A product recommendation method, system and storage medium applied to the private domain and e-commerce platforms

Abstract

The present application provides a product recommendation method, system and storage medium applicable to the private domain and e-commerce platforms. By respectively obtaining the behavior data of a target user on the private domain platform and the public domain platform, a user behavior analysis model is established based on the behavior data of the target user on the private domain platform and the public domain platform to customize the tagging of the user and construct a user portrait. The tagging results of the products are correlated and matched with the user tags in the user portrait, and a connection channel between the user behavior and the corresponding products is established according to the matching situation between the products and the user. Different connection channels are opened for the target users in different scenarios. The present application optimizes the processing and matching mode of product and customer data in the private domain e-commerce scenario, improves the accuracy of product recommendation and the data processing efficiency, and effectively improves the user experience.

Description

Technical Field

[0001] This application belongs to the technical field of big data processing. More specifically, it relates to a product recommendation method, system, and storage medium applied to private domains and e-commerce platforms. Background Art

[0002] Currently, the existing private domain e-commerce selects products from the supply chain and the recommendation technology is based on the data processing mode of its own platform. This results in relatively low user perception and service efficiency, unable to well meet the service needs of merchants and customers, and also unable to reflect better human efficiency. Therefore, there is an urgent need to propose a big data processing technology for private domains and e-commerce platforms to solve the problems existing in the process of product introduction, product selection, and customer complete data collection, integration, and recommendation on the supply chain side. Summary of the Invention

[0003] In view of this, this application provides a product recommendation method, system, and storage medium applied to private domains and e-commerce platforms, which optimizes the product and customer data processing and matching modes in the private domain e-commerce scenario, improves the accuracy of product recommendation and data processing efficiency, and effectively improves the user experience.

[0004] The specific technical solutions of this application are as follows:

[0005] The first aspect of this application provides a product recommendation method applied to private domains and e-commerce platforms, which is characterized by including the following steps:

[0006] Obtain the behavior data of the target user on the private domain platform, perform data embedding on the behavior data and send the data to the processing center in the form of a script, obtain the behavior data of the target user on the public domain platform and send it to the processing center;

[0007] Read the behavior data in the processing center queue and transfer and store the data, and then store it in the open data processing data warehouse through the method of writing metadata;

[0008] Establish a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform, and customize the target user according to the user behavior analysis model to build a user portrait;

[0009] Obtain product data for aggregation analysis and perform personalized tagging on it according to product characteristic parameters, and perform relevance matching between the tagging results of the products and the user tags in the user portrait;

[0010] Establish a connection channel between user behavior and corresponding products according to the matching situation between products and users, open different connection channels for target users in different scenarios, and display the corresponding products in the form of a recommendation list.

[0011] Preferably, the process of embedding points for the behavior data and sending the data to the processing center in the form of a script is as follows:

[0012] After receiving the click, favorite, forward, or order signal from the target user, embed points for the generated behavior data respectively;

[0013] Use the lightweight transfer mode to collect the behavior data with embedded points into the data transmission channel, and then the data transmission channel asynchronously sends the data log to the processing center through the script.

[0014] Preferably, the process of obtaining the behavior data of the target user on the public domain platform and sending it to the processing center is as follows:

[0015] Obtain the public information of the target user on the private domain platform and synchronize it to the ERP and third-party systems, and obtain the behavior data of the target user on the public domain platform through the ERP and third-party systems;

[0016] Use the ERP and third-party systems to store and send the behavior data of the target user on the public domain platform to the processing center.

[0017] Preferably, the process of establishing a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform is as follows:

[0018] Statistically analyze the store information and product information of the target user's behavior data under each behavior node. The behavior nodes include search, screening, favorite, price comparison, and commission operation;

[0019] Extract the static information of the target user, and perform basic tagging on the user behavior according to the static information, store information, and product information of the target user under each behavior node.

[0020] Preferably, the user behavior analysis model includes:

[0021] Infer the dynamic usage habits of the user based on the behavior nodes and occurrence frequencies of the user, so as to generate scenario tags;

[0022] Collect location data and price data according to the static information and product information of the user, so as to generate attribute tags;

[0023] Generate preference tags according to the behavior data, store information, and product information of the user.

