CN113435969A

CN113435969A - Product recommendation method and device, electronic equipment and storage medium

Info

Publication number: CN113435969A
Application number: CN202110701556.3A
Authority: CN
Inventors: 刘冬
Original assignee: Weikun Shanghai Technology Service Co Ltd
Current assignee: Weikun Shanghai Technology Service Co Ltd
Priority date: 2021-06-23
Filing date: 2021-06-23
Publication date: 2021-09-24
Also published as: WO2022267160A1

Abstract

The application relates to the technical field of artificial intelligence, and particularly discloses a product recommendation method, a device, electronic equipment and a storage medium, wherein the product recommendation method comprises the following steps: acquiring downlink data and uplink behavior data of a user; performing feature extraction on the online behavior data to obtain a first feature vector; performing feature extraction on the downlink data to obtain a second feature vector; fusing the first feature vector and the second feature vector to obtain a fused feature vector; establishing a behavior portrait of the user according to the fusion feature vector; determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the field to which the behavior corresponding to the offline behavior data and the online behavior data of the user belongs; determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

Description

Product recommendation method and device, electronic equipment and storage medium

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a product recommendation method, a product recommendation device, electronic equipment and a storage medium.

Background

With the advent of the intelligent era, most businesses analyze the daily behaviors of users so as to determine the preferences of the users and accurately recommend commodities to the users. However, the traditional analysis method can only collect behavior data for a specific scene, so that the preference of a user is analyzed, the analysis result is too comprehensive, the accuracy is not high, and the recommendation accuracy of a product is low.

Disclosure of Invention

In order to solve the problems in the prior art, the embodiment of the application provides a product recommendation method, a product recommendation device, an electronic device and a storage medium, which can comprehensively and accurately analyze the preference of a user, and then improve the product recommendation accuracy.

In a first aspect, an embodiment of the present application provides a product recommendation method, including:

acquiring downlink data and uplink behavior data of a user;

performing feature extraction on the online behavior data to obtain a first feature vector;

performing feature extraction on the offline behavior data to obtain a second feature vector;

fusing the first feature vector and the second feature vector to obtain a fused feature vector;

establishing a behavior portrait of the user according to the fusion feature vector;

determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the fields to which the behaviors corresponding to the offline behavior data and the online behavior data of the user belong;

and determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

In a second aspect, an embodiment of the present application provides a product recommendation device, including:

the data acquisition module is used for acquiring the offline behavior data and the online behavior data of the user;

the characteristic extraction module is used for extracting the characteristics of the data of the line uplink to obtain a first characteristic vector; performing feature extraction on the offline behavior data to obtain a second feature vector;

the image establishing module is used for fusing the first feature vector and the second feature vector to obtain a fused feature vector; establishing a behavior portrait of the user according to the fusion feature vector;

the product recommendation module is used for determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the fields of the offline behavior data and the behaviors corresponding to the online behavior data of the user; and determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

In a third aspect, an embodiment of the present application provides an electronic device, including: a processor coupled to the memory, the memory for storing a computer program, the processor for executing the computer program stored in the memory to cause the electronic device to perform the method of the first aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having a computer program stored thereon, the computer program causing a computer to perform the method according to the first aspect.

In a fifth aspect, embodiments of the present application provide a computer program product comprising a non-transitory computer readable storage medium storing a computer program, the computer operable to cause the computer to perform a method according to the first aspect.

The implementation of the embodiment of the application has the following beneficial effects:

in the embodiment of the application, the fusion characteristics fusing the offline behavior and the online behavior of the user are obtained by acquiring the offline behavior data and the online behavior data of the user, and then respectively extracting and fusing the characteristics of the offline behavior data and the online behavior data. And then constructing a behavior portrait of the user according to the fusion characteristics, then determining preference information and a behavior field of the user, and finally determining a corresponding product to be recommended to the user according to the preference information and the behavior field. Therefore, online data and offline data are fused, so that behavior portrait of a client is more accurate and comprehensive, meanwhile, the selection range of products is further narrowed by combining the behavior field, the products are limited to products in the user behavior related field, the recommendation accuracy is improved, and the user experience is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a network architecture diagram of a product recommendation system according to an embodiment of the present disclosure;

