CN115757956A

CN115757956A - Combined product recommendation method and device, electronic equipment and storage medium

Info

Publication number: CN115757956A
Application number: CN202211458360.7A
Authority: CN
Inventors: 罗冬阳
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2022-11-17
Filing date: 2022-11-17
Publication date: 2023-03-07

Abstract

The invention provides a combined product recommendation method and device, electronic equipment and a computer-readable storage medium. The invention provides a combined product recommendation method, which comprises the following steps: acquiring characteristic data and consumption data of a current user; determining a guest group corresponding to a current user in a preset guest group according to the feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group; inputting feature data of a current user into a machine learning model trained in advance to obtain probabilities that the current user purchases different preset products respectively; and determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively. The combined product recommendation method improves the rationality of combined product recommendation, thereby improving the accuracy of combined product recommendation.

Description

Combined product recommendation method and device, electronic equipment and storage medium

Technical Field

The invention relates to the technical field of intelligent decision, in particular to a combined product recommendation method and device, electronic equipment and a computer readable storage medium.

Background

In the current big data era with developed internet, many internet enterprises can have multiple products under their own flag. For example, in the car insurance service, when a user selects a business insurance, a certain insurance grade or a plurality of insurance grades of a plurality of alternative business insurance grades can be selected; most of the cognitive information of the users may be limited, so that the combined product is recommended as accurately as possible, which is very significant for improving the sales volume and the operating profit of the product. However, the existing combined product recommendation scheme has fewer dimensions, which results in lower rationality of combined product recommendation and thus lower accuracy of combined product recommendation.

Disclosure of Invention

The invention aims to provide a combined product recommendation method, a combined product recommendation device, electronic equipment and a computer readable storage medium, so as to solve the technical problem that the combined product recommendation accuracy is low in the prior art.

The technical scheme of the invention is as follows, and provides a combined product recommendation method, which comprises the following steps:

acquiring characteristic data and consumption data of a current user;

determining a guest group corresponding to a current user in a preset guest group according to feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group, wherein the preset guest group is determined according to feature data and consumption data of a preset number of users, and the preset guest group comprises a guest group where each user in the preset number is located and an initial recommended product combination corresponding to each guest group;

inputting feature data of a current user into a pre-trained machine learning model to obtain probabilities that the current user purchases different preset products respectively, wherein the machine learning module is obtained by training according to sample feature data and sample consumption data;

and determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively.

Further, the step of determining the preset guest cluster includes:

acquiring characteristic data and consumption data of a preset number of users;

portraying the feature data to obtain user portrait data, and determining all guest groups of the preset number of users and the guest group of each user in the preset number according to the user portrait data;

and determining an initial recommended product combination corresponding to each customer group according to the customer group where each user is located in the preset number and the consumption data of each user.

Further, the determining all the guest groups of the preset number of users and the guest group of each user in the preset number according to the user portrait data includes:

and determining a passenger group factor and the representative attribute of the passenger group factor according to the user image data, and determining all passenger groups of the preset number of users and the passenger group of each user in the preset number according to the passenger group factor and the representative attribute of the passenger group factor.

Further, determining a guest group corresponding to the current user in a preset guest group according to the feature data of the current user, including:

according to the characteristic data of the current user, determining a guest group factor corresponding to the current user and a representative attribute corresponding to the guest group factor, carrying out similarity comparison on the representative attribute corresponding to the guest group factor and the representative attributes of the guest group factors corresponding to all guest groups, and determining the guest group corresponding to the current user according to a similarity comparison result.

Further, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively, including:

if the preset product is a product in the initial recommended product combination, taking the preset product as a final recommended product when the probability that the current user purchases the preset product is greater than or equal to a first preset probability;

if the preset product is not a product in the initial recommended product combination, taking the preset product as a final recommended product when the probability that the current user purchases the preset product is greater than or equal to a second preset probability;

and taking all the final recommended products as final recommended product combinations.

Further, the consumption data of the current user comprises products purchased by the current user in history and the time for purchasing the products, and the method further comprises the step of determining a pre-recommended product according to the products purchased by the current user in history and the time for purchasing the products;

correspondingly, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively, including: and determining a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended products and the probability that the current user purchases different preset products respectively.