[0024] Preferably, the process of customizing tags for the user according to the user behavior analysis model to construct a user portrait is as follows:

[0025] Monitor and identify the special requirement information of the target user, and clean the behavior data of the target user according to the special requirement information;

[0026] Based on the cleaned behavior data, use the user behavior analysis model to generate customized tags for the behavior data of the target user.

[0027] Preferably, obtaining commodity data for aggregation analysis and performing personalized tagging according to commodity characteristic parameters specifically includes:

[0028] Retrieve the basic commodity data of the private domain platform and the popular commodity data of the public domain platform, classify the commodity data according to the data type and value, and enter classification tags for the commodity data;

[0029] Monitor whether there is ambiguity in the classification tags of the same commodity from the private domain platform and the public domain platform. If so, perform denoising processing on the tagging results of the commodity according to the customary markings of the commodity.

[0030] Preferably, obtaining commodity data for aggregation analysis and performing personalized tagging according to commodity characteristic parameters further includes:

[0031] Enter the SPU basic data of the commodity according to the commodity characteristic parameters, and generate the SPU tag of the commodity according to the SPU basic data;

[0032] Extract the supply chain information of the commodity, and count the node data of the supply chain to generate the supply chain tag of the commodity;

[0033] Perform basic tagging on the commodity data according to the SPU tag and supply chain information of the commodity.

[0034] The second aspect of this application provides a commodity recommendation system applied to private domain and e-commerce platforms, including a memory and a processor. The memory includes a commodity recommendation program applied to private domain and e-commerce platforms. When the commodity recommendation program applied to private domain and e-commerce platforms is executed by the processor, the following steps are implemented:

[0035] Obtain the behavior data of the target user on the private domain platform, perform data embedding on the behavior data and send the data to the processing center in the form of a script, obtain the behavior data of the target user on the public domain platform and send it to the processing center;

[0036] Read the behavior data in the processing center queue and transfer and store the data, and then store it in the open data processing data warehouse through the method of writing metadata;

[0037] Establish a user behavior analysis model according to the behavior data of the target user on the private domain platform and the public domain platform, and perform customized tagging on the user according to the user behavior analysis model to construct a user portrait;

[0038] Obtain commodity data for aggregation analysis and perform personalized tagging according to commodity characteristic parameters, and perform correlation matching between the tagging results of the commodity and the user tags in the user portrait;

[0039] Establish a connection channel between user behavior and corresponding products according to the matching situation between products and users, open different connection channels for target users in different scenarios, and display the corresponding products in the form of a recommended list.

[0040] Preferably, the behavior data is buried and sent to the processing center in the form of a script, specifically as follows:

[0041] After receiving the click, favorite, forward or order signal from the target user respectively, bury the generated behavior data;

[0042] Use the lightweight transfer mode to collect the buried behavior data to the data transmission channel, and then the data transmission channel asynchronously sends the data log to the processing center through the script.

[0043] Preferably, obtaining the behavior data of the target user on the public domain platform and sending it to the processing center is specifically as follows:

[0044] Obtain the public information of the target user on the private domain platform and synchronize it to the ERP and third-party systems, and obtain the behavior data of the target user on the public domain platform through the ERP and third-party systems;

[0045] Use the ERP and third-party systems to store and send the behavior data of the target user on the public domain platform to the processing center.

[0046] Preferably, establishing a user behavior analysis model according to the behavior data of the target user on the private domain platform and the public domain platform is specifically as follows:

[0047] Statistically analyze the store information and product information of the target user's behavior data under each behavior node. The behavior nodes include search, screening, favorite, price comparison and commission operations;

[0048] Extract the static information of the target user, and conduct basic tagging of user behavior according to the static information, store information and product information of the target user under each behavior node.

[0049] Preferably, the user behavior analysis model includes:

[0050] Infer the dynamic usage habits of users based on the behavior nodes and occurrence frequencies of users, so as to generate scenario tags;

[0051] Collect location data and price data according to the static information and product information of users, so as to generate attribute tags;

[0052] Generate preference tags according to the behavior data, store information and product information of users.

