popularity value; obtaining a preset model according to the statistical popularity value of each product within a preset time period; inputting the at least one attribute information of each product into the preset model to determine the predicted popularity value of each product; determining the comprehensive popularity value of each product according to the initial popularity value of each product, the statistical popularity value of each product, and the predicted popularity value of each product; recommending products to the user according to the comprehensive popularity values of the products.

2. The method according to claim 1, characterized in that, obtaining at least one attribute information of each product and determining the initial popularity value of each product according to the at least one attribute information of each product includes: obtaining at least one attribute information of each product and determining the type of each attribute information of each product; determining the value of the popularity value weight corresponding to the type according to the type of each attribute information of each product; determining the initial popularity value of each product according to the value of the popularity value weight corresponding to the type.

3. The method according to claim 2, characterized in that, determining the initial popularity value of each product according to the value of the popularity value weight corresponding to the type includes: adding up the values of the popularity value weights corresponding to the type to obtain a first processing result; performing a normalization process on the first processing result to obtain the initial popularity value of each product.

4. The method according to claim 1, characterized in that, obtaining the preset model according to the statistical popularity value of each product within a preset time period includes: setting the products with high statistical popularity values within a preset time period as positive samples, and setting the products with low statistical popularity values within a preset time period as negative samples; classifying the at least one attribute information of each product within a preset time period and performing numerical processing on the classified values to obtain a second processing result; training the second processing result according to a preset algorithm to obtain the preset model.

5. The method according to claim 4, characterized in that, the number of products with high statistical popularity values within a preset time period is the same as the number of products with low statistical popularity values within a preset time period.

6. The method according to claim 1, characterized in that, Determine the statistical heat value of each product within a preset time period according to the current statistical heat value and the historical statistical heat value in the same period, including: Calculate the time decay factor of the current statistical heat value and the time decay factor of the historical statistical heat value in the same period; Determine the statistical heat value of each product within a preset time period according to the value of the current statistical heat value, the time decay factor of the current statistical heat value, the value of the historical statistical heat value in the same period, and the time decay factor of the historical statistical heat value in the same period.

7. A product recommendation device, characterized in that, the device includes: An attribute information acquisition module, configured to acquire at least one attribute information of each product, and determine the initial heat value of each product according to the at least one attribute information of each product, where the attribute information is information used to characterize the current characteristics of the product, and the initial heat value of each attribute information of each product is preset by the user; A statistical heat value acquisition module, configured to acquire the statistical heat value of each product within a preset time period; wherein, the statistical heat value is calculated according to user behavior within a preset time period; A predicted heat value determination module, configured to input at least one attribute information of each product into a preset model to determine the predicted heat value of each product; A comprehensive heat value determination module, configured to determine the comprehensive heat value of each product according to the initial heat value of each product, the statistical heat value of each product, and the predicted heat value of each product; A recommendation module, configured to recommend products to a user according to the comprehensive heat values of the products; The statistical heat value acquisition module includes: A weight value determination unit, configured to allocate different weight values according to the types of user behavior obtained, where the types of user behavior include user purchasing products, user clicking on products, and user sharing products; A current statistical heat value determination unit, configured to determine the current statistical heat value according to the number of times of the user behavior and the weight value; A statistical heat value determination unit, configured to determine the statistical heat value of each product within a preset time period according to the current statistical heat value and the historical statistical heat value in the same period; The device further includes: a preset model determination module, The preset model determination module is configured to obtain a preset model according to the statistical heat value of each product within a preset time period.

8. An electronic device, characterized in that, includes: A processor, and a memory communicatively connected to the processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the method according to any one of claims 1-6.

10. A computer program product, characterized in that, includes a computer program, and when the computer program is executed by a processor, it implements the method according to any one of claims 1-6.

Images