fig. 2 is a schematic hardware structure diagram of a product recommendation device according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of a product recommendation method according to an embodiment of the present application;

fig. 4 is a schematic flowchart of a method for acquiring online behavior data of a user through offline data of the user according to an embodiment of the present application;

fig. 5 is a schematic flowchart of a method for acquiring offline data and identity information of a user according to an embodiment of the present disclosure;

fig. 6 is a schematic flowchart of a method for acquiring offline behavior data of a user according to online behavior data of the user according to an embodiment of the present application;

fig. 7 is a schematic flowchart of a method for acquiring online behavior data and identity information of a user according to an embodiment of the present disclosure;

fig. 8 is a block diagram illustrating functional modules of a product recommendation device according to an embodiment of the present disclosure;

fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments obtained by a person of ordinary skill in the art without any inventive work based on the embodiments in the present application are within the scope of protection of the present application.

The terms "first," "second," "third," and "fourth," etc. in the description and claims of this application and in the accompanying drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

Reference herein to "an embodiment" means that a particular feature, result, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. 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.

First, as shown in fig. 1, fig. 1 is a network architecture diagram of a product recommendation system according to the present application, and the product recommendation system 100 includes an online acquisition system 101, an offline acquisition system 102, and an analysis system 103. The online acquisition system 101 comprises a plurality of triggers distributed in an online target webpage, and when the target webpage is accessed, the triggers are triggered to execute relevant online behavior acquisition; the offline acquisition system 102 comprises a plurality of collectors distributed in offline target sites, and when the target sites are accessed, the collectors are triggered to execute the acquisition of related offline; the analysis system 103 is connected to the online acquisition system 101 and the offline acquisition system 102, and is configured to receive data acquired by the online acquisition system 101 and the offline acquisition system 102, and perform analysis processing.

In addition, the product recommendation method provided by the application can be applied to the scenes of virtual service recommendation, entity commodity recommendation, virtual commodity recommendation and other services. In the application, the process of the product recommendation method is mainly described by taking entity commodity recommendation as an example, and the product recommendation methods in other scenes are similar to the product recommendation method in the entity commodity recommendation scene, and are not described here.

Based on this, the plurality of collectors in the offline acquisition system 102 may be cameras or the like disposed in offline brick and mortar stores.

Next, referring to fig. 2, fig. 2 is a schematic diagram of a hardware structure of a product recommendation device according to an embodiment of the present disclosure. The product recommendation device 200 includes at least one processor 201, a communication line 202, a memory 203, and at least one communication interface 204.

In this embodiment, the processor 201 may be a general processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more ics for controlling the execution of programs according to the present disclosure.

The communication link 202, which may include a path, carries information between the aforementioned components.

The communication interface 204 may be any transceiver or other device (e.g., an antenna, etc.) for communicating with other devices or communication networks, such as an ethernet, RAN, Wireless Local Area Network (WLAN), etc.

The memory 203 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

In this embodiment, the memory 203 may be independent and connected to the processor 201 through the communication line 202. The memory 203 may also be integrated with the processor 201. The memory 203 provided in the embodiments of the present application may generally have a nonvolatile property. The memory 203 is used for storing computer execution instructions for executing the scheme of the application, and is controlled by the processor 201 to execute. The processor 201 is configured to execute computer-executable instructions stored in the memory 103, thereby implementing the methods provided in the embodiments described below.

In alternative embodiments, computer-executable instructions may also be referred to as application code, which is not specifically limited in this application.

In alternative embodiments, processor 201 may include one or more CPUs, such as CPU0 and CPU1 of FIG. 1.

In an alternative embodiment, the product recommendation device 200 can include multiple processors, such as processor 201 and processor 207 in FIG. 2. Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).