[0053] Preferably, customizing tags for users according to the user behavior analysis model to construct a user portrait is specifically as follows:

[0054] Monitor and identify the special requirement information of the target user, and clean the behavior data of the target user according to the special requirement information;

[0055] Based on the cleaned behavior data, use the user behavior analysis model to generate customized labels for the behavior data of the target user.

[0056] Preferably, obtaining commodity data for aggregation analysis and performing personalized labeling according to commodity characteristic parameters specifically includes:

[0057] Retrieve the basic commodity data of the private domain platform and the popular commodity data of the public domain platform, classify the commodity data according to the data type and value, and enter classification labels for the commodity data;

[0058] Monitor whether there is ambiguity in the classification labels of the same commodity from the private domain platform and the public domain platform. If so, perform denoising processing on the labeling result of the commodity according to the customary label of the commodity.

[0059] Preferably, obtaining commodity data for aggregation analysis and performing personalized labeling according to commodity characteristic parameters further includes:

[0060] Enter the SPU basic data of the commodity according to the commodity characteristic parameters, and generate the SPU label of the commodity according to the SPU basic data;

[0061] Extract the supply chain information of the commodity, and count the node data of the supply chain to generate the supply chain label of the commodity;

[0062] Perform basic labeling on the commodity data according to the SPU label and supply chain information of the commodity.

[0063] The third aspect of the present application provides a computer-readable storage medium, which includes a commodity recommendation program applied to the private domain and e-commerce platforms. When the commodity recommendation program applied to the private domain and e-commerce platforms is executed by a processor, the steps of the commodity recommendation method applied to the private domain and e-commerce platforms are implemented.

[0064] In summary, the present application provides a product recommendation method, system, and storage medium applicable to the private domain and e-commerce platforms. By separately obtaining the behavioral data of the target user on the private domain platform and the public domain platform, a user behavior analysis model is established based on the behavioral data of the target user on the private domain platform and the public domain platform to customize label the user and construct a user portrait. The labeling results of the products are correlated and matched with the user labels in the user portrait, and a connection channel between the user behavior and the corresponding products is established according to the matching situation between the products and the user. Different connection channels are opened for the target users in different scenarios. The present application optimizes the processing and matching mode of product and customer data in the private domain e-commerce scenario, improves the accuracy of product recommendation and the data processing efficiency, and effectively improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0066] Figure 1 It is a flowchart of a product recommendation method applicable to the private domain and e-commerce platforms of the present application;

[0067] Figure 2 It is a block diagram of a product recommendation system applicable to the private domain and e-commerce platforms of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] In order to make the objectives, features, and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application are clearly and completely described below. Obviously, the embodiments described below are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0069] Please refer to Figure 1 , Figure 1 It is a flowchart of a product recommendation method applicable to the private domain and e-commerce platforms of the present application.

[0070] The first aspect of the embodiments of the present application provides a product recommendation method applicable to the private domain and e-commerce platforms, including the following steps:

[0071] S102: Obtain the behavioral data of the target user on the private domain platform, perform data embedding on the behavioral data and send the data to the processing center in the form of a script, obtain the behavioral data of the target user on the public domain platform and send it to the processing center;

[0072] S104: Read the behavior data in the processing center queue, transfer and store the data, and then store it in the open data processing data warehouse by means of writing metadata.

[0073] S106: Establish a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform, and customize the tagging of the user according to the user behavior analysis model to construct a user portrait.

[0074] S108: Obtain commodity data for aggregation analysis, perform personalized tagging on it according to commodity characteristic parameters, and perform correlation matching between the tagging results of the commodity and the user tags in the user portrait.

[0075] S110: Establish a connection channel between the user behavior and the corresponding commodity according to the matching situation between the commodity and the user, open different connection channels for the target users in different scenarios, and display the corresponding commodities in the form of a recommended list.