Further, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended product and the probability that the current user purchases different preset products, respectively, includes:

if the preset product is a product in the initial recommended product combination or the pre-recommended product, taking the preset product as a final recommended product when the probability that the current user purchases the preset product is greater than or equal to a first preset probability;

if the preset product is not the product in the initial recommended product combination and is not the pre-recommended product, taking the preset product as a final recommended product when the probability that the current user purchases the preset product is greater than or equal to a second preset probability;

According to another technical scheme, the invention provides a combined product recommendation device, which comprises a data acquisition module, an initial recommendation module, a probability acquisition module and a final recommendation module;

the data acquisition module is used for acquiring characteristic data and consumption data of a current user;

the initial recommendation module is used for determining a guest group corresponding to a current user in a preset guest cluster according to feature data of the current user and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest cluster, wherein the preset guest cluster is determined according to feature data and consumption data of a preset number of users, and the preset guest cluster comprises a guest group where each user in the preset number is located and an initial recommended product combination corresponding to each guest group;

the probability acquisition module is used for inputting the feature data of the current user into a machine learning model trained in advance to obtain the probability that the current user purchases different preset products respectively, wherein the machine learning module is obtained by training according to the sample feature data and the sample consumption data;

and the final recommending module is used for determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively.

Another aspect of the present invention provides an electronic device, including a memory and a processor, where the memory stores a computer program executable by the processor, and the processor implements the combination product recommendation method according to any one of the above aspects when executing the computer program.

Another aspect of the present invention provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the combination product recommendation method according to any one of the above aspects.

The invention has the beneficial effects that: acquiring characteristic data and consumption data of a current user; determining a guest group corresponding to a current user in a preset guest group according to feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group, wherein the preset guest group is determined according to feature data and consumption data of a preset number of users, and the preset guest group comprises the guest group where each user in the preset number is located and the initial recommended product combination corresponding to each guest group; inputting feature data of a current user into a pre-trained machine learning model to obtain probabilities that the current user purchases different preset products respectively, wherein the machine learning module is obtained by training according to sample feature data and sample consumption data; determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively; by the technical scheme, the rationality of the combined product recommendation is improved, and therefore the accuracy of the combined product recommendation is improved.

Drawings

FIG. 1 is a flowchart illustrating a method for recommending a combination product according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a combined product recommendation device according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In the description of the present application, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance, nor order. The terms "comprising," "including," "having," and variations thereof in this specification mean "including, but not limited to," unless expressly specified otherwise.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

Fig. 1 is a flowchart illustrating a method for recommending a combination product according to an embodiment of the present invention. It should be noted that the method for recommending a combined product according to the present invention is not limited to the flow sequence shown in fig. 1 if the results are substantially the same. As shown in fig. 1, the method for recommending a combination product mainly includes the following steps:

s1, acquiring feature data and consumption data of a current user;

the characteristic data can include user age, user gender, user income level, constellation, service index data and the like, for example, for the car insurance service, the service index data can include data of car age, car purchase price, whether the car purchase price belongs to a price sensitive type, whether the car purchase price belongs to a renewal user or not and the like; the consumption data comprises products historically purchased by the user, and the historically purchased products can be historically purchased products in a period of time;

s2, determining a guest group corresponding to the current user in a preset guest group according to feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group, wherein the preset guest group is determined according to feature data and consumption data of a preset number of users, and the preset guest group comprises a guest group where each user in the preset number is located and an initial recommended product combination corresponding to each guest group;

the preset number of users can be all users, the all users are all client groups corresponding to the current company business, and the initial recommended product combination can be at least one product;

s3, inputting the feature data of the current user into a machine learning model trained in advance to obtain the probability that the current user purchases different preset products respectively, wherein the machine learning module is obtained by training according to sample feature data and sample consumption data;

the modeling and training of the machine learning model (logistic regression model or support vector machine model and the like) can be performed on different preset products purchased by a user respectively according to the sample characteristic data and the sample consumption data and by adopting a sklern data analysis framework. After the machine learning model is trained, the trained machine learning model can acquire the probability of the user purchasing the preset product according to the characteristic data of the user and the consumption data. Some products may have periodicity, such as car insurance, the policy is generally once a year sign, if the user is still in the using period, the machine learning model judges whether the user purchases the product, so that the meaning is not great, when modeling and training the machine learning model, the characteristics and the periodicity of the product are combined, for example, for once a year sign product, the probability of purchasing the product in the future can be predicted according to the data of the latest 12 months.