In an alternative embodiment, if the product recommendation apparatus 200 is a server, the product recommendation apparatus 200 may further include an output device 205 and an input device 206. The output device 205 is in communication with the processor 201 and may display information in a variety of ways. For example, the output device 205 may be a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display device, a Cathode Ray Tube (CRT) display device, a projector (projector), or the like. The input device 206 is in communication with the processor 201 and may receive user input in a variety of ways. For example, the input device 206 may be a mouse, a keyboard, a touch screen device, or a sensing device, among others.

The product recommendation device 200 may be a general-purpose device or a special-purpose device. The present embodiment does not limit the type of the product recommendation device 200.

Hereinafter, a product recommendation method disclosed in the present application will be explained:

referring to fig. 3, fig. 3 is a schematic flowchart of a product recommendation method according to an embodiment of the present disclosure. The product recommendation method comprises the following steps:

301: and acquiring downlink behavior data and uplink behavior data of the user.

Due to the popularization of intelligent equipment and the Internet, the daily life of people is closely related to the network. Nowadays, most of the activities in daily life of people can be completed through intelligent equipment and a network. Meanwhile, the behavior (online behavior) performed by the smart device on the network cannot completely get rid of the relationship with the actual behavior (offline behavior), and likewise, the behavior performed in the real world cannot completely get rid of the relationship with the behavior on the network. Based on this, it can be simply considered that the on-line behavior and the off-line behavior are mutually interleaved, and the complete behavior data of the user is formed together.

Due to the above-mentioned interleaving relationship between the on-line behavior and the off-line behavior, in this embodiment, a method for acquiring the on-line behavior data of the user through the off-line behavior data of the user is provided, and as shown in fig. 4, the method includes:

401: and acquiring the downlink of the user as data and identity information.

In this embodiment, the identity information may be a predefined multidimensional unified ID, and specifically, the dimensions thereof are as follows: biometric IDs such as face, retina, iris, fingerprint, palm print, signature, voice, etc.; user electronic ID, such as system account, digital certificate, mobile phone number, identification number, etc.; and the user physical ID, such as an identity card, a bank card, a mobile phone and the like. Therefore, by collecting any one item of the multi-dimensional unified ID, the other relevant dimension ID can be obtained through the multi-dimensional unified ID.

Illustratively, following the scenario example of the physical goods recommendation described above, in this scenario, the offline scenario may be considered as a product purchase to an offline physical store. Therefore, the collected identity information may be a face of a biometric ID in the multi-dimensional unified ID. Based on this, the present embodiment provides a method for acquiring offline data and identity information of a user, as shown in fig. 5, the method includes:

501: behavior images of a user are received from an offline acquisition device disposed at a target location.

For example, in a physical product recommendation scenario, the target location may be an offline physical store, and the capture device may be a camera disposed on a shelf. Based on this, when the user selects the commodity, the action image when the user selects can be obtained through the camera.

502: and analyzing the behavior image frame by frame to obtain at least one sub-action image of the user.

For example, in a scenario of entity commodity recommendation, a set of standard actions may be preset, for example: standard actions of picking up goods, checking goods, putting back goods, taking goods away, exchanging goods and the like. And comparing the standard actions with action images obtained frame by frame in the action image, and taking the frame image with the similarity larger than a threshold value as a sub-action image.

In alternative embodiments, each action may correspond to a plurality of standard actions. In other words, each of the actions of picking up a product, viewing a product, putting back a product, taking away a product, exchanging a product, etc. may correspond to a standard action set, and each standard action set may contain at least one standard action picture.

503: and for each sub-action image in the at least one sub-action image, respectively performing feature extraction on each sub-action image to obtain at least one sub-action feature corresponding to the at least one sub-action image one to one.

504: and determining the downlink of the user as data according to at least one sub-action characteristic.

For example, the at least one sub-action feature may be arranged in a first order, resulting in a sub-action sequence, where the first order is an order in which the sub-action images corresponding to each of the at least one sub-action feature appear in the action image. Thus, the sequence of sub-actions that can be obtained can be regarded as one dynamic behavior represented by a feature corresponding to a plurality of static sub-actions.

