CN117726403A

CN117726403A - Product recommendation method and device, storage medium and electronic equipment

Info

Publication number: CN117726403A
Application number: CN202311737427.5A
Authority: CN
Inventors: 徐洪水; 张馨月; 何毅恒; 李文渊
Original assignee: Tianyi Electronic Commerce Co Ltd
Current assignee: Tianyi Electronic Commerce Co Ltd
Priority date: 2023-12-15
Filing date: 2023-12-15
Publication date: 2024-03-19

Abstract

The application discloses a product recommendation method and device, a storage medium and electronic equipment, and relates to the technical field of computers. The method comprises the following steps: obtaining N target products, wherein the N target products are products to be recommended to a target object; acquiring a data information set based on N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of a target object; determining S prediction models, wherein the S prediction models at least comprise: a first prediction model, a second prediction model, and a third prediction model; and determining a target recommendation sequence for recommending N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence. By the method and the device, the problem of low accuracy of recommending the product to the user in the related technology is solved.

Description

Product recommendation method and device, storage medium and electronic equipment

Technical Field

The present application relates to the field of computer technologies, and in particular, to a product recommendation method and apparatus, a storage medium, and an electronic device.

Background

In the related art, a machine learning model is generally used to score the commodity, and the commodity is recommended according to the score. However, with the improvement of commodity richness and the improvement of user behavior richness, the traditional recommendation method (such as GBDT and DNN) based on a single model is difficult to accurately describe and capture interest preference and interest change of the user, and therefore the accuracy of recommending products to the user is low.

Aiming at the problem of low accuracy of recommending products to users in the related technology, no effective solution is proposed at present.

Disclosure of Invention

The main objective of the present application is to provide a product recommendation method and apparatus, a storage medium, and an electronic device, so as to solve the problem of low accuracy in recommending products to users in related technologies.

In order to achieve the above object, according to one aspect of the present application, there is provided a recommendation method of a product. The method comprises the following steps: obtaining N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; acquiring a data information set based on the N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1; determining S prediction models, wherein the prediction models are used for predicting and recommending the recommendation sequence of the N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; and determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

Further, the T historical behavior data at least includes: the method comprises the steps of determining a target recommendation sequence of recommending N target products according to a data information set and S prediction models, wherein the data information set comprises first historical behavior data, second historical behavior data and third historical behavior data, the first historical behavior data is the historical behavior data of the target objects on the target products, the second historical behavior data is the historical behavior data of each of the M objects on the target products, the third historical behavior data is the historical behavior data of each of the M objects on the target products, and the target recommendation sequence of recommending the N target products comprises: inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a first recommendation sequence; inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a second recommendation sequence; inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a third recommendation sequence; and determining the target recommendation order of recommending the N target products based on the second recommendation order and the third recommendation order.

Further, determining the target recommendation order to recommend the N target products based on the second recommendation order and the third recommendation order includes: determining a score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; determining a score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and determining the target recommendation sequence for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

Further, the S prediction models are obtained by: obtaining a target training sample set, wherein the target training sample set at least comprises Q sample behavior data, attribute information of W sample products and attribute information of P sample objects obtained in a history process, the Q sample behavior data at least comprises behavior data of the P sample objects on the W sample products, and Q, W and P are positive integers larger than 1; learning and training each original prediction model in S original prediction models by using the target training sample set to obtain the S prediction models, wherein the S original prediction models at least comprise: the first, second and third original prediction models.

Further, learning and training each original prediction model in the S original prediction models by using the target training sample set, and obtaining the S prediction models includes: classifying the Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of the sample objects on the sample products, the second sample behavior data is behavior data of each sample object of the P sample objects on the sample products, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product; learning and training the first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain the first prediction model; obtaining a prediction result output by the first prediction model, and learning and training the second original prediction model in the S original prediction models by using the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the second prediction model; learning and training the third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model; and obtaining the S prediction models based on the first prediction model, the second prediction model and the third prediction model.

Further, obtaining the prediction result output by the first prediction model includes: acquiring the second sample behavior data and the third sample behavior data; inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict and recommend the recommendation sequence of the W sample products, and outputting the predicted recommendation sequence of the W sample products; and taking the predicted recommendation sequence of recommending the W sample products as the prediction result.

Further, obtaining the target training sample set includes: acquiring the portrait information of each sample object in the P sample objects, and determining the attribute information of the P sample objects based on the portrait information of each sample object, wherein the attribute information of the P sample objects at least comprises: ID information of each sample object and sex information of each sample object; acquiring attribute information of the W sample products, wherein the attribute information of the W sample products at least comprises: ID information of each sample product and category information of each sample product; acquiring behavior data of the P sample objects on the W sample products; determining the Q sample behavior data according to the behavior data of the P sample objects on the W sample products; and acquiring the target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

In order to achieve the above object, according to another aspect of the present application, there is provided a recommendation device for a product. The device comprises: the first acquisition unit is used for acquiring N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; the second obtaining unit is configured to obtain a data information set based on the N target products, where the data information set at least includes: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1; the first determining unit is configured to determine S prediction models, where the prediction models are used to predict and recommend a recommendation order of the N target products, and the S prediction models at least include: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; the first processing unit is used for determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

Further, the T historical behavior data at least includes: first historical behavior data, second historical behavior data and third historical behavior data, wherein the first historical behavior data is the historical behavior data of the target object on the target product, the second historical behavior data is the historical behavior data of each object of the M objects on the target product, the third historical behavior data is the historical behavior data of each object of the M objects on each target product, and the first processing unit comprises: the first processing module is used for inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a first recommendation sequence; the second processing module is used for inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a second recommendation sequence; the third processing module is used for inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a third recommendation sequence; the first determining module is configured to determine the target recommendation order of recommending the N target products based on the second recommendation order and the third recommendation order.