And S4, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively. Wherein the final recommended product combination may be at least one product.

According to the embodiment of the invention, the characteristic data and the consumption data of the current user are obtained; determining a guest group corresponding to a current user in a preset guest group according to the feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group; inputting the characteristic data of the current user into a machine learning model trained in advance to obtain the probability that the current user purchases different preset products respectively; determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively; the rationality of the recommendation of the combined product is improved, and therefore the accuracy of the recommendation of the combined product is improved.

In an optional embodiment, the determining of the preset guest cluster includes:

acquiring characteristic data and consumption data of a preset number of users;

The acquired feature data of the user may include desensitization data or detail data, and the feature data may be used to perform a 360-degree user portrait on the user.

In an embodiment, the portraying the feature data to obtain user portrayal data includes: and performing service standard portrait, equal frequency portrait or equal width portrait on the characteristic data to obtain user portrait data. Wherein, some data indexes of the characteristic data can be performed with business standard portrait, for example, age data can be performed with business standard portrait, 0-3 years old can be labeled as infant stage, 4-13 years old can be labeled as juvenile stage, etc., so as to realize portrait by means of business standard (business experience); other data indicators of the feature data may also be rendered as equi-frequency images, e.g., the data indicators may be equally labeled as good, neutral, and bad; other data indexes of the characteristic data can also be subjected to equal-width portrayal, for example, the numerical values 1-3 are marked as small, the numerical values 4-6 are marked as medium, and the numerical values 7-9 are marked as large; when the user portrays, the different portrayal modes can be flexibly used.

In an optional implementation manner, the determining, according to the user portrait data, all guest groups of the preset number of users and a guest group in which each user in the preset number is located includes:

In a specific embodiment, a guest group factor and a representative attribute of the guest group factor are determined according to the user image data, and a region difference can be considered in determining the guest group factor and the representative attribute of the guest group factor; if the number of the passenger group factors is more, the passenger group is richer, and the acquisition complexity of the passenger group is relatively increased; if one passenger group includes 7 passenger group factors, the representative attributes of the first passenger group factor are 3, the representative attributes of the second passenger group factor are 5, the representative attributes of the third passenger group factor are 4, the representative attributes of the fourth passenger group factor are 3, the representative attributes of the fifth passenger group factor are 5, the representative attributes of the sixth passenger group factor are 5, and the representative attributes of the seventh passenger group factor are 4, the number of the passenger group is 3 × 5 × 4 × 3 × 5 =18000. Note that, if the age is the guest group factor, the attribute at the juvenile stage is a representative attribute.

In another specific embodiment, all the guest groups of the preset number of users and the guest group where each user in the preset number is located are determined according to the guest group factors and the representative attributes of the guest group factors, and a guest group table can be obtained according to the guest group where each user in the preset number is located, as shown in table 1.

TABLE 1 customer group table

User ID	Belonging to the guest group	Purchasing a product
			252155	Passenger group B	A，B，C
262155	Guest group A	A，D
			262175	Passenger group B	C，D
263175	Passenger group B	A，B，C
			…	…	…

In table 1, the last column indicates products historically purchased by different users, and the initial recommended product combination corresponding to the current user is determined according to the guest group corresponding to the current user, the guest group in which each user in the preset number is located, and the initial recommended product combination corresponding to each guest group, where the initial recommended product combination corresponding to the guest group is obtained by counting and sorting products historically purchased by all users in the guest group, and products historically purchased by previous preset names are used as the initial recommended product combination, for example, as shown in table 1, the initial recommended product combinations corresponding to the guest group B are a, B, and C.

