Product recommendation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a product recommendation method and device, electronic equipment and a storage medium, and belongs to the technical field of financial science and technology and artificial intelligence. The method comprises the following steps: acquiring historical behavior data of a target object, and acquiring a propaganda text template of a candidate product; constructing a user portrait of the target object based on the historical behavior data; on the basis of the user portrait, the historical behavior data and a preset text generation model, performing recommendation copywriting generation on the candidate product to obtain a product recommendation text; based on the user portrait and the historical behavior data, screening out a target propaganda template from the propaganda text template; and performing product recommendation on the target object based on the product recommendation text and the target publicity template. According to the embodiment of the invention, the product recommendation accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the fields of financial technology and artificial intelligence technology, and in particular to a product recommendation method and device, electronic device and storage medium. Background Art

[0002] Product recommendation refers to making targeted product recommendations to users based on their interests and hobbies. For example, in the financial field, pension insurance can be recommended to a 58-year-old user who frequently checks retirement pension information.

[0003] At present, common product recommendation methods usually simply recommend products to all users without considering user needs and preferences, resulting in inaccurate product recommendations, which in turn causes users to be unable to pay attention to products that are closely related to their preferences, reducing users' desire to purchase products. Therefore, how to improve the accuracy of product recommendations has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a product recommendation method and device, an electronic device and a storage medium, aiming at improving the accuracy of product recommendations.

[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a product recommendation method, the method comprising:

[0006] Obtain historical behavior data of the target object and obtain promotional text templates of candidate products;

[0007] Based on the historical behavior data, construct a user profile of the target object;

[0008] Based on the user portrait, the historical behavior data and a preset text generation model, generating a recommendation copy for the candidate product to obtain a product recommendation text;

[0009] Based on the user portrait and the historical behavior data, a target promotional template is selected from the promotional text templates;

[0010] Based on the product recommendation text and the target promotion template, product recommendations are made to the target object.

[0011] In some embodiments, generating a recommendation text for the candidate product based on the user portrait, the historical behavior data, and a preset text generation model to obtain a product recommendation text includes:

[0012] Based on the user portrait and the historical behavior data, screening out a target product from the candidate products;

[0013] Generate product recommendation indication information based on the user portrait, the historical behavior data and the target product;

[0014] The text generation model is used to generate text for the user portrait and the product recommendation indication information to obtain the product recommendation text.

[0015] In some embodiments, generating product recommendation indication information based on the user portrait, the historical behavior data and the target product includes:

[0016] Determine product recommendation objects based on the user portrait;

[0017] Determining information emphasis content based on the historical behavior data;

[0018] Determining a recommended text type based on the historical behavior data and the target product;

[0019] The product recommendation indication information is constructed based on the product recommendation object, the information emphasis content and the recommendation text type.

[0020] In some embodiments, the step of generating text for the user portrait and the product recommendation indication information by using the text generation model to obtain the product recommendation text includes:

[0021] The user portrait and the product recommendation indication information are combined to obtain an indication text;

[0022] Based on the text generation model, context learning is performed on the instruction text to obtain text semantics;

[0023] Based on the text semantics, the product recommendation text is generated.

[0024] In some embodiments, the step of selecting a target promotional template from the promotional text templates based on the user portrait and the historical behavior data includes:

[0025] Perform feature analysis on the user portrait to obtain the product preference tendency of the target object;

[0026] Based on the product preference tendency, screening out products of interest to the target object from the candidate products;

[0027] Determine the type of promotional template based on the historical behavior data;

[0028] Based on the product of interest and the promotional template type, the promotional text template is screened to obtain the target promotional template.

[0029] In some embodiments, the recommending a product to the target object based on the product recommendation text and the target promotion template includes:

[0030] Performing a target query on the target promotion template to obtain a text filling position;

[0031] Based on the product recommendation text and the text filling position, obtain the target product promotion text for the target promotion template;

[0032] Based on the target product promotion text, the text is transmitted to the target object to obtain a user preferred product recommendation text.