[0076] It should be noted that the private domain platforms in S102 include APP terminals, applets, official accounts, etc. The behavior data can be the operation data generated by the target user under the platform operation nodes. The behavior data of the public domain platform is generally the operation data containing non-private information of the user. In S104, the data in the queue can be read by the datacenter data center service and stored in the odps data warehouse. In S106, the user portrait can reflect the behavior preferences and potential behavior purposes of the target user, and can be used for accurate matching according to the user needs. In S108, according to the characteristic information such as the price, source and use of the commodity, the commodity can be tagged and associated with the user tag to match suitable commodity recommendations for the user. In S110, the application scenario can be set to meet the special needs of customers and promote products more pertinently.

[0077] According to the embodiments of the present application, specifically, the behavior data is buried and the data is sent to the processing center in the form of a script as follows:

[0078] Respectively perform data burying on the generated behavior data after receiving the click, favorite, forward or order signal from the target user.

[0079] Collect the buried behavior data to the data transmission channel by using the lightweight transmission mode, and then the data transmission channel asynchronously sends the data log to the processing center through the script.

[0080] It should be noted that the buried point data mainly includes enterprises, stores, users, commodities, behavior tags, etc. In the embodiments of the present application, filebeat can collect the buried point data read from multiple servers into logstash, and logstash asynchronously sends the log to kafka through the script.

[0081] According to the embodiments of the present application, obtaining the behavior data of the target user on the public domain platform and sending it to the processing center specifically includes:

[0082] Obtaining the public information of the target user on the private domain platform and synchronizing it to the ERP and third-party systems, and obtaining the behavior data of the target user on the public domain platform through the ERP and third-party systems;

[0083] Using the ERP and third-party systems to store the behavior data of the target user on the public domain platform and send it to the processing center.

[0084] According to the embodiments of the present application, establishing a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform specifically includes:

[0085] Counting the store information and product information of the target user's behavior data under each behavior node, and the behavior nodes include search, screening, collection, price comparison, and commission operation;

[0086] Extracting the static information of the target user, and performing basic labeling of the user behavior according to the static information, store information, and product information of the target user under each behavior node.

[0087] According to the embodiments of the present application, the user behavior analysis model includes:

[0088] Inferring the dynamic usage habits of the user based on the behavior nodes and occurrence frequencies of the user, so as to generate scenario tags;

[0089] Collecting location data and price data according to the static information and product information of the user, so as to generate attribute tags;

[0090] Generating preference tags according to the behavior data, store information, and product information of the user.

[0091] It should be noted that the scenario tags include apps, mini-programs, official accounts, and custom platforms, the attribute tags include gender, age, consumption level, occupation, etc., and the preference tags include women's clothing, mother and baby, children, makeup, virtual goods, etc.

[0092] According to the embodiments of the present application, customizing the labeling of the user according to the user behavior analysis model to construct a user portrait specifically includes:

[0093] Monitoring and identifying the special requirement information of the target user, and cleaning the behavior data of the target user according to the special requirement information;

[0094] Based on the cleaned behavior data, using the user behavior analysis model to generate customized tags for the behavior data of the target user.

[0095] It should be noted that the special requirement information can be custom filtering conditions generated through user settings. The user tags generated by cleaning data according to the user's custom setting parameters are more authentic.

[0096] According to the embodiments of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters specifically includes:

[0097] Retrieve the basic commodity data of the private domain platform and the popular commodity data of the public domain platform, classify the commodity data according to the data type and value, and enter classification labels for the commodity data;

[0098] Monitor whether there is ambiguity in the classification labels of the same commodity from the private domain platform and the public domain platform. If so, perform denoising processing on the labeling results of the commodity according to the conventional markings of the commodity.

[0099] It should be noted that the basic commodity data is mainly the characteristic parameters of the commodity, such as size, brand, specification and other information. Popular commodities refer to commodities within the public domain platform whose click volume or sales volume is above a preset range. The sources of these data are extensive and highly referenceable. Classification labels include women's clothing, beauty products, personal care, food, children's clothing, furniture, accessories, mother and baby, etc.

[0100] According to the embodiments of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters further includes:

[0101] Enter the SPU basic data of the commodity according to the commodity characteristic parameters, and generate the SPU label of the commodity according to the SPU basic data;

[0102] Extract the supply chain information of the commodity, and count the node data of the supply chain to generate the supply chain label of the commodity;

[0103] Perform basic labeling on the commodity data according to the SPU label and supply chain information of the commodity.