And then, matching the sub-action sequence with a preset behavior sequence, and determining the line descending of the user as data. For example, the preset behavior sequence may be: select products, unselect products, hesitant selection, etc. Specifically, the action sequence of picking up a product is usually initiated by the action of picking up a product, and ended by the action of picking up a product, and other actions such as checking a product, putting back a product, and exchanging a product may be interspersed in between. Therefore, the behavior data of the user in the online entity shop can be determined by matching the sub-action sequence with the preset behavior sequence.

In addition, in an optional implementation manner, each behavior sequence matched by the user may be timed, the time taken by the user to complete each behavior is determined, and the time taken by the user to complete each behavior is also counted into the offline data of the user, so that the behavior data of the user is improved, and the subsequent analysis on the preference of the user is more complete and accurate.

505: and according to the behavior image, carrying out face recognition on the user and determining the identity information of the user.

In an alternative embodiment, the collecting device may also be a pressure sensor disposed on the shelf, so as to determine the commodity taken up by the user, the position taken up, the action of putting back, the action of taking multiple times, and the like according to the pressure change of the shelf. Therefore, the behavior data shot by the camera is supplemented, so that dead corners are prevented, and the condition that the behavior of products on the inner side of the shelf of a user cannot be collected by the camera is avoided.

Meanwhile, when the user pays, the information of the user such as the fingerprint, the mobile phone number, the bank card and the like can be acquired and used as the identity information of the user.

402: and determining the network account information and the equipment information associated with the user according to the identity information.

In this embodiment, as described above, the identity information is a predefined multidimensional unified ID, and therefore, the network account information and the device information of the user can be determined according to the association relationship between the constituent items in the multidimensional unified ID by obtaining the complete user identity information matched with any one of the constituent items.

403: and determining social dynamic information of the user according to the network account information.

In this embodiment, the social dynamic information of the user may be various types of dynamic shares sent by the user through the associated network account information. For example, the dynamic sharing sent by the network account in the previous preset time period may be obtained starting from the current time point. Specifically, assuming that the current time is 2021 year, 5 month and 27 day, the dynamic share issued by the network account in the previous month, i.e., from 2021 year, 4 month and 27 day to 2021 year, 5 month and 27 day, can be obtained. Furthermore, data screening can be performed on the obtained dynamic shares, and dynamic shares related to shopping are screened out and used as social dynamic information of the user.

404: and determining the page number of the target webpage browsed by the equipment corresponding to the equipment information and the browsing time of the target webpage according to the equipment information.

405: and taking the social dynamic information of the user, the page number of the target webpage browsed by the equipment corresponding to the equipment information and the browsing duration of the target webpage as the online behavior data of the user.

From this, under the condition that the user triggered collector under the line, can be according to the line downlink data of user for confirming its identity information, then through identity information's correlation, further acquire this user's online behavior data, then acquire the line of customer simultaneously through once gathering and be data and online behavior data, promote collection efficiency.

In addition, in this embodiment, a method for acquiring offline behavior data of a user according to the online behavior data of the user is further provided, as shown in fig. 6, the method includes:

601: and acquiring online behavior data and identity information of the user.

As described above, the identity information may be a predefined multi-dimensional uniform ID, and thus, the above scenario example of physical commodity recommendation is followed, in which an online scenario may be regarded as accessing an online network store for product purchase. Therefore, the collected identity information can be the system account of the electronic ID of the user in the multi-dimensional unified ID. Based on this, the present embodiment provides a method for acquiring online behavior data and identity information of a user, as shown in fig. 7, the method includes:

701: and receiving the page number of the target webpage browsed by the user, the browsing duration of the target webpage, the operation data in the target webpage and the login information from an online trigger arranged on the target website.

For example, taking the physical product recommendation scenario as an example, the target website may be an online web store, and when a user accesses the web store, the trigger may be triggered to start recording the number of pages of the target web page browsed by the user, the browsing duration of the target web page, the operation data in the target web page, and the login information.

702: and taking the page number of the target webpage browsed by the user, the browsing time of the target webpage and the operation data in the target webpage as the online behavior data of the user.