[0033] In some embodiments, after recommending a product to the target object based on the product recommendation text and the target promotion template, the method further includes:

[0034] Acquire text feedback information of the target object based on the user's preferred product recommendation text feedback;

[0035] Adjusting the text generation model based on the text feedback information to obtain a text generation update model;

[0036] Based on the text generation update model and the target promotion template, product recommendations are made to the target object to obtain a target product recommendation text.

[0037] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a product recommendation device, the device comprising:

[0038] The information acquisition module is used to obtain the historical behavior data of the target object and obtain the promotional text template of the candidate product;

[0039] A portrait building module, used to build a user portrait of the target object based on the historical behavior data;

[0040] A copywriting generation module, used to generate a recommendation copy for the candidate product based on the user portrait, the historical behavior data and a preset text generation model to obtain a product recommendation text;

[0041] A template screening module, used to screen out a target promotional template from the promotional text templates based on the user portrait and the historical behavior data;

[0042] The text generation module is used to make product recommendations to the target object based on the product recommendation text and the target promotion template.

[0043] To achieve the above objectives, a third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0044] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0045] The product recommendation method and device, electronic device and storage medium proposed in the present application obtain the historical behavior data of the target object, and then build a user portrait of the target object based on the historical behavior data, clarify the user interests of the target object, so as to understand the user's product needs and improve the accuracy of product recommendations. Secondly, based on the user portrait, historical behavior data and a preset text generation model, recommendation copy is generated for candidate products to obtain product recommendation text, so as to improve the user's understanding of the product and improve the user's product recommendation experience comfort. Finally, based on the user portrait and historical behavior data, a target promotional template is screened from the promotional text template, and product recommendations are made to the target object based on the product recommendation text and the target promotional template, which can improve the user's product recommendation experience comfort and improve the accuracy of product recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of a product recommendation method provided in an embodiment of the present application;

[0047] Figure 2 yes Figure 1 Flow chart of step S103 in FIG.

[0048] Figure 3 yes Figure 2 Flow chart of step S202 in FIG.

[0049] Figure 4 yes Figure 2 Flow chart of step S203 in FIG.

[0050] Figure 5 yes Figure 1 Flow chart of step S104 in FIG.

[0051] Figure 6 yes Figure 1 Flow chart of step S105 in FIG.

[0052] Figure 7 is a flow chart of a product recommendation method provided by another embodiment of the present application;

[0053] Figure 8 is a schematic diagram of the structure of a product recommendation device provided in an embodiment of the present application;

[0054] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0058] First, some nouns involved in this application are analyzed:

[0059] User portrait: A user portrait is a detailed description generated by analyzing and integrating data such as user behavior, preferences, needs, and characteristics. It is designed to help companies deeply understand target users, thereby achieving personalized marketing, product design, and service optimization. By building user portraits, companies can better identify potential customers, predict user needs, and develop more targeted strategies to improve user experience and satisfaction.

[0060] Instructions: In natural language processing and generation tasks, instructions are input information or instructions given to the model to guide the model to generate a specific type of output. It can be a short question, a description, or a specific task instruction. Through clear prompts, the model can understand the user's intention and generate relevant text, answers, or suggestions. Effective instruction design can significantly improve the quality and relevance of generated content.

[0061] Product recommendation refers to making targeted product recommendations to users based on their interests and hobbies. For example, in the financial field, pension insurance can be recommended to a 58-year-old user who frequently checks retirement pension information.

[0062] At present, common product recommendation methods usually simply recommend products to all users without considering user needs and preferences, resulting in inaccurate product recommendations, which in turn causes users to be unable to pay attention to products that are closely related to their preferences, reducing users' desire to purchase products. Therefore, how to improve the accuracy of product recommendations has become a technical problem that needs to be solved urgently.

[0063] Based on this, the embodiments of the present application provide a product recommendation method and device, an electronic device and a storage medium, aiming to improve the accuracy of the estimated weight of the insured person.

[0064] The product recommendation method and device, electronic device and storage medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the product recommendation method in the embodiments of the present application is described.