Further, the first determining module includes: the first determining submodule is used for determining the score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; the second determining submodule is used for determining the score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; the third determination submodule is used for respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and a fourth determining sub-module, configured to determine the target recommendation order for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model, and the weights corresponding to the third prediction model.

Further, the S prediction models are obtained by: a third obtaining unit, configured to obtain a target training sample set, where the target training sample set includes at least Q sample behavior data obtained in a history process, attribute information of W sample products, and attribute information of P sample objects, the Q sample behavior data includes at least behavior data of the P sample objects on the W sample products, and Q, W and P are both positive integers greater than 1; the first training unit is configured to perform learning training on each of S original prediction models by using the target training sample set, so as to obtain S prediction models, where the S original prediction models at least include: the first, second and third original prediction models.

Further, the first training unit includes: a fourth processing module, configured to perform classification processing on the Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data, and third sample behavior data, where the first sample behavior data is behavior data of the sample object on the sample product, the second sample behavior data is behavior data of each sample object of the P sample objects on the sample product, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product; the first training module is used for learning and training the first original prediction model in the S original prediction models by utilizing the second sample behavior data and the third sample behavior data to obtain the first prediction model; the second training module is used for acquiring a prediction result output by the first prediction model, and learning and training the second original prediction model in the S original prediction models by utilizing the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the second prediction model; the third training module is used for learning and training the third initial prediction model in the S initial prediction models by utilizing the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the third prediction model; and the second determining module is used for obtaining the S prediction models based on the first prediction model, the second prediction model and the third prediction model.

Further, the second training module includes: a first obtaining sub-module, configured to obtain the second sample behavior data and the third sample behavior data; the first output sub-module is used for inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict and recommend the recommendation sequence of the W sample products, and outputting the predicted recommendation sequence of the W sample products; and a fifth determining sub-module, configured to take the predicted recommended order of recommending the W sample products as the prediction result.

Further, the third acquisition unit includes: a fifth processing module, configured to obtain image information of each sample object in the P sample objects, and determine attribute information of the P sample objects based on the image information of each sample object, where the attribute information of the P sample objects at least includes: ID information of each sample object and sex information of each sample object; the first obtaining module is configured to obtain attribute information of the W sample products, where the attribute information of the W sample products includes at least: ID information of each sample product and category information of each sample product; the second acquisition module is used for acquiring behavior data of the P sample objects on the W sample products; a third determining module, configured to determine the Q sample behavior data according to the behavior data of the P sample objects on the W sample products; and a third obtaining module, configured to obtain the target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products, and the Q sample behavior data.

In order to achieve the above object, according to another aspect of the present application, there is provided a computer-readable storage medium storing a program, wherein the program performs the recommendation method of the product of any one of the above.

To achieve the above object, according to another aspect of the present application, there is provided an electronic device including one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the recommended method of the product of any of the above.

Through the application, the following steps are adopted: obtaining N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; acquiring a data information set based on N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of target objects, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on N target products, the M pieces of historical behavior data at least comprise target objects, and T and M are positive integers larger than 1; s prediction models are determined, wherein the prediction models are used for predicting the recommendation sequence of recommending N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; according to the data information set and the S prediction models, determining the target recommendation sequence of recommending N target products, recommending N target products to the target object according to the target recommendation sequence, and solving the problem of low accuracy of recommending products to users in the related technology. The method comprises the steps of obtaining a plurality of target products to be recommended to a user, obtaining a data information set based on the plurality of target products, determining a target recommendation sequence for recommending the plurality of target products according to the data information set and a plurality of prediction models, and recommending the plurality of target products to the user according to the target recommendation sequence, so that the problem of low accuracy of recommending the products to the user in the related art is avoided, and the effect of improving the accuracy of recommending the products to the user is achieved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application, illustrate and explain the application and are not to be construed as limiting the application. In the drawings:

FIG. 1 is a flowchart I of a method of recommending a product according to an embodiment of the present application;

FIG. 2 is a second flowchart of a method of recommending a product according to an embodiment of the present application;

FIG. 3 is a flowchart III of a method of recommending a product provided in accordance with an embodiment of the present application;

FIG. 4 is a schematic diagram of a recommendation device for products provided in accordance with an embodiment of the present application;

fig. 5 is a schematic diagram of an electronic device provided according to an embodiment of the present application.

Detailed Description

It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be noted that, related information (including, but not limited to, user equipment information, user personal information, etc.) and data (including, but not limited to, data for presentation, analyzed data, etc.) related to the present disclosure are information and data authorized by a user or sufficiently authorized by each party. For example, an interface is provided between the system and the relevant user or institution, before acquiring the relevant information, the system needs to send an acquisition request to the user or institution through the interface, and acquire the relevant information after receiving the consent information fed back by the user or institution.

For convenience of description, the following will describe some terms or terms related to the embodiments of the present application:

DNN: DNN is an abbreviation for deep neural network (Deep Neural Network), which is an artificial neural network composed of multiple layers of neurons.

The invention will be described with reference to preferred implementation steps, and FIG. 1 is a flowchart of a method for recommending a product according to an embodiment of the present application, as shown in FIG. 1, the method includes the steps of:

step S101, N target products are obtained, wherein the N target products are the products to be recommended to the target object, and N is a positive integer greater than 1.

For example, the target object may be a user, and the N target products may be a plurality of commodities to be recommended to the user.

Step S102, acquiring a data information set based on N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of target objects, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on N target products, the M pieces of historical behavior data at least comprise target objects, and T and M are positive integers larger than 1.

For example, the above-mentioned T pieces of history behavior data may include history behavior data of the user (the above-mentioned target object), including merchandise exposure, clicking, purchasing data, etc., and may also include clicking, purchasing data, transaction frequency data, etc. of the same merchandise by different users in the history process and clicking, purchasing data, transaction frequency data, etc. of different merchandise by different users in the history process. The attribute information of each target product may include a commodity ID (Identification) mark, category information, and the like; the attribute information of the target object may include a user ID (Identification) Identification, sex information, and the like.