[0104] It should be noted that the SPU basic data refers to real-time information such as price range division, sales volume, commission, discount, activities, etc. The supply chain information refers to supplier, manufacturer, transporter, retailer and customer information.

[0105] In another embodiment of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters specifically includes:

[0106] Obtain the after-sales information of the target user, where the after-sales information includes logistics information, maintenance information, guidance information and feedback information, and screen the commodity description characteristics in the after-sales information;

[0107] Perform data correction on the basic labeling result of the commodity data according to the commodity description characteristics to generate the personalized labeling result of the commodity.

[0108] In another embodiment of the present application, enabling different connection channels for target users in different scenarios specifically includes:

[0109] Obtain the scenario requirement information of the target user, and clean the user's behavior data according to the scenario requirement information of the user;

[0110] Analyze the hot data of the commodity in this special scenario, store the cleaned commodity data and the corresponding hot data in the database and perform persistent processing, then synchronously store the cleaned behavior data and commodity data in the open data processing data warehouse, and finally asynchronously store the cleaned user behavior data and commodity data in the es database.

[0111] It should be noted that in the embodiment of the present application, the cleaned commodity data and the corresponding hot data can be stored in redis and perform persistent processing, then synchronously store the cleaned behavior data and commodity data in odps, and finally asynchronously store the cleaned user behavior data and commodity data in es, realizing the matching of personalized, high-quality, popular, and explosive commodities for users in different scenarios.

[0112] Please refer to Figure 2 , Figure 2 which is a block diagram of a commodity recommendation system applied to the private domain and e-commerce platform of the present application.

[0113] The second aspect of the embodiment of the present application provides a commodity recommendation system applied to the private domain and e-commerce platform, including a memory 21 and a processor 22. The memory 21 includes a commodity recommendation program applied to the private domain and e-commerce platform. When the commodity recommendation program applied to the private domain and e-commerce platform is executed by the processor 22, the following steps are implemented:

[0114] Obtain the behavior data of the target user on the private domain platform, perform data embedding on the behavior data and send the data to the processing center in the form of a script, obtain the behavior data of the target user on the public domain platform and send it to the processing center;

[0115] Read the behavior data in the processing center queue and transfer and store the data, and then store it in the open data processing data warehouse through the method of writing metadata;

[0116] Establish a user behavior analysis model according to the behavior data of the target user on the private domain platform and the public domain platform, and customize the target user according to the user behavior analysis model to construct a user portrait;

[0117] Obtain commodity data for aggregation analysis and perform personalized marking on it according to commodity characteristic parameters, and perform correlation matching between the marking results of the commodities and the user tags in the user portrait;

[0118] Establish a connection channel between user behavior and corresponding products according to the matching situation between products and users, open different connection channels for target users in different scenarios, and display the corresponding products in the form of a recommended list.

[0119] According to the embodiments of the present application, specifically, the behavior data is buried and sent to the processing center in the form of a script as follows:

[0120] Respectively, after receiving signals of click, collection, forwarding or ordering from the target user, bury the generated behavior data;

[0121] Collect the buried behavior data to the data transmission channel using a lightweight transmission mode, and then the data transmission channel asynchronously sends the data log to the processing center through a script.

[0122] According to the embodiments of the present application, specifically, obtaining the behavior data of the target user on the public domain platform and sending it to the processing center is as follows:

[0123] Obtain the public information of the target user on the private domain platform and synchronize it to the ERP and third-party systems, and obtain the behavior data of the target user on the public domain platform through the ERP and third-party systems;

[0124] Use the ERP and third-party systems to store and send the behavior data of the target user on the public domain platform to the processing center.

[0125] According to the embodiments of the present application, specifically, establishing a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform is as follows:

[0126] Statistically analyze the store information and product information of the target user's behavior data at each behavior node. The behavior nodes include search, screening, collection, price comparison, and commission operations;

[0127] Extract the static information of the target user, and perform basic tagging on the user behavior according to the static information, store information, and product information of the target user at each behavior node.