[0065] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0066] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0067] The product recommendation method provided in the embodiment of the present application relates to the fields of financial technology and artificial intelligence technology. The product recommendation method provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the product recommendation method, etc., but is not limited to the above forms.

[0068] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0069] Figure 1 is an optional flowchart of the product recommendation method provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S105.

[0070] Step S101, obtaining historical behavior data of the target object and obtaining a promotional text template of a candidate product;

[0071] Step S102, constructing a user profile of the target object based on historical behavior data;

[0072] Step S103, generating recommendation text for the candidate products based on the user portrait, historical behavior data and a preset text generation model to obtain a product recommendation text;

[0073] Step S104, based on the user portrait and historical behavior data, select a target promotional template from the promotional text templates;

[0074] Step S105: recommend products to the target object based on the product recommendation text and the target promotion template.

[0075] Steps S101 to S105 shown in the embodiment of the present application construct a profile of the target object by acquiring the historical behavior data of the target object, obtain a user profile, and then obtain a promotional text template of the candidate product, and generate a recommendation text for the candidate product based on the user portrait, historical behavior data and a preset text generation model to obtain a product recommendation text. Secondly, based on the user portrait and historical behavior data, a target promotional template is screened out from the promotional text templates. Finally, based on the product recommendation text and the target promotional template, a product recommendation is made to the target object. Therefore, this application obtains the historical behavior data of the target object, and then builds a user portrait of the target object based on the historical behavior data, clarifies the user interests of the target object, so as to understand the user's product needs and improve the accuracy of product recommendations. Secondly, based on the user portrait, historical behavior data and the preset text generation model, recommendation copy is generated for candidate products to obtain product recommendation text, so as to improve the user's understanding of the product and improve the user's product recommendation experience comfort. Finally, based on the user portrait and historical behavior data, the target promotional template is screened out from the promotional text template, and product recommendations are made to the target object based on the product recommendation text and the target promotional template, which can improve the user's product recommendation experience comfort and improve the accuracy of product recommendations.

[0076] In step S101 of some embodiments, the target object may be a potential purchaser of the candidate product, for example, a user who frequently browses the candidate product page. The historical behavior data may be the behavior data of the target object on the Internet in the past month, such as browsing history, purchase behavior, social interaction, etc. The candidate product may be a product that an enterprise or company needs to sell, for example, medical insurance and pension insurance that a financial enterprise needs to sell. The promotional text template may be a popular science text, case analysis text, cost analysis text, and activity promotion text template for the candidate product.

[0077] This application can use a web crawler to crawl the target object's behavior records on the Internet, thereby obtaining the target object's historical behavior records. It can also summarize the basic information of the candidate products to obtain the common characteristics of the candidate products, and then generate a promotional text template for the candidate products based on the common characteristics and pre-set wording.

[0078] In step S102 of some embodiments, the preference prediction of the target object can be realized based on the historical behavior data of the target object, so as to obtain the product preference of the target object, and then a portrait of the target object is generated based on the product preference and the basic information of the target object, so as to obtain the user portrait of the target object, wherein the basic information of the target object may be the age, gender, occupation and other information of the target object.

[0079] In step S103 of some embodiments, after determining the user portrait and historical behavior data of the target object, a product recommendation text of the product of interest to the target object among the candidate products can be generated based on a pre-trained text generation model.

[0080] For details, see Figure 2 In some embodiments, step S103 may include but is not limited to steps S201 to S203:

[0081] Step S201, screening out target products from candidate products based on user portraits and historical behavior data;

[0082] Step S202, generating product recommendation indication information based on the user portrait, historical behavior data and target products;

[0083] Step S203, generating text for the user portrait and product recommendation indication information through a text generation model to obtain a product recommendation text.

[0084] In step S201 of some embodiments, the interests and hobbies of the target object can be clarified by analyzing the user portrait and historical behavior data of the target object, and then one or more products that best match the interests and hobbies of the target object are screened out from the candidate products, namely the target products.