In a specific embodiment, the configuration information table may be output by using big data, where the configuration information table includes an initial recommended product combination for each customer group, and if a new product exists, the product configuration may be performed by using the configuration information table.

In an optional embodiment, determining, in a preset guest cluster, a guest cluster corresponding to a current user according to feature data of the current user includes:

In a specific embodiment, the guest group factor corresponding to the current user needs to be the same as the guest group factors corresponding to all the guest groups, and the representative attribute corresponding to the guest group factor is compared with the representative attributes of the guest group factors corresponding to all the guest groups to obtain a similarity comparison result, which may be a similarity score, where if one guest group includes n guest group factors, the similarity score may be calculated as a similarity score = weight x1 of the 1 st guest group factor + weight x2+ … of the 2 nd guest group factor + weight xn of the nth guest group factor, the weights of the first guest group factor to the nth guest group factor may be preset, and x1... Xn represents 1 or 0, and if the representative attributes of the two guest group factors are the same, 1 is taken, and otherwise 0 is taken.

In an optional embodiment, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probabilities that the current user purchases different preset products, respectively, includes:

In one embodiment, the feature data of the current user is input into the pre-trained machine learning model to obtain probabilities that the current user purchases different preset products, for example, if there are several types (for example, 5 types) of the preset products, the probabilities that the current user purchases the different preset products are predicted by the pre-trained machine learning model, and the probability values are between 0 and 1, and the probability table of the current user purchasing the preset products is shown in table 2.

Table 2 probability table for current user to purchase preset products

Product name	Probability of
		Product A	0.71
Product B	0.42
		Product C	0.83

In a specific embodiment, the second preset probability may be greater than the first preset probability, for example, the first preset probability may be 0.2, and the second preset probability may be 0.5; if the preset product is not a product in the initial recommended product combination, when the probability that the current user purchases the preset product is greater than or equal to 0.5, determining that the user will purchase the preset product, and taking the preset product as a final recommended product; as shown in table 2, the product a and the product C can be used as the final recommended products.

If the preset product is a product in the initial recommended product combination, when the probability that the current user purchases the preset product is greater than or equal to 0.2, the user is judged to purchase the preset product, and the preset product is taken as a final recommended product.

In an optional embodiment, the consumption data of the current user includes products purchased by the current user in history and time of purchasing the products, the method further includes determining pre-recommended products according to the products purchased by the current user in history and the time of purchasing the products, and correspondingly, determining a final recommended product combination according to an initial recommended product combination corresponding to the current user and probabilities that the current user purchases different pre-recommended products, respectively, including: and determining a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended products and the probability that the current user purchases different preset products respectively.

In a specific embodiment, the consumption data of the current user includes products purchased by the current user in a history and time of purchasing the products, and it is determined whether to recommend the products purchased in the history or not according to the products purchased by the current user in the history and the time of purchasing the products, that is, it is determined that a product is to be pre-recommended, for example, if the current time is greater than or equal to the recommended time of the product a, it is considered that the user is using the product a and does not recommend the product a, otherwise, the product a is recommended as the pre-recommended product.

In an optional embodiment, determining a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended product, and the probability that the current user purchases different preset products, respectively, includes:

In a specific embodiment, a final recommended product combination is determined according to the initial recommended product combination corresponding to the current user, the pre-recommended product, and the probability that the current user purchases different pre-set products, respectively, and the three information data, and if the first pre-set probability is 0.2 and the second pre-set probability is 0.5, a final recommended product table can be obtained, as shown in table 3.

Table 3 final recommended products table

Product name	Type of product	Probability of purchase	Whether or not to recommend
				Product A	Pre-recommended product	0.20	Is that
Product B	General products	0.49	Whether or not
				Product(s)C	General products	0.5	Is that
Product D	Pre-recommended product	0.19	Whether or not
				Product E	Initial recommended product portfolio	0.20	Is that
Product F	Initial recommended product portfolio	0.19	Whether or not

If the preset product is a product in the initial recommended product combination or the pre-recommended product, taking the preset product as a final recommended product when the probability that the current user purchases the preset product is greater than or equal to 0.2; if the preset product is not a product in the initial recommended product combination and is not the pre-recommended product, when the probability that the current user purchases the preset product is greater than or equal to 0.5, the preset product is taken as a final recommended product, and then in table 3, a product a, a product C, and a product E are final recommended products.