Step S103, S prediction models are determined, wherein the prediction models are used for predicting the recommendation sequence of recommending N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorization machine model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorization machine model, and S is a positive integer greater than 1.

For example, the first predictive model described above may be a gradient boost decision tree (GBDT, gradient Boosting Decision Trees) predictive model; the second predictive model may be a factorizer (FM, factorization Machine) predictive model; the third prediction model may be a model of a combination of a multi-layer perceptron and a factorizer.

Step S104, determining a target recommendation sequence of recommending N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

For example, a Gradient Boost Decision Tree (GBDT) prediction model (the first prediction model described above), a Factorizer (FM) prediction model (the second prediction model described above), and a model of a combination of a multi-layer perceptron and a factorizer (the third prediction model described above) may be fused to obtain a fusion model. The acquired historical behavior data, user ID identification and sex information, and commodity ID identification and category information may then be input into a fusion model, and the recommendation order of the plurality of commodities to be recommended to the user (the above-described target object) may be predicted by the fusion model, and then the plurality of commodities to be recommended may be recommended to the user according to the recommendation order.

Through the steps S101 to S104, the plurality of target products to be recommended to the user are obtained, the data information set is obtained based on the plurality of target products, then the target recommendation sequence for recommending the plurality of target products is determined according to the data information set and the plurality of prediction models, and the plurality of target products are recommended to the user according to the target recommendation sequence, so that the problem of low accuracy of recommending the products to the user in the related art is avoided, and the effect of improving the accuracy of recommending the products to the user is achieved.

Optionally, in the product recommendation method provided in the embodiment of the present application, obtaining the target training sample set includes: acquiring portrait information of each sample object in the P sample objects, and determining attribute information of the P sample objects based on the portrait information of each sample object, wherein the attribute information of the P sample objects at least comprises: ID information of each sample object and sex information of each sample object; acquiring attribute information of W sample products, wherein the attribute information of the W sample products at least comprises: ID information of each sample product and category information of each sample product; acquiring behavior data of P sample objects on W sample products; according to the behavior data of the P sample objects on the W sample products, determining Q sample behavior data; and acquiring a target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

For example, user portrait information of the user on the platform may be acquired first, and then ID identification information, sex information, and the like of the user may be obtained based on the user portrait information of the user. Then the ID identification, category information and the like of the commodity can be obtained; and acquiring behavior data of a plurality of sample objects on a plurality of sample products, wherein the behavior data can be data such as clicking, purchasing, transaction frequency and the like of the sample objects on the sample products, and then summarizing the acquired ID identification information and sex information of the user, ID identification and category information of the commodity, and the data such as clicking, purchasing, transaction frequency and the like of the sample objects on the sample products as a training sample set for training a prediction model.

Through the scheme, the training sample set for training the prediction model can be conveniently obtained.

Optionally, in the product recommendation method provided in the embodiment of the present application, S prediction models are obtained by: obtaining a target training sample set, wherein the target training sample set at least comprises Q sample behavior data, attribute information of W sample products and attribute information of P sample objects obtained in a history process, the Q sample behavior data at least comprises behavior data of P sample objects on the W sample products, and Q, W and P are positive integers larger than 1; learning and training each original prediction model in the S original prediction models by using the target training sample set to obtain S prediction models, wherein the S original prediction models at least comprise: the first, second and third original prediction models.

For example, the ID identification information and the sex information of the user, the ID identification and the category information of the commodity, and the click, purchase, transaction frequency, and other data of the sample object on the sample product in the training sample set may be used to learn and train the plurality of original prediction models, and a trained Gradient Boost Decision Tree (GBDT) prediction model, a factor decomposition machine (FM) prediction model, and a model of a combination of a multi-layer perceptron and a factor decomposition machine may be obtained, and then the trained Gradient Boost Decision Tree (GBDT) prediction model, the factor decomposition machine (FM) prediction model, and the model of a combination of a multi-layer perceptron and a factor decomposition machine may be used as the S prediction models.

Through the scheme, the training sample set can be used for conveniently learning and training the original prediction model, and a trained model can be obtained quickly and accurately.

Optionally, in the product recommendation method provided in the embodiment of the present application, obtaining a prediction result output by the first prediction model includes: acquiring second sample behavior data and third sample behavior data; inputting the second sample behavior data and the third sample behavior data into a first prediction model to predict the recommendation sequence of the recommended W sample products, and outputting the predicted recommendation sequence of the recommended W sample products; taking the predicted recommendation sequence of the recommended W sample products as a prediction result.

For example, the second sample behavior data may be click, purchase, transaction frequency data, etc. of different sample objects on the same sample product; the third sample behavior data may be click, purchase, transaction frequency data, etc. of different sample objects on different sample products; the first predictive model described above may be a Gradient Boost Decision Tree (GBDT) predictive model. For example, click, purchase adding, transaction frequency data and the like of different sample objects on the same sample product may be input into a Gradient Boost Decision Tree (GBDT) prediction model, and then a predicted recommended sequence for recommending a plurality of sample products may be output, and the output predicted recommended sequence for recommending a plurality of sample products may be used as the prediction result.

In summary, by inputting the click and purchase data of the commodity by the user into the gradient-lifting decision tree prediction model, the recommendation sequence of the recommended commodity can be rapidly and accurately predicted.

Optionally, in the product recommendation method provided in the embodiment of the present application, learning and training each of the S original prediction models by using the target training sample set includes: classifying Q sample behavior data in a target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of sample objects on sample products, the second sample behavior data is behavior data of each sample object in P sample objects on sample products, and the third sample behavior data is behavior data of each sample object in P sample objects on each sample product; learning and training a first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain a first prediction model; obtaining a prediction result output by the first prediction model, and learning and training a second original prediction model in the S original prediction models by using the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a second prediction model; learning and training a third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model; and obtaining S prediction models based on the first prediction model, the second prediction model and the third prediction model.