[0128] According to the embodiments of the present application, the user behavior analysis model includes:

[0129] Infer the dynamic usage habits of the user based on the user's behavior nodes and occurrence frequencies, so as to generate scenario tags;

[0130] Collect location data and price data according to the user's static information and product information, so as to generate attribute tags;

[0131] Generate preference tags according to the user's behavior data, store information, and product information.

[0132] According to the embodiments of the present application, customizing and labeling users based on the user behavior analysis model to construct user portraits specifically includes:

[0133] Monitoring and identifying the special requirement information of the target user, and cleaning the behavior data of the target user according to the special requirement information;

[0134] Based on the cleaned behavior data, using the user behavior analysis model to generate customized labels for the behavior data of the target user.

[0135] According to the embodiments of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters specifically includes:

[0136] Retrieving the basic commodity data of the private domain platform and the popular commodity data of the public domain platform, classifying the commodity data according to the data type and value, and entering classification labels for the commodity data;

[0137] Monitoring whether there is ambiguity in the classification labels of the same commodity from the private domain platform and the public domain platform. If so, denoising the labeling result of the commodity according to the customary label of the commodity.

[0138] According to the embodiments of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters further includes:

[0139] Entering the SPU basic data of the commodity according to the commodity characteristic parameters, and generating the SPU label of the commodity according to the SPU basic data;

[0140] Extracting the supply chain information of the commodity, and statistically generating the supply chain label of the commodity based on the node data of the supply chain;

[0141] Performing basic labeling on the commodity data according to the SPU label and supply chain information of the commodity.

[0142] In another embodiment of the present application, obtaining commodity data for aggregation analysis and performing personalized labeling on it according to commodity characteristic parameters specifically includes:

[0143] Obtaining the after-sales information of the target user, where the after-sales information includes logistics information, maintenance information, guidance information, and feedback information, and screening the commodity description features in the after-sales information;

[0144] According to the commodity description features, correcting the data of the basic labeling result of the commodity data to generate the personalized labeling result of the commodity.

[0145] In another embodiment of the present application, opening different connection channels for target users in different scenarios specifically includes:

[0146] Obtaining the scenario requirement information of the target user, and cleaning the behavior data of the user according to the scenario requirement information of the user;

[0147] Analyze the hot data of the product in this special scenario, store the cleaned product data and the corresponding hot data in the database and perform persistent processing, then synchronously store the cleaned behavior data and product data into the open data processing data warehouse, and finally asynchronously store the cleaned user behavior data and product data into the es database.

[0148] The third aspect of the embodiments of the present application provides a computer-readable storage medium, which includes a product recommendation program applied to the private domain and the e-commerce platform. When the product recommendation program applied to the private domain and the e-commerce platform is executed by a processor, the steps of the product recommendation method applied to the private domain and the e-commerce platform are implemented. For details, please refer to Figure 1 the description of the method steps, which will not be elaborated here.

[0149] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces. The indirect coupling or communication connection of the devices or units can be electrical, mechanical, or other forms.

[0150] The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] In addition, each functional unit in the embodiments of the present invention can be all integrated in a processing unit, or each unit can be separately used as a unit, or two or more units can be integrated in a unit; the above integrated unit can be implemented in the form of hardware, or in the form of a hardware plus software functional unit.

[0152] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including those of the above method embodiments. The aforementioned storage medium includes various media that can store program codes, such as removable storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0153] Alternatively, if the above integrated units are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the embodiments of the present invention, in essence, or the parts that contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program codes, such as removable storage devices, ROM, RAM, magnetic disks, or optical discs.