According to the combined product recommendation method provided by the embodiment of the invention, the characteristic data and the consumption data of the current user are obtained; determining a guest group corresponding to a current user in a preset guest group according to the feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group; inputting the characteristic data of the current user into a machine learning model trained in advance to obtain the probability that the current user purchases different preset products respectively; determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probability that the current user purchases different preset products respectively; the rationality of the combined product recommendation is improved, and therefore the accuracy of the combined product recommendation is improved.

The combined product recommendation method provided by the embodiment of the invention fuses the initial recommended product combination corresponding to the current user, the pre-recommended product and the probability that the current user purchases different preset products respectively, and determines the final recommended product combination, so that the rationality of combined product recommendation is greatly improved on the premise of fully utilizing data resources. The combined product recommendation method provided by the embodiment of the invention determines the initial recommended product combination by creating the portrait, considers the periodicity of the product in the machine learning model modeling process and the process of determining the pre-recommended product, and can well ensure the validity of the product recommendation result.

The combined product recommendation method provided by the embodiment of the invention can be constructed based on artificial intelligence, and related data is acquired and processed based on an artificial intelligence technology, so that unattended artificial intelligence combined product recommendation is realized. Among them, artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.

The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

Fig. 2 is a schematic structural diagram of a combined product recommendation apparatus according to an embodiment of the present invention, and as shown in fig. 2, the combined product recommendation apparatus 20 includes a data obtaining module 21, an initial recommendation module 22, a probability obtaining module 23, and a final recommendation module 24;

the data acquisition module 21 is configured to acquire feature data and consumption data of a current user;

the initial recommendation module 22 is configured to determine, according to feature data of a current user, a guest group corresponding to the current user in a preset guest cluster, and determine, according to the guest group corresponding to the current user and the preset guest cluster, an initial recommended product combination corresponding to the current user, where the preset guest cluster is determined according to feature data and consumption data of a preset number of users, and the preset guest cluster includes a guest group where each user in the preset number is located and an initial recommended product combination corresponding to each guest group;

the probability obtaining module 23 is configured to input feature data of a current user into a machine learning model trained in advance, so as to obtain probabilities that the current user purchases different preset products respectively, where the machine learning module is obtained by training according to sample feature data and sample consumption data;

the final recommending module 24 is configured to determine a final recommended product combination according to the initial recommended product combination corresponding to the current user and probabilities that the current user purchases different preset products respectively.

In an optional embodiment, the combined product recommendation apparatus 20 further includes a preset guest cluster determining module, where the preset guest cluster determining module is configured to obtain feature data and consumption data of a preset number of users; portraying the feature data to obtain user portrait data, and determining all guest groups of the preset number of users and the guest group of each user in the preset number according to the user portrait data; and determining an initial recommended product combination corresponding to each customer group according to the customer group where each user is located in the preset number and the consumption data of each user.

In an optional implementation manner, the preset guest cluster determining module is further configured to determine a guest cluster factor and a representative attribute of the guest cluster factor according to the user portrait data, and determine all guest clusters of the preset number of users and a guest cluster in which each user in the preset number is located according to the guest cluster factor and the representative attribute of the guest cluster factor.

In an optional embodiment, the initial recommendation module 22 is further configured to determine, according to the feature data of the current user, a guest group factor corresponding to the current user and a representative attribute corresponding to the guest group factor, perform similarity comparison between the representative attribute corresponding to the guest group factor and the representative attributes of the guest group factors corresponding to all guest groups, and determine a guest group corresponding to the current user according to a result of the similarity comparison.

In an optional embodiment, the determining, by the final recommending module 24, a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probabilities that the current user purchases different preset products respectively includes:

In an optional embodiment, the consumption data of the current user includes products purchased by the current user in history and time of purchasing the products, the combined product recommendation apparatus 20 further includes a pre-recommended product determination module, configured to determine a pre-recommended product according to the products purchased by the current user in history and the time of purchasing the products; correspondingly, the final recommending module 24 is further configured to determine a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended product, and the probability that the current user purchases different preset products, respectively.