For example, the first sample behavior data may be exposure, clicking, purchasing data, etc. of the sample object to the same sample product in the history process; the second sample behavior data may be click, purchase adding, transaction frequency data, etc. of different sample objects on the same sample product in the history process; the third sample behavior data may be click, purchase, transaction frequency data, etc. of different sample objects on different sample products in the history process; the first raw predictive model may be an untrained Gradient Boost Decision Tree (GBDT) predictive model; the second original predictive model may be an untrained Factorizer (FM) predictive model; the third initial prediction model may be a model of a combination of an untrained multi-layer perceptron and a factorizer; the first prediction model may be a trained Gradient Boost Decision Tree (GBDT) prediction model; the second prediction model may be a trained Factorization (FM) prediction model; the third prediction model may be a model of a combination of a trained multi-layer perceptron and a factorizer.

For example, click, purchase, transaction frequency data and the like of the same sample product in a history process can be firstly used for different sample objects, and untrained Gradient Boost Decision Tree (GBDT) prediction models can be learned and trained by using click, purchase, transaction frequency data and the like of different sample products in a history process to obtain trained Gradient Boost Decision Tree (GBDT) prediction models; acquiring a recommendation sequence of a plurality of predicted sample products output by a trained Gradient Boost Decision Tree (GBDT) prediction model, and learning and training a training-free factor decomposition machine (FM) prediction model by using the recommendation sequence of the plurality of predicted sample products output by the trained Gradient Boost Decision Tree (GBDT) prediction model, clicking, purchasing, transaction frequency data and the like of different sample objects on the same sample product in a historical process, clicking, purchasing, transaction frequency data and the like of different sample products, ID identification, category information and the like of the sample products, user ID identification, sex information and the like of different sample objects in the historical process to obtain a trained factor decomposition machine (FM) prediction model; and then, learning and training the model of the untrained combination of the multi-layer perceptron and the factoring machine by utilizing the recommendation sequence of the plurality of predicted sample products outputted by the trained Gradient Boost Decision Tree (GBDT) prediction model, exposure, clicking, purchasing data and the like of the same sample product by a sample object in a history process, clicking, purchasing, transaction frequency data and the like of different sample products by different sample objects in a history process, ID identification, category information and the like of the sample products, user ID identification, sex information and the like, and obtaining the model of the trained combination of the multi-layer perceptron and the factoring machine. And then the trained gradient lifting decision tree (GBDT) prediction model, the trained Factorization Machine (FM) prediction model and the model combined by the trained multi-layer perceptron and the factorization machine can be used as the S prediction models.

By the scheme, a trained gradient lifting decision tree (GBDT) prediction model, a trained Factorization Machine (FM) prediction model and a trained model combining a multi-layer perceptron and the factorization machine can be quickly and accurately obtained.

Fig. 2 is a second flowchart of a product recommendation method according to an embodiment of the present application, as shown in fig. 2, in the product recommendation method according to an embodiment of the present application, T pieces of historical behavior data at least include: the method comprises the steps of determining a target recommendation sequence of recommending N target products according to a data information set and S prediction models, wherein the first historical behavior data, the second historical behavior data and the third historical behavior data are the historical behavior data of the target object on the target product, the second historical behavior data are the historical behavior data of each object of M objects on the target product, and the third historical behavior data are the historical behavior data of each object of M objects on the target product, and the determining the target recommendation sequence of recommending the N target products comprises the following steps:

step S201, inputting second historical behavior data and third historical behavior data into a first prediction model to predict and recommend the recommendation sequence of N target products, and outputting a first recommendation sequence;

Step S202, inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into a second prediction model to predict the recommendation sequence of recommending N target products, and outputting a second recommendation sequence;

step S203, inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into a third prediction model to predict the recommendation sequence of recommending N target products, and outputting a third recommendation sequence;

step S204, determining a target recommendation sequence for recommending N target products based on the second recommendation sequence and the third recommendation sequence.

For example, the first historical behavior data may be historical behavior data of a user (the target object described above), including merchandise exposure, clicking, purchasing data, and the like; the second historical behavior data can be click, purchase adding, transaction frequency data and the like of the same commodity in the historical process of different users; the third historical behavior data can be click, purchase adding, transaction frequency data and the like of different commodities in a historical process of different users; the attribute information of each target product may include ID identification, category information, and the like of each commodity; the attribute information of the target object may include a user ID identification, sex information, and the like. The first prediction model may be a trained Gradient Boost Decision Tree (GBDT) prediction model; the second prediction model may be a trained Factorization (FM) prediction model; the third prediction model may be a model of a combination of a trained multi-layer perceptron and a factorizer.

For example, click, purchase adding, transaction frequency data and the like of the same commodity by different users in a history process, and click, purchase adding, transaction frequency data and the like of different commodities by different users in a history process can be input into a trained Gradient Boost Decision Tree (GBDT) prediction model, and a prediction output result (the first recommendation sequence) of the Gradient Boost Decision Tree (GBDT) prediction model is output; then, the predicted output result of the Gradient Boosting Decision Tree (GBDT) prediction model, the click, additional purchase, transaction frequency data and the like of different users on the same commodity in the history process, the click, additional purchase, transaction frequency data and the like of different users on different commodities in the history process, the ID identification, category information and the like of each commodity, the ID identification, sex information and the like of the user and the like can be input into a trained Factoring Machine (FM) prediction model, and the predicted output result (the second recommended sequence) of the Factoring Machine (FM) prediction model is output; then inputting the predicted output result of the Gradient Boost Decision Tree (GBDT) prediction model, historical behavior data of a user (the target object) and the like, including commodity exposure, clicking, purchasing data and the like, clicking, purchasing, transaction frequency data and the like of the same commodity by different users in the historical process, clicking, purchasing, transaction frequency data and the like of different commodities by different users in the historical process, ID identification, category information and the like of each commodity, ID identification, sex information and the like of the user and the like into a trained model of a combination of a multi-layer perceptron and a factor decomposer, and outputting the predicted output result of the model of the combination of the multi-layer perceptron and the factor decomposer (the third recommendation sequence); the recommendation order in which a plurality of items are finally recommended to the user may then be determined based on the prediction output result of the Factoring Machine (FM) prediction model (the second recommendation order described above) and the prediction output result of the model of the combination of the multi-layer perceptron and the factoring machine (the third recommendation order described above).