703: and determining the identity information of the user according to the login information of the user.

In an alternative embodiment, the identity information may also be a face of a biometric ID in a multi-dimensional unified ID. Based on this, in an optional embodiment, in step 701, the device information for browsing the target webpage may also be obtained, so that according to the device information, a camera of the device is called to obtain the facial data of the user; and then, according to the face data, carrying out face recognition on the user to determine the identity information of the user.

602: and determining the network account information and the equipment information associated with the user according to the identity information.

In this embodiment, the method for determining the network account information and the device information associated with the user according to the identity information is similar to the method for determining the network account information and the device information associated with the user according to the identity information in step 402, and is not described herein again.

603: and determining the activity participation information of the user according to the network account information.

In this embodiment, the activity participation information of the user may be various types of information participating in offline activities, which is sent by the user through the associated network account information. For example, the information about participation in offline activities sent by the network account in a preset time period may be obtained starting from the current time point. Specifically, assuming that the current time is 2021 year, 5 month and 27 day, information on participation in offline activities issued by the network account in the previous month, i.e., from 2021 year, 4 month and 27 day to 2021 year, 5 month and 27 day, can be acquired. Furthermore, the acquired information participating in the offline activity can be subjected to data screening, and the information participating in the offline activity related to shopping is screened out and used as the activity participation information of the user.

604: and determining the number of places where the equipment corresponding to the equipment information reaches the target place and the stay time at the target place according to the equipment information.

605: and taking the activity participation information of the user, the number of places where the equipment corresponding to the equipment information arrives at the target place and the stay time at the target place as the downlink data of the user.

Therefore, under the condition that the user triggers the on-line trigger, the identity information of the user can be determined according to the on-line data of the user, then the off-line data of the user is further acquired through the association of the identity information, then the off-line data and the on-line behavior data of the user are acquired simultaneously through one-time acquisition, and the acquisition efficiency is improved.

302: and performing feature extraction on the online behavior data to obtain a first feature vector.

In this embodiment, keyword extraction may be performed on the online data, and then word embedding may be performed on each extracted keyword, so as to obtain a corresponding word vector. Finally, the word vectors are spliced to obtain a first feature vector of the data on the line.

303: and performing feature extraction on the offline behavior data to obtain a second feature vector.

In this embodiment, the method for extracting the features of the data in the downlink line to obtain the second feature vector is similar to the method for extracting the features of the data in the uplink line to obtain the first feature vector in step 302, and is not described herein again.

304: and fusing the first feature vector and the second feature vector to obtain a fused feature vector.

In this embodiment, the corresponding weight may be determined according to a specific gravity between the online behavior and the offline behavior of the user, and then the first feature vector and the second feature vector are subjected to weighted summation according to the weight to obtain the fused feature vector.

305: and establishing a behavior portrait of the user according to the fusion feature vector.

In the embodiment, an intelligent analysis model can be constructed according to a neural network framework, a large number of training sets are used for tuning the intelligent analysis model, and the intelligent analysis model with better performance is established. The intelligent analysis model can be intelligently adapted to input massive different types of behavior data, and accurately and automatically output the behavior portrait of a user through model calculation.

306: and determining preference information and a behavior field of the user according to the behavior portrait.

In this embodiment, the behavior field is used to label a field to which the offline behavior data and the behavior corresponding to the online behavior data of the user belong. Illustratively, the preference information of the user may be expressed in the form of a two-dimensional table. The behavior of the user is recorded, meanwhile, the behavior tag of the user is defined according to the behavior of the user, and the behavior field of the user is determined.

Specifically, user a travels to the badminton stadium for two hours today, during which training courses are consulted and sports leads are purchased. Thus, from the identity ID, the user's behavior data, the user's behavior tag is defined: user ID, badminton racquet, badminton, sports apparel, badminton training, sports drinks; the behavioral fields are: sports shuttlecocks.