In an optional embodiment, the determining, by the final recommending module 24, a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended product, and the probability that the current user purchases different preset products respectively includes:

The combined product recommendation device 20 provided by the embodiment of the invention acquires the characteristic data and the consumption data of the current user through the data acquisition module 21; determining a guest group corresponding to a current user in a preset guest group through an initial recommendation module 22 according to feature data of the current user, and determining an initial recommended product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest group; inputting the feature data of the current user into a machine learning model trained in advance through a probability acquisition module 23 to obtain the probability that the current user purchases different preset products respectively; determining a final recommended product combination through a final recommending module 24 according to the initial recommended product combination corresponding to the current user and the probabilities that the current user purchases different preset products respectively; the rationality of the combined product recommendation is improved, and therefore the accuracy of the combined product recommendation is improved.

Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. As shown in fig. 3, the electronic device 30 includes a processor 31 and a memory 32 communicatively coupled to the processor 31.

The memory 32 stores program instructions for implementing the combination product recommendation method of any of the above embodiments.

The processor 31 is operative to execute program instructions stored in the memory 32 to make a composed product recommendation.

The processor 31 may also be referred to as a CPU (Central Processing Unit). The processor 31 may be an integrated circuit chip having signal processing capabilities. The processor 31 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The storage medium of the embodiment of the present invention stores program instructions that can implement all the methods described above, and may be nonvolatile or volatile. The program instructions may be stored in the storage medium in the form of a software product, and include several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or terminal devices, such as a computer, a server, a mobile phone, and a tablet.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of modules is merely a division of logical functions, and an actual implementation may have another division, for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or modules, and may be in an electrical, mechanical or other form.

In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each module may exist alone physically, or two or more modules are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit. The above description is only an embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent processes performed by the present invention or directly or indirectly applied to other related technical fields are also included in the scope of the present invention.

While the foregoing is directed to embodiments of the present invention, it will be understood by those skilled in the art that various changes may be made without departing from the spirit and scope of the invention.

Claims

1. A combination product recommendation method, comprising the steps of:

acquiring characteristic data and consumption data of a current user;

2. The combination product recommendation method according to claim 1, wherein the step of determining the preset guest cluster comprises:

acquiring characteristic data and consumption data of a preset number of users;

3. The method of claim 2, wherein the determining, according to the user profile data, all the guest groups of the preset number of users and the guest group in which each user of the preset number is located comprises:

4. The combination product recommendation method according to claim 3, wherein determining the guest group corresponding to the current user in a preset guest group according to the feature data of the current user comprises:

5. The method for recommending a blended product according to claim 1, wherein determining a final recommended product combination according to the initial recommended product combination corresponding to the current user and the probabilities that the current user purchases different preset products, respectively, comprises:

6. The combination product recommendation method of claim 1, wherein the consumption data of the current user comprises products historically purchased by the current user and time of purchase of the products, the method further comprising determining a pre-recommended product based on the products historically purchased by the current user and the time of purchase of the products;

7. The method for recommending a combination product according to claim 6, wherein determining a final recommended product combination according to the initial recommended product combination corresponding to the current user, the pre-recommended product, and the probability that the current user purchases different pre-set products, respectively, comprises:

8. A combined product recommendation device is characterized by comprising a data acquisition module, an initial recommendation module, a probability acquisition module and a final recommendation module;

the initial recommendation module is used for determining a guest group corresponding to the current user in a preset guest cluster according to the feature data of the current user and determining an initial recommendation product combination corresponding to the current user according to the guest group corresponding to the current user and the preset guest cluster, wherein the preset guest cluster is determined according to the feature data and consumption data of a preset number of users, and the preset guest cluster comprises the guest group where each user in the preset number is located and the initial recommendation product combination corresponding to each guest group;

9. An electronic device comprising a memory, a processor, the memory storing a computer program executable by the processor, wherein the processor implements the combination product recommendation method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements a combined product recommendation method according to any one of claims 1 to 7.