Through the scheme, the recommendation sequence of finally recommending a plurality of commodities to the user can be rapidly and accurately determined by using the fusion model.

Optionally, in the product recommendation method provided in the embodiment of the present application, determining, based on the second recommendation order and the third recommendation order, a target recommendation order for recommending N target products includes: determining a score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; determining a score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and determining a target recommendation sequence for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

For example, the product may be recommended according to the score of each product. For example, the score value corresponding to each commodity may be determined according to the prediction output result (the second recommendation order described above) of the Factoring Machine (FM) prediction model; and determining the corresponding score value of each commodity according to the prediction output result (the third recommendation sequence) of the model combined by the multi-layer perceptron and the factoring machine; and then determining weights corresponding to the factor decomposition machine (FM) prediction model, the multi-layer perceptron and the factor decomposition machine combined model respectively, and finally recommending a plurality of commodities to a user according to the score value corresponding to each commodity determined by the prediction output result of the factor decomposition machine (FM) prediction model, the score value corresponding to each commodity determined by the prediction output result of the multi-layer perceptron and the factor decomposition machine combined model, and the weights corresponding to the factor decomposition machine (FM) prediction model, the multi-layer perceptron and the factor decomposition machine combined model respectively.

By the scheme, the recommendation sequence of finally recommending a plurality of commodities to the user can be rapidly and accurately determined according to the score value of each commodity.

For example, the embodiment provides a commodity recommendation method and system based on a fusion model, and the embodiment belongs to the technical field of computers, and particularly relates to the field of commodity recommendation systems. Moreover, in the embodiment, a commodity recommendation method based on a fusion model is provided, so that recommendation accuracy can be improved.

For example, the technical problem that the present embodiment mainly solves is to overcome the problem that the traditional single model is difficult to accurately recommend in the commodity recommendation system, and provide a recommendation method based on multi-model fusion, so as to improve the accuracy of recommendation, and improve the shopping experience of users and the sales of platform commodities.

For example, fig. 3 is a flowchart III of a product recommendation method provided according to an embodiment of the present application, and as shown in fig. 3, the commodity recommendation method based on multi-model fusion provided in this embodiment provides the following technical solutions:

s1, acquiring user figures of a user on a platform and historical behavior data of the user, wherein the user figures comprise commodity exposure, clicking, purchasing and purchasing data;

S2, determining a training sample set and a test sample set according to the data acquired in the S1, wherein the data occurrence time node of the training sample set is earlier than that of the test sample set; the sample set needs to contain unique ID identification of the user on the platform so as to distinguish different users;

s3, constructing model training features according to the sample set determined in the S2, wherein the features comprise user behavior sequence features and commodity features, the feature types can be divided into finer granularity according to the types of feature values, and the feature types can be specifically called posterior statistical features such as commodity clicking, purchasing, transaction frequency and the like; user ID identification, gender, merchandise ID identification, category information, etc. may be referred to as category type features; clicking, purchasing, trading frequency and the like of different users on different commodities can be called as displaying cross features;

s4, training a plurality of models according to the training data set determined in the S2 and different types of training features in the S3. Specifically, training a Gradient Boosting Decision Tree (GBDT) prediction model by using posterior statistical features and display cross features; training a Factorization (FM) prediction model by using posterior statistical features, category features, display cross features and a prediction output result of a promotion decision tree prediction model; training a model of a combination of a multi-layer perceptron and a factorizer by using posterior statistical features, category type features, display cross features, user behavior sequence features and prediction output results of a gradient lifting decision tree prediction model;

S5, predicting the test sample set in S2 by using the prediction model trained in S4, and recommending commodities to the user in S1.

In addition, compared with the prior art, the embodiment has the following advantages:

1. the characteristics of different attributes are divided, different models are trained based on the characteristics, the characteristics of various models are fully utilized, and the functions of the models are exerted as much as possible;

2. the results of various prediction models are fused, so that the advantages of various models are fully absorbed, and compared with a single model recommendation method, the commodity recommendation accuracy is greatly improved.

In summary, according to the product recommendation method provided by the embodiment of the application, N target products are obtained, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; acquiring a data information set based on N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of target objects, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on N target products, the M pieces of historical behavior data at least comprise target objects, and T and M are positive integers larger than 1; s prediction models are determined, wherein the prediction models are used for predicting the recommendation sequence of recommending N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; according to the data information set and the S prediction models, determining the target recommendation sequence of recommending N target products, recommending N target products to the target object according to the target recommendation sequence, and solving the problem of low accuracy of recommending products to users in the related technology. The method comprises the steps of obtaining a plurality of target products to be recommended to a user, obtaining a data information set based on the plurality of target products, determining a target recommendation sequence for recommending the plurality of target products according to the data information set and a plurality of prediction models, and recommending the plurality of target products to the user according to the target recommendation sequence, so that the problem of low accuracy of recommending the products to the user in the related art is avoided, and the effect of improving the accuracy of recommending the products to the user is achieved.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.

The embodiment of the application also provides a recommending device for the product, and the recommending device for the product can be used for executing the recommending method for the product. The following describes a recommending device for products provided in the embodiments of the present application.

Fig. 4 is a schematic diagram of a recommendation device for products provided according to an embodiment of the present application. As shown in fig. 4, the apparatus includes: a first acquisition unit 401, a second acquisition unit 402, a first determination unit 403, and a first processing unit 404.