[0085] In step S202 of some embodiments, the product recommendation indication information is used to indicate the product information that needs to be included in the text output by the text generation model.

[0086] This application can determine the recommendation object of the target product, that is, the product recommendation object, based on the user portrait, determine the product content that needs to be emphasized in the product recommendation indication information, that is, the information emphasis content, based on the historical behavior data, determine the type of product recommendation text that needs to be generated, that is, the recommendation text type, and finally, based on the above-mentioned product recommendation object, information emphasis content and recommendation text type, clarify the content of the product recommendation indication information and construct the product recommendation indication information.

[0087] For details, see Figure 3 In some embodiments, step S202 may include but is not limited to steps S301 to S304:

[0088] Step S301, determining product recommendation objects based on user portraits;

[0089] Step S302, determining information emphasis content based on historical behavior data;

[0090] Step S303, determining the recommended text type based on the historical behavior data and the target product;

[0091] Step S304: construct product recommendation indication information based on the product recommendation object, information emphasis content and recommendation text type.

[0092] In step S301 of some embodiments, based on the user portrait of the target object, the basic information of the target object, such as age, gender, etc., can be clarified, and then based on the basic information, it is determined to which type of people the target product needs to be recommended and the product recommendation object. For example, if the user portrait of the target object shows that the target object is a 35-year-old married woman, then the product recommendation object can be a 35-year-old married woman.

[0093] In step S302 of some embodiments, the product content that the target object is concerned about can be clarified based on the historical behavior data of the target object, and then the product content that needs to be emphasized in the product recommendation indication information is determined based on the product content that the target object is concerned about, that is, the information emphasis content. For example, the target object has browsed a variety of children's education insurance and medical insurance, and mentioned in a recent consultation that she hopes to plan for her child's future and family health protection, then the information emphasis content can be "emphasizing how the target product can help her achieve family protection and her child's future."

[0094] In step S303 of some embodiments, the product information of the target product that the target object needs to know is determined based on the historical behavior data of the target object, and then the text type of the product recommendation text is determined based on the product information of the target product that the target object needs to know. For example, if the target object has browsed popular science information on medical insurance many times, the recommended text type may be a popular science text.

[0095] In step S304 of some embodiments, based on the above-mentioned product recommendation object, information emphasis content and recommended text type, the text content that needs to be output by the text generation model can be clarified, and product recommendation indication information is generated based on the determined text content that needs to be output by the current generation model. For example, the product recommendation object is a 45-year-old male, the information emphasis content is "emphasis on wealth appreciation and a comfortable retirement life", and the recommended text type is a popular science article on pension insurance, then the product recommendation indication information can be "write a marketing copy for high-end pension insurance for a 45-year-old male customer B, emphasizing wealth appreciation and a comfortable retirement life."

[0096] In steps S301 to S304 of some embodiments, the product recommendation object is determined based on the user portrait, the information emphasis content is determined based on the historical behavior data, the recommended text type is determined based on the historical behavior data and the target product, and then the product recommendation indication information is constructed based on the product recommendation object, the information emphasis content and the recommended text type. This enables text recommendations for products that the target object is interested in, thereby improving the accuracy of product recommendations and the user experience.

[0097] In step S203 of some embodiments, the above-mentioned product recommendation indication information and user portrait are input into a pre-trained text generation model, and then the text generation model is used to perform feature learning on the product recommendation indication information and user portrait, so as to generate a product recommendation text for the target product.

[0098] For details, see Figure 4 In some embodiments, step S203 may include but is not limited to steps S401 to S403:

[0099] Step S401, combining the user portrait and the product recommendation indication information to obtain an indication text;

[0100] Step S402, based on the text generation model, context learning is performed on the instruction text to obtain text semantics;

[0101] Step S403: Generate product recommendation text based on text semantics.

[0102] In step S401 of some embodiments, the user portrait of the target object and the product recommendation indication information of the target object are textually spliced ​​to obtain an indication text.