Meanwhile, in the present embodiment, for each behavior tag, different preference values may be given according to a specific behavior, for example: use 1 point, buy 2 points, share 1 point to others. Based on this, the preference information of the user may be as shown in table 1:

TABLE 1

User ID	Badminton racket	Sports clothes	Badminton training	Sports beverage
					1001	3	2	3	2

307: and determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

In the embodiment, according to the preference information of the user, one type of product can be determined as an alternative product, and then the alternative product is screened through the action field, so that a final product is obtained and recommended to the user.

In an optional implementation manner, after the preference information and the behavior field of the user are obtained, matching can be performed in the same behavior field according to the behavior field of the user. For example, the user preference information may be converted into a feature vector, a similarity between the feature vector of the user preference information and feature vectors of preference information of other users in the same behavior field is calculated, and the other users whose similarities exceed a threshold value are used as a matching result.

Based on the above, the cataloged similarity of the mutually matched users in a certain action field is high, so that the commodities purchased by the mutually matched users can be compared, and then the commodity purchased by one of the users is recommended to other matched users. Illustratively, it is collected that a user A logs in an electronic mall through an electronic account and purchases swimming equipment on line. Meanwhile, the user A and the user B are mutually matched users in the badminton field in the motion behavior field, the user A has a record of purchasing commodities by swimming equipment, the user B does not have the record, and the swimming commodities purchased by the user A can be recommended to the user B based on user collaborative filtering, so that the recommendation efficiency is improved.

In summary, in the product recommendation method provided by the present invention, the offline behavior data and the online behavior data of the user are obtained, and then the offline behavior data and the online behavior data are respectively subjected to feature extraction and fusion, so as to obtain the fusion feature fusing the offline behavior and the online behavior of the user. And then constructing a behavior portrait of the user according to the fusion characteristics, then determining preference information and a behavior field of the user, and finally determining a corresponding product to be recommended to the user according to the preference information and the behavior field. Therefore, online data and offline data are fused, so that behavior portrait of a client is more accurate and comprehensive, meanwhile, the selection range of products is further narrowed by combining the behavior field, the products are limited to products in the user behavior related field, the recommendation accuracy is improved, and the user experience is improved.

Referring to fig. 8, fig. 8 is a block diagram illustrating functional modules of a product recommendation device according to an embodiment of the present disclosure. As shown in fig. 8, the product recommendation device 800 includes:

a data obtaining module 801, configured to obtain data of a user in a downlink and data of a user in an uplink;

the feature extraction module 802 is configured to perform feature extraction on the online behavior data to obtain a first feature vector; performing feature extraction on the offline behavior data to obtain a second feature vector;

an image creating module 803, configured to fuse the first feature vector and the second feature vector to obtain a fused feature vector; establishing a behavior portrait of the user according to the fusion feature vector;

the product recommendation module 804 is used for determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the fields of the behaviors corresponding to the offline behavior data and the online behavior data of the user; and determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

In the embodiment of the present invention, in acquiring data of a user in a downlink line and data of an uplink line, the data acquiring module 801 is specifically configured to:

acquiring downlink data and identity information of a user;

determining network account information and equipment information associated with the user according to the identity information;

determining social dynamic information of a user according to the network account information;

determining the page number of a target webpage browsed by equipment corresponding to the equipment information and the browsing duration of the target webpage according to the equipment information;

and taking the social dynamic information of the user, the page number of the target webpage browsed by the equipment corresponding to the equipment information and the browsing duration of the target webpage as the online behavior data of the user.

In the embodiment of the present invention, in terms of acquiring data and identity information of a downlink of a user, the data acquiring module 801 is specifically configured to:

receiving behavior images of a user from an offline acquisition device arranged at a target location;

analyzing the behavior image frame by frame to obtain at least one sub-action image of the user;

for each sub-action image in at least one sub-action image, respectively carrying out feature extraction on each sub-action image to obtain at least one sub-action feature, wherein the at least one sub-action feature is in one-to-one correspondence with the at least one sub-action image;

determining the line descending of the user as data according to at least one sub-action characteristic;

and according to the behavior image, carrying out face recognition on the user and determining the identity information of the user.