Specifically, the first obtaining unit 401 is configured to obtain N target products, where the N target products are products to be recommended to a target object, and N is a positive integer greater than 1;

the second obtaining unit 402 is configured to obtain a data information set based on N target products, where the data information set at least includes: t pieces of historical behavior data, attribute information of each target product and attribute information of target objects, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on N target products, the M pieces of historical behavior data at least comprise target objects, and T and M are positive integers larger than 1;

The first determining unit 403 is configured to determine S prediction models, where the prediction models are used for predicting a recommendation order of recommending N target products, and the S prediction models at least include: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1;

the first processing unit 404 is configured to determine a target recommendation order for recommending N target products according to the data information set and the S prediction models, and recommend N target products to the target object according to the target recommendation order.

In summary, in the product recommendation device provided in the embodiment of the present application, N target products are obtained through the first obtaining unit 401, where N target products are products to be recommended to a target object, and N is a positive integer greater than 1; the second obtaining unit 402 obtains a data information set based on N target products, where the data information set at least includes: t pieces of historical behavior data, attribute information of each target product and attribute information of target objects, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on N target products, the M pieces of historical behavior data at least comprise target objects, and T and M are positive integers larger than 1; the first determining unit 403 determines S prediction models, where the prediction models are used for predicting a recommendation order of recommending N target products, and the S prediction models include at least: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; the first processing unit 404 determines a target recommendation sequence of recommending N target products according to the data information set and the S prediction models, and recommends N target products to the target object according to the target recommendation sequence, thereby solving the problem of low accuracy of recommending products to the user in the related art. The method comprises the steps of obtaining a plurality of target products to be recommended to a user, obtaining a data information set based on the plurality of target products, determining a target recommendation sequence for recommending the plurality of target products according to the data information set and a plurality of prediction models, and recommending the plurality of target products to the user according to the target recommendation sequence, so that the problem of low accuracy of recommending the products to the user in the related art is avoided, and the effect of improving the accuracy of recommending the products to the user is achieved.

Optionally, in the product recommendation device provided in the embodiment of the present application, the T pieces of historical behavior data at least include: the first processing unit includes: the first processing module is used for inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of N target products, and outputting the first recommendation sequence; the second processing module is used for inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict the recommendation sequence of recommending N target products and outputting the second recommendation sequence; the third processing module is used for inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of N target products, and outputting the third recommendation sequence; the first determining module is used for determining a target recommendation sequence for recommending N target products based on the second recommendation sequence and the third recommendation sequence.

Optionally, in the recommending apparatus for a product provided in the embodiment of the present application, the first determining module includes: the first determining submodule is used for determining the score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; the second determining submodule is used for determining the score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; the third determination submodule is used for respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and the fourth determining submodule is used for determining a target recommendation sequence for recommending N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

Optionally, in the recommendation device for a product provided in the embodiment of the present application, S prediction models are obtained by: the third acquisition unit is used for acquiring a target training sample set, wherein the target training sample set at least comprises Q sample behavior data acquired in a history process, attribute information of W sample products and attribute information of P sample objects, the Q sample behavior data at least comprises behavior data of P sample objects on the W sample products, and Q, W and P are positive integers larger than 1; the first training unit is configured to perform learning training on each of the S original prediction models by using a target training sample set to obtain S prediction models, where the S original prediction models at least include: the first, second and third original prediction models.

Optionally, in the product recommendation device provided in the embodiment of the present application, the first training unit includes: the fourth processing module is used for classifying Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of sample objects on sample products, the second sample behavior data is behavior data of each sample object of the P sample objects on sample products, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product; the first training module is used for learning and training a first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain a first prediction model; the second training module is used for acquiring a prediction result output by the first prediction model, and learning and training a second original prediction model in the S original prediction models by utilizing the prediction result, second sample behavior data, third sample behavior data, attribute information of W sample products and attribute information of P sample objects to obtain a second prediction model; the third training module is used for learning and training a third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model; and the second determining module is used for obtaining S prediction models based on the first prediction model, the second prediction model and the third prediction model.

Optionally, in the product recommendation device provided in the embodiment of the present application, the second training module includes: the first acquisition sub-module is used for acquiring second sample behavior data and third sample behavior data; the first output sub-module is used for inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict the recommendation sequence of the recommended W sample products and outputting the predicted recommendation sequence of the recommended W sample products; and a fifth determining sub-module, configured to take the predicted recommended order of the recommended W sample products as a prediction result.

Optionally, in the product recommendation device provided in the embodiment of the present application, the third obtaining unit includes: a fifth processing module, configured to obtain the portrait information of each sample object in the P sample objects, and determine attribute information of the P sample objects based on the portrait information of each sample object, where the attribute information of the P sample objects at least includes: ID information of each sample object and sex information of each sample object; the first obtaining module is configured to obtain attribute information of W sample products, where the attribute information of W sample products at least includes: ID information of each sample product and category information of each sample product; the second acquisition module is used for acquiring behavior data of the P sample objects on the W sample products; the third determining module is used for determining Q sample behavior data according to the behavior data of the P sample objects on the W sample products; and the third acquisition module is used for acquiring a target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

The recommendation device for products includes a processor and a memory, where the first acquiring unit 401, the second acquiring unit 402, the first determining unit 403, the first processing unit 404, and the like are stored as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one, and the accuracy of recommending products to the user is improved by adjusting kernel parameters.

The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.

An embodiment of the present invention provides a computer-readable storage medium having stored thereon a program that, when executed by a processor, implements a recommendation method for the product.

The embodiment of the invention provides a processor which is used for running a program, wherein the program runs to execute a recommendation method of the product.