[0103] In step S402 and step S403 of some embodiments, the indicative text is encoded using a text generation model to obtain an encoded text, and then the encoded text is contextually understood and attention weights are allocated to obtain the text semantics of the indicative text. The text generation model then generates a product recommendation text for the target product based on the text semantics of the indicative text.

[0104] In steps S401 to S403 of some embodiments, the user portrait and product recommendation indication information are spliced ​​to obtain an indication text, and the indication text is contextually learned based on a text generation model to obtain text semantics. Then, based on the text semantics, a product recommendation text is generated, so that the generated product recommendation text can meet the target object's need to understand the target product, thereby increasing the target object's desire to purchase the target product.

[0105] In steps S201 to S203 of some embodiments, target products are screened out from candidate products based on user portraits and historical behavior data, and products that need to be recommended to the target objects are identified. Product recommendation indication information is then generated based on the user portraits, historical behavior data, and target products, and text content that needs to be displayed in the product recommendation text is identified. Finally, a text generation model is used to generate text for the user portraits and product recommendation indication information to obtain a product recommendation text, thereby improving the accuracy of product recommendations and user experience.

[0106] In step S104 of some embodiments, based on the user portrait of the target object, the target object's preference for products, that is, the product preference tendency, can be clarified, and then based on the product preference tendency, the products of interest to the target object can be screened out from the candidate products. Furthermore, based on the historical behavior data of the target object, the promotional template type of the promotional text template required by the target object can also be determined, and then the promotional text templates are screened according to the products of interest and the promotional template type to obtain the target promotional template.

[0107] For details, see Figure 5 In some embodiments, step S104 may include but is not limited to steps S501 to S505:

[0108] Step S501, performing feature analysis on the user portrait to obtain the product preference of the target object;

[0109] Step S502, based on product preference, screening out products of interest to the target object from candidate products;

[0110] Step S503, determining the type of promotional template based on historical behavior data;

[0111] Step S504, based on the product of interest and the type of promotional template, the promotional text template is screened to obtain a target promotional template.

[0112] In step S501 of some embodiments, the user portrait is constructed based on the user's historical behavior data and basic information. Therefore, by analyzing the user portrait of the target object, the behavior characteristics of the target object can be obtained, and based on the behavior characteristics, the product preference tendency of the target object can be predicted.

[0113] In step S502 of some embodiments, products that meet the product preference tendency are screened out from the candidate products, namely, the products of interest to the target object.

[0114] In step S503 of some embodiments, the product content that the target object wants to know can be predicted based on the historical behavior data of the target object, so as to clarify the promotional template type of the required promotional text template.

[0115] In step S504 of some embodiments, a target promotional template can be screened out from the promotional text templates according to the target object's product of interest and the promotional template type of the promotional text template, wherein the target promotional template refers to a template used to describe the product of interest, and the type of the target promotional template is a promotional template type, for example, the promotional text template is a popular science template for medical insurance, a case analysis template for medical insurance, a cost analysis template for medical insurance, an activity promotion template for medical insurance, a popular science template for pension insurance, a case analysis template for pension insurance, a cost analysis template for pension insurance, and an activity promotion template for pension insurance. When the product of interest is medical insurance and the promotional template type is a cost analysis template, the target promotional template is a cost analysis template for medical insurance.

[0116] In steps S501 to S504 of some embodiments, the product preference of the target object is obtained by performing feature analysis on the user portrait, and then the products of interest to the target object are screened out from the candidate products based on the product preference. Furthermore, the promotional template type is determined based on the historical behavior data, and then the promotional text template is screened based on the products of interest and the promotional template type to obtain the target promotional template, which can ensure the accuracy of the target promotional template and thus improve the accuracy of product recommendations.

[0117] In step S105 of some embodiments, the product recommendation text is merged with the target promotion template to obtain the user preferred product recommendation text of the target object, and then the user preferred product recommendation text is sent to the target object, thereby achieving product recommendation for the target object.

[0118] For details, see Figure 6 In some embodiments, step S105 may include but is not limited to steps S601 to S603:

[0119] Step S601, performing a target query on the target promotion template to obtain a text filling position;

[0120] Step S602, based on the product recommendation text and the text filling position, obtain the target product promotion text for the target promotion template;

[0121] Step S603: Based on the target product promotion text, text transmission is performed on the target object to obtain a user preferred product recommendation text.