[0154] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A product recommendation method applied to the private domain and e-commerce platforms, characterized in that, It includes the following steps: Obtain the behavior data of the target user on the private domain platform, perform data embedding on the behavior data and send the data to the processing center in the form of a script, obtain the behavior data of the target user on the public domain platform and send it to the processing center; Read the behavior data in the processing center queue, transfer and store the data, and then store it in the open data processing data warehouse by means of metadata writing; Establish a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform, and customize label the user according to the user behavior analysis model to build a user portrait; Obtain product data for aggregation analysis and perform personalized labeling on it according to product characteristic parameters, and perform correlation matching between the labeling results of the products and the user labels in the user portrait; Establish a connection channel between user behavior and corresponding products according to the matching situation between products and users, open different connection channels for target users in different scenarios, and display the corresponding products in the form of a recommended list; Performing data embedding on the behavior data and sending the data to the processing center in the form of a script specifically means: Perform data embedding on the generated behavior data respectively after receiving signals of click, collection, forwarding or ordering from the target user; Collect the embedded behavior data to the data transmission channel using a lightweight transmission mode, and then the data transmission channel asynchronously sends the data log to the processing center through a script; Establishing a user behavior analysis model based on the behavior data of the target user on the private domain platform and the public domain platform specifically means: Statistically analyze the store information and product information of the target user's behavior data under each behavior node. The behavior nodes include search, screening, collection, price comparison and commission operations; Extract the static information of the target user, and perform basic labeling on the user behavior according to the static information, store information and product information of the target user under each behavior node; Obtain product data for aggregation analysis and perform personalized labeling on it according to product characteristic parameters specifically means: Retrieve the basic product data of the private domain platform and the popular product data of the public domain platform, classify the product data according to the data type and value, and enter classification labels for the product data; Monitor whether there is ambiguity in the classification labels of the same product from the private domain platform and the public domain platform. If so, denoise the labeling results of the product according to the common markings of the product; Opening different connection channels for target users in different scenarios specifically means: Obtain the scenario requirement information of the target user, and clean the behavior data of the user according to the scenario requirement information of the user; Analyze the hot data of the product in this special scenario, store the cleaned product data and the corresponding hot data in the database for persistent processing, then synchronously store the cleaned behavior data and product data in the open data processing data warehouse, and finally asynchronously store the cleaned user behavior data and product data in the es database.

2. The product recommendation method applied to the private domain and e-commerce platform according to claim 1, wherein Obtaining the behavior data of the target user on the public domain platform and sending it to the processing center specifically means: Obtain the public information of the target user on the private domain platform and synchronize it to the ERP and third-party systems, and obtain the behavior data of the target user on the public domain platform through the ERP and third-party systems; Use ERP and third-party systems to store and send the behavioral data of target users on the public domain platform to the processing center.

3. The product recommendation method applied to the private domain and e-commerce platform according to claim 1, wherein, The user behavior analysis model includes: Infer the dynamic usage habits of users based on the behavior nodes and occurrence frequencies of users, so as to generate scenario tags; Collect location data and price data based on the static information of users and commodity information, so as to generate attribute tags; Generate preference tags based on the behavior data, store information, and commodity information of users.

4. The method for recommending products applied to the private domain and e-commerce platforms according to claim 1, wherein Customize tagging users according to the user behavior analysis model to construct user portraits, specifically: Monitor and identify the special requirement information of target users, and clean the behavior data of target users according to the special requirement information; Based on the cleaned behavior data, use the user behavior analysis model to generate customized tags for the behavior data of target users.

5. The product recommendation method applied to the private domain and e-commerce platform according to claim 1, characterized in that, Obtain commodity data for aggregation analysis and perform personalized tagging according to the commodity characteristic parameters, and further include: Enter the SPU basic data of the commodity according to the commodity characteristic parameters, and generate the SPU tag of the commodity according to the SPU basic data; Extract the supply chain information of the commodity, and count the node data of the supply chain to generate the supply chain tag of the commodity; Perform basic tagging on the commodity data according to the SPU tag and supply chain information of the commodity.

6. A product recommendation system applied to the private domain and e-commerce platforms, characterized in that, It includes a memory and a processor. The memory includes a commodity recommendation program applied to the private domain and e-commerce platforms. When the commodity recommendation program applied to the private domain and e-commerce platforms is executed by the processor, the steps of the commodity recommendation method applied to the private domain and e-commerce platforms as described in any one of claims 1 to 5 are implemented.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a commodity recommendation program applied to the private domain and e-commerce platforms. When the commodity recommendation program applied to the private domain and e-commerce platforms is executed by the processor, the steps of the commodity recommendation method applied to the private domain and e-commerce platforms as described in any one of claims 1 to 5 are implemented.

Images