As shown in fig. 5, an embodiment of the present invention provides an electronic device, where the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, the following steps are implemented: obtaining N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; acquiring a data information set based on the N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1; determining S prediction models, wherein the prediction models are used for predicting and recommending the recommendation sequence of the N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; and determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

The processor also realizes the following steps when executing the program: the T historical behavior data includes at least: the method comprises the steps of determining a target recommendation sequence of recommending N target products according to a data information set and S prediction models, wherein the data information set comprises first historical behavior data, second historical behavior data and third historical behavior data, the first historical behavior data is the historical behavior data of the target objects on the target products, the second historical behavior data is the historical behavior data of each of the M objects on the target products, the third historical behavior data is the historical behavior data of each of the M objects on the target products, and the target recommendation sequence of recommending the N target products comprises: inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a first recommendation sequence; inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a second recommendation sequence; inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a third recommendation sequence; and determining the target recommendation order of recommending the N target products based on the second recommendation order and the third recommendation order.

The processor also realizes the following steps when executing the program: determining the target recommendation order to recommend the N target products based on the second recommendation order and the third recommendation order includes: determining a score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; determining a score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and determining the target recommendation sequence for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

The processor also realizes the following steps when executing the program: the S prediction models are obtained by the following modes: obtaining a target training sample set, wherein the target training sample set at least comprises Q sample behavior data, attribute information of W sample products and attribute information of P sample objects obtained in a history process, the Q sample behavior data at least comprises behavior data of the P sample objects on the W sample products, and Q, W and P are positive integers larger than 1; learning and training each original prediction model in S original prediction models by using the target training sample set to obtain the S prediction models, wherein the S original prediction models at least comprise: the first, second and third original prediction models.

The processor also realizes the following steps when executing the program: learning and training each original prediction model in the S original prediction models by using the target training sample set, wherein the step of obtaining the S prediction models comprises the following steps: classifying the Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of the sample objects on the sample products, the second sample behavior data is behavior data of each sample object of the P sample objects on the sample products, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product; learning and training the first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain the first prediction model; obtaining a prediction result output by the first prediction model, and learning and training the second original prediction model in the S original prediction models by using the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the second prediction model; learning and training the third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model; and obtaining the S prediction models based on the first prediction model, the second prediction model and the third prediction model.

The processor also realizes the following steps when executing the program: the obtaining of the prediction result output by the first prediction model comprises the following steps: acquiring the second sample behavior data and the third sample behavior data; inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict and recommend the recommendation sequence of the W sample products, and outputting the predicted recommendation sequence of the W sample products; and taking the predicted recommendation sequence of recommending the W sample products as the prediction result.

The processor also realizes the following steps when executing the program: obtaining the target training sample set includes: acquiring the portrait information of each sample object in the P sample objects, and determining the attribute information of the P sample objects based on the portrait information of each sample object, wherein the attribute information of the P sample objects at least comprises: ID information of each sample object and sex information of each sample object; acquiring attribute information of the W sample products, wherein the attribute information of the W sample products at least comprises: ID information of each sample product and category information of each sample product; acquiring behavior data of the P sample objects on the W sample products; determining the Q sample behavior data according to the behavior data of the P sample objects on the W sample products; and acquiring the target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

The device herein may be a server, PC, PAD, cell phone, etc.

The present application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: obtaining N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1; acquiring a data information set based on the N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1; determining S prediction models, wherein the prediction models are used for predicting and recommending the recommendation sequence of the N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1; and determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: the T historical behavior data includes at least: the method comprises the steps of determining a target recommendation sequence of recommending N target products according to a data information set and S prediction models, wherein the data information set comprises first historical behavior data, second historical behavior data and third historical behavior data, the first historical behavior data is the historical behavior data of the target objects on the target products, the second historical behavior data is the historical behavior data of each of the M objects on the target products, the third historical behavior data is the historical behavior data of each of the M objects on the target products, and the target recommendation sequence of recommending the N target products comprises: inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a first recommendation sequence; inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a second recommendation sequence; inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a third recommendation sequence; and determining the target recommendation order of recommending the N target products based on the second recommendation order and the third recommendation order.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: determining the target recommendation order to recommend the N target products based on the second recommendation order and the third recommendation order includes: determining a score value corresponding to each target product based on the second recommendation sequence to obtain N first score values; determining a score value corresponding to each target product based on the third recommendation sequence to obtain N second score values; respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model; and determining the target recommendation sequence for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: the S prediction models are obtained by the following modes: obtaining a target training sample set, wherein the target training sample set at least comprises Q sample behavior data, attribute information of W sample products and attribute information of P sample objects obtained in a history process, the Q sample behavior data at least comprises behavior data of the P sample objects on the W sample products, and Q, W and P are positive integers larger than 1; learning and training each original prediction model in S original prediction models by using the target training sample set to obtain the S prediction models, wherein the S original prediction models at least comprise: the first, second and third original prediction models.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: learning and training each original prediction model in the S original prediction models by using the target training sample set, wherein the step of obtaining the S prediction models comprises the following steps: classifying the Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of the sample objects on the sample products, the second sample behavior data is behavior data of each sample object of the P sample objects on the sample products, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product; learning and training the first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain the first prediction model; obtaining a prediction result output by the first prediction model, and learning and training the second original prediction model in the S original prediction models by using the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the second prediction model; learning and training the third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model; and obtaining the S prediction models based on the first prediction model, the second prediction model and the third prediction model.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: the obtaining of the prediction result output by the first prediction model comprises the following steps: acquiring the second sample behavior data and the third sample behavior data; inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict and recommend the recommendation sequence of the W sample products, and outputting the predicted recommendation sequence of the W sample products; and taking the predicted recommendation sequence of recommending the W sample products as the prediction result.