In an embodiment of the present invention, in determining the downlink of the user as data according to at least one sub-action feature, the data obtaining module 801 is specifically configured to:

arranging at least one sub-action feature according to a first sequence to obtain a sub-action sequence, wherein the first sequence is the sequence of sub-action images corresponding to each sub-action feature in at least one sub-action feature appearing in the action image;

and matching the sub-action sequence with a preset behavior sequence, and determining the line descending of the user as data.

acquiring online behavior data and identity information of a user;

determining activity participation information of a user according to the network account information;

according to the equipment information, determining the number of places where the equipment corresponding to the equipment information reaches the target place and the stay time of the equipment at the target place;

and taking the activity participation information of the user, the number of places where the equipment corresponding to the equipment information arrives at the target place and the stay time at the target place as the downlink data of the user.

In an embodiment of the present invention, in terms of acquiring online behavior data and identity information of a user, the data acquiring module 801 is specifically configured to:

receiving the page number of a target webpage browsed by a user, the browsing duration of the target webpage, operation data in the target webpage and login information from an online trigger arranged on the target website;

taking the page number of a target webpage browsed by a user, the browsing duration of the target webpage and operation data in the target webpage as online behavior data of the user;

and determining the identity information of the user according to the login information of the user.

receiving the page number of a target webpage browsed by a user, the browsing duration of the target webpage, operation data in the target webpage and equipment information of the target webpage from an online trigger arranged on the target website;

according to the equipment information, calling a camera of the equipment to acquire the facial data of the user;

and according to the face data, carrying out face recognition on the user and determining the identity information of the user.

Referring to fig. 9, fig. 9 is a schematic structural diagram of an electronic device provided in the embodiment of the present application, where the electronic device 900 is disposed in a first tenant system. As shown in fig. 9, the electronic device 900 includes a transceiver 901, a processor 902, and a memory 903. Connected to each other by a bus 904. The memory 903 is used to store computer programs and data, and may transfer the data stored in the memory 903 to the processor 902.

The processor 902 is configured to read the computer program in the memory 903 to perform the following operations:

acquiring downlink data and uplink behavior data of a user;

In an embodiment of the present invention, in obtaining the offline behavior data and the online behavior data of the user, the processor 902 is specifically configured to perform the following operations:

acquiring downlink data and identity information of a user;

In an embodiment of the present invention, in terms of acquiring data and identity information of a downlink of a user, the processor 902 is specifically configured to perform the following operations:

In an embodiment of the present invention, in determining the downlink of the user as data according to at least one sub-action feature, the processor 902 is specifically configured to perform the following operations:

acquiring online behavior data and identity information of a user;

In an embodiment of the present invention, in obtaining the online behavior data and the identity information of the user, the processor 902 is specifically configured to perform the following operations:

It should be understood that the product recommendation device in the present application may include a smart Phone (e.g., an Android Phone, an iOS Phone, a Windows Phone, etc.), a tablet computer, a palm computer, a notebook computer, a Mobile Internet device MID (MID), a robot, a wearable device, etc. The product recommendation device is merely an example, and is not exhaustive, and includes but is not limited to the product recommendation device. In practical applications, the product recommendation device may further include: intelligent vehicle-mounted terminal, computer equipment and the like.

Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented by combining software and a hardware platform. With this understanding in mind, all or part of the technical solutions of the present invention that contribute to the background can be embodied in the form of a software product, which can be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods according to the embodiments or some parts of the embodiments.

Accordingly, the present application also provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement part or all of the steps of any one of the product recommendation methods as described in the above method embodiments. For example, the storage medium may include a hard disk, a floppy disk, an optical disk, a magnetic tape, a magnetic disk, a flash memory, and the like.

Embodiments of the present application also provide a computer program product comprising a non-transitory computer readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the product recommendation methods as described in the above method embodiments.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are all alternative embodiments and that the acts and modules referred to are not necessarily required by the application.