[0122] In step S601 of some embodiments, the text filling position is obtained by querying the target area that needs to be filled, supplemented or modified in the target promotion template.

[0123] In step S602 of some embodiments, the generated product recommendation text is filled into the text filling position of the target promotion template, so as to supplement the instance of the target promotion template and obtain the target product promotion text that the user is interested in.

[0124] In step S603 of some embodiments, the target product promotion text is transmitted to the target object through message subscription or the like, and a user preferred product recommendation text can be obtained.

[0125] In steps S601 to S603 of some embodiments, a target query is performed on the target promotional template to obtain the text filling position. Secondly, based on the product recommendation text and the text filling position, the target product promotional text is obtained for the target promotional template. Finally, based on the target product promotional text, text transmission is performed on the target object to obtain the user preferred product recommendation text, thereby improving the accuracy of product recommendations and thus enhancing the user's comfort level in product recommendation experience.

[0126] In some embodiments, in order to further improve the accuracy of product recommendations, feedback information from the target object on the user's preferred product recommendation text, i.e., text feedback information, can be collected, and then the parameters of the above-mentioned text generation model can be adjusted according to the text feedback information to improve the model accuracy, thereby improving the accuracy of the product recommendation text output by the model and improving the accuracy of product recommendations.

[0127] For details, see Figure 6 In some embodiments, the product recommendation method may also include but is not limited to steps S701 to S703:

[0128] Step S701, obtaining text feedback information of the target object based on the user's preferred product recommendation text feedback;

[0129] Step S702, adjusting the text generation model based on the text feedback information to obtain an updated text generation model;

[0130] Step S703, generating an update model and a target promotion template based on the text, recommending products to the target object, and obtaining a target product recommendation text.

[0131] In step S701 of some embodiments, the target object can provide feedback on the user preferred product recommendation text by scoring the user preferred product recommendation text, so that the server can obtain text feedback information of the target object based on the user preferred product recommendation text feedback.

[0132] In step S702 of some embodiments, when the text feedback information indicates that the user's preferred product recommendation text does not meet the needs of the target object, it is necessary to adjust the parameters of the text generation model to obtain an updated text generation model to improve the output precision of the text generation model and the accuracy of the model.

[0133] In step S703 of some embodiments, a product recommendation update text is generated based on a text generation update model, and then a product recommendation is made to a target object based on the product recommendation update text and a target promotional template, thereby obtaining a target product recommendation text.

[0134] In steps S701 to S703 of some embodiments, by obtaining text feedback information of the target object based on the user preference product recommendation text feedback, and then adjusting the text generation model according to the text feedback information to obtain the text generation update model, the model accuracy can be improved, thereby improving the accuracy of the product recommendation text output by the model. Furthermore, according to the text generation update model and the target promotional template, product recommendations are made to the target object to obtain the target product recommendation text, which can improve the accuracy of the target product recommendation text, thereby improving the accuracy of the product recommendation.

[0135] In one application scenario, taking financial products as an example, when the historical behavior data of the target object obtained is multiple browsing of popular science introductions of medical insurance, and the promotional text templates of the candidate products obtained are popular science templates, case analysis templates, and cost explanation templates of various insurances, based on the historical behavior data of the target object multiple browsing of popular science introductions of medical insurance, a user portrait of the target object who is interested in medical insurance can be constructed, and then based on the generated user portrait, the above-obtained historical behavior data and the pre-trained insurance product text generation model, a product recommendation text for medical insurance can be generated. Secondly, based on the above-mentioned user portrait and historical behavior data, the popular science template of medical insurance can be screened out from various popular science templates, case analysis templates, and cost explanation templates as the target promotional template. Finally, based on the above-mentioned product recommendation text of medical insurance and the popular science template of medical insurance, product recommendations can be made to the target object.