When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: obtaining the target training sample set includes: acquiring the portrait information of each sample object in the P sample objects, and determining the attribute information of the P sample objects based on the portrait information of each sample object, wherein the attribute information of the P sample objects at least comprises: ID information of each sample object and sex information of each sample object; acquiring attribute information of the W sample products, wherein the attribute information of the W sample products at least comprises: ID information of each sample product and category information of each sample product; acquiring behavior data of the P sample objects on the W sample products; determining the Q sample behavior data according to the behavior data of the P sample objects on the W sample products; and acquiring the target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. A method of recommending a product, comprising:

obtaining N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1;

acquiring a data information set based on the N target products, wherein the data information set at least comprises: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1;

Determining S prediction models, wherein the prediction models are used for predicting and recommending the recommendation sequence of the N target products, and the S prediction models at least comprise: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1;

and determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

2. The method of claim 1, wherein the T historical behavior data comprises at least: the method comprises the steps of determining a target recommendation sequence of recommending N target products according to a data information set and S prediction models, wherein the data information set comprises first historical behavior data, second historical behavior data and third historical behavior data, the first historical behavior data is the historical behavior data of the target objects on the target products, the second historical behavior data is the historical behavior data of each of the M objects on the target products, the third historical behavior data is the historical behavior data of each of the M objects on the target products, and the target recommendation sequence of recommending the N target products comprises:

Inputting the second historical behavior data and the third historical behavior data into the first prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a first recommendation sequence;

inputting the first recommendation sequence, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the second prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a second recommendation sequence;

inputting the first recommendation sequence, the first historical behavior data, the second historical behavior data, the third historical behavior data, the attribute information of each target product and the attribute information of the target object into the third prediction model to predict and recommend the recommendation sequence of the N target products, and outputting a third recommendation sequence;

and determining the target recommendation order of recommending the N target products based on the second recommendation order and the third recommendation order.

3. The method of claim 2, wherein determining the target recommendation order to recommend the N target products based on the second recommendation order and the third recommendation order comprises:

Determining a score value corresponding to each target product based on the second recommendation sequence to obtain N first score values;

determining a score value corresponding to each target product based on the third recommendation sequence to obtain N second score values;

respectively determining the weight corresponding to the second prediction model and the weight corresponding to the third prediction model;

and determining the target recommendation sequence for recommending the N target products based on the N first score values, the N second score values, the weights corresponding to the second prediction model and the weights corresponding to the third prediction model.

4. The method of claim 1, wherein the S predictive models are obtained by:

obtaining a target training sample set, wherein the target training sample set at least comprises Q sample behavior data, attribute information of W sample products and attribute information of P sample objects obtained in a history process, the Q sample behavior data at least comprises behavior data of the P sample objects on the W sample products, and Q, W and P are positive integers larger than 1;

learning and training each original prediction model in S original prediction models by using the target training sample set to obtain the S prediction models, wherein the S original prediction models at least comprise: the first, second and third original prediction models.

5. The method of claim 4, wherein learning each of the S original predictive models using the set of target training samples comprises:

classifying the Q sample behavior data in the target training sample set to obtain first sample behavior data, second sample behavior data and third sample behavior data, wherein the first sample behavior data is behavior data of the sample objects on the sample products, the second sample behavior data is behavior data of each sample object of the P sample objects on the sample products, and the third sample behavior data is behavior data of each sample object of the P sample objects on each sample product;

learning and training the first original prediction model in the S original prediction models by using the second sample behavior data and the third sample behavior data to obtain the first prediction model;

obtaining a prediction result output by the first prediction model, and learning and training the second original prediction model in the S original prediction models by using the prediction result, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain the second prediction model;

Learning and training the third initial prediction model in the S initial prediction models by using the prediction result, the first sample behavior data, the second sample behavior data, the third sample behavior data, the attribute information of the W sample products and the attribute information of the P sample objects to obtain a third prediction model;

and obtaining the S prediction models based on the first prediction model, the second prediction model and the third prediction model.

6. The method of claim 5, wherein obtaining the prediction result output by the first prediction model comprises:

acquiring the second sample behavior data and the third sample behavior data;

inputting the second sample behavior data and the third sample behavior data into the first prediction model to predict and recommend the recommendation sequence of the W sample products, and outputting the predicted recommendation sequence of the W sample products;

and taking the predicted recommendation sequence of recommending the W sample products as the prediction result.

7. The method of claim 4, wherein obtaining a set of target training samples comprises:

Acquiring the portrait information of each sample object in the P sample objects, and determining the attribute information of the P sample objects based on the portrait information of each sample object, wherein the attribute information of the P sample objects at least comprises: ID information of each sample object and sex information of each sample object;

acquiring attribute information of the W sample products, wherein the attribute information of the W sample products at least comprises: ID information of each sample product and category information of each sample product;

acquiring behavior data of the P sample objects on the W sample products;

determining the Q sample behavior data according to the behavior data of the P sample objects on the W sample products;

and acquiring the target training sample set based on the attribute information of the P sample objects, the attribute information of the W sample products and the Q sample behavior data.

8. A recommendation device for a product, comprising:

the first acquisition unit is used for acquiring N target products, wherein the N target products are products to be recommended to a target object, and N is a positive integer greater than 1;

the second obtaining unit is configured to obtain a data information set based on the N target products, where the data information set at least includes: t pieces of historical behavior data, attribute information of each target product and attribute information of the target object, wherein the T pieces of historical behavior data at least comprise historical behavior data of M objects on the N target products, the M pieces of historical behavior data at least comprise the target objects, and T and M are positive integers larger than 1;

The first determining unit is configured to determine S prediction models, where the prediction models are used to predict and recommend a recommendation order of the N target products, and the S prediction models at least include: the system comprises a first prediction model, a second prediction model and a third prediction model, wherein the first prediction model at least comprises a decision tree model, the second prediction model at least comprises a factorizer model, the third prediction model at least comprises a model formed by combining a multi-layer perceptron model and the factorizer model, and S is a positive integer greater than 1;

the first processing unit is used for determining a target recommendation sequence for recommending the N target products according to the data information set and the S prediction models, and recommending the N target products to the target object according to the target recommendation sequence.

9. A computer-readable storage medium, characterized in that the storage medium stores a program, wherein the program performs the recommendation method of the product according to any one of claims 1 to 7.

10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of recommending an article of manufacture of any of claims 1-7.