In the above embodiments, the description of each embodiment has its own emphasis, and for parts not described in detail in a certain embodiment, reference may be made to the description of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is merely a logical division, and other divisions may be realized in practice, for example, a plurality of units 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 of some interfaces, devices or units, and may be an electric or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The integrated unit may be implemented in the form of hardware, or may be implemented in the form of a software program module.

The integrated units, if implemented in the form of software program modules and sold or used as stand-alone products, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be substantially implemented or a part of or all or part of the technical solution contributing to the prior art may be embodied in the form of a software product stored in a memory, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned memory comprises: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable memory, and the memory may include: flash Memory disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

The foregoing detailed description of the embodiments of the present application has been presented to illustrate the principles and implementations of the present application, and the above description of the embodiments is only provided to help understand the methods and their core ideas of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A method for recommending products, the method comprising:

acquiring downlink data and uplink behavior data of a user;

performing feature extraction on the downlink data to obtain a second feature vector;

determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the field to which the behavior corresponding to the offline behavior data and the online behavior data of the user belongs;

determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

2. The method of claim 1, wherein the obtaining downlink behavior data and uplink behavior data of the user comprises:

acquiring downlink data and identity information of the user;

determining the network account information and the equipment information associated with the user according to the identity information;

determining social dynamic information of the user according to the network account information;

3. The method of claim 2, wherein the obtaining downlink of the user as data and identity information comprises:

receiving behavior images of the user from an offline acquisition device arranged at a target location;

for each sub-action image in the at least one sub-action image, respectively performing feature extraction on each sub-action image to obtain at least one sub-action feature, wherein the at least one sub-action feature is in one-to-one correspondence with the at least one sub-action image;

determining the line descending of the user as data according to the at least one sub-action characteristic;

and according to the behavior image, carrying out face recognition on the user, and determining the identity information of the user.

4. The method of claim 3, wherein the determining the line descending of the user as data according to the at least one sub-action feature comprises:

arranging the at least one sub-action feature according to a first sequence to obtain a sub-action sequence, wherein the first sequence is the sequence of sub-action images corresponding to each sub-action feature in the at least one sub-action feature appearing in the behavior image;

and matching the sub-action sequence with a preset behavior sequence, and determining the downlink of the user as data.

5. The method of claim 1, wherein the obtaining downlink behavior data and uplink behavior data of the user comprises:

acquiring online behavior data and identity information of the user;

determining activity participation information of the user according to the network account information;

according to the equipment information, determining the number of places where the equipment corresponding to the equipment information reaches a target place and the stay time of the equipment at the target place;

and taking the activity participation information of the user, the number of places where the equipment corresponding to the equipment information reaches the target place and the stay time at the target place as the downlink data of the user.

6. The method of claim 5, wherein the obtaining online behavior data and identity information of the user comprises:

receiving the page number of the target webpage browsed by the user, the browsing duration of the target webpage, operation data in the target webpage and login information from an online trigger arranged on a target website;

taking the number of pages of the target webpage browsed by the user, the browsing duration of the target webpage and the operation data in the target webpage as the online behavior data of the user;

7. The method of claim 5, wherein the obtaining online behavior data and identity information of the user comprises:

receiving the page number of the target webpage browsed by the user, the browsing duration of the target webpage, operation data in the target webpage and equipment information for browsing the target webpage from an online trigger arranged on a target website;

calling a camera of the equipment according to the equipment information to acquire the facial data of the user;

and according to the facial data, carrying out face recognition on the user and determining the identity information of the user.

8. A product recommendation device, the device comprising:

the characteristic extraction module is used for extracting characteristics of the online behavior data to obtain a first characteristic vector; performing feature extraction on the downlink data to obtain a second feature vector;

the product recommendation module is used for determining preference information and a behavior field of the user according to the behavior portrait, wherein the behavior field is used for marking the fields of the behaviors of the user corresponding to the offline behavior data and the online behavior data; determining products according to the preference information and the behavior field of the user, and recommending the products to the user.

9. An electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, the one or more programs including instructions for performing the steps in the method of any of claims 1-7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which is executed by a processor to implement the method according to any one of claims 1-7.