[0136] This application obtains the historical behavior data of the target object, and then builds a user portrait of the target object based on the historical behavior data, clarifies the user interests of the target object, so as to understand the user's product needs and improve the accuracy of product recommendations. Secondly, based on the user portrait, historical behavior data and a preset text generation model, recommendation copy is generated for candidate products to obtain product recommendation text, so as to improve the user's understanding of the product and improve the user's product recommendation experience comfort. Finally, based on the user portrait and historical behavior data, the target promotional template is screened out from the promotional text template, and product recommendations are made to the target object based on the product recommendation text and the target promotional template, which can improve the user's product recommendation experience comfort and improve the accuracy of product recommendations.

[0137] See also Figure 8 The embodiment of the present application also provides a product recommendation device, which can implement the above-mentioned product recommendation method, and the device includes:

[0138] The information acquisition module 801 is used to acquire the historical behavior data of the target object and obtain the promotional text template of the candidate product;

[0139] A portrait construction module 802 is used to construct a user portrait of a target object based on historical behavior data;

[0140] The copywriting generation module 803 is used to generate recommendation copywriting for candidate products based on user portraits, historical behavior data and a preset text generation model to obtain product recommendation text;

[0141] The template screening module 804 is used to screen out target promotional templates from promotional text templates based on user portraits and historical behavior data;

[0142] The text generation module 805 is used to make product recommendations to the target object based on the product recommendation text and the target promotion template.

[0143] The specific implementation of the product recommendation device is basically the same as the specific implementation of the above-mentioned product recommendation method, and will not be repeated here.

[0144] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above product recommendation method when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.

[0145] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0146] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0147] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the product recommendation method of the embodiment of this application;

[0148] Input / output interface 903, used to implement information input and output;

[0149] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0150] A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0151] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0152] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above-mentioned product recommendation method is implemented.

[0153] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0154] The product recommendation method, product recommendation device, electronic device and storage medium provided in the embodiments of the present application construct a portrait of the target object by acquiring the historical behavior data of the target object, obtain the user portrait, and then obtain the promotional text template of the candidate product, and generate recommendation copy for the candidate product based on the user portrait, historical behavior data and a preset text generation model to obtain product recommendation text. Secondly, based on the user portrait and historical behavior data, a target promotional template is screened out from the promotional text templates. Finally, based on the product recommendation text and the target promotional template, a product recommendation is made to the target object.

[0155] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0156] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0157] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0159] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0160] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0161] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0162] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0163] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0164] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0165] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A product recommendation method, characterized in that: The method comprises: Obtain historical behavior data of the target object and obtain promotional text templates of candidate products; Based on the historical behavior data, construct a user profile of the target object; Based on the user portrait, the historical behavior data and a preset text generation model, generating a recommendation copy for the candidate product to obtain a product recommendation text; Based on the user portrait and the historical behavior data, a target promotional template is selected from the promotional text templates; Based on the product recommendation text and the target promotion template, product recommendations are made to the target object.

2. The method according to claim 1, characterized in that Generating a recommendation text for the candidate product based on the user portrait, the historical behavior data and a preset text generation model to obtain a product recommendation text includes: Based on the user portrait and the historical behavior data, screening out a target product from the candidate products; Generate product recommendation indication information based on the user portrait, the historical behavior data and the target product; The text generation model is used to generate text for the user portrait and the product recommendation indication information to obtain the product recommendation text.

3. The method according to claim 2, characterized in that The generating product recommendation indication information based on the user portrait, the historical behavior data and the target product includes: Determine product recommendation objects based on the user portrait; Determining information emphasis content based on the historical behavior data; Determining a recommended text type based on the historical behavior data and the target product; The product recommendation indication information is constructed based on the product recommendation object, the information emphasis content and the recommendation text type.

4. The method according to claim 2, characterized in that: The step of generating text for the user portrait and the product recommendation indication information by using the text generation model to obtain the product recommendation text includes: The user portrait and the product recommendation indication information are combined to obtain an indication text; Based on the text generation model, context learning is performed on the instruction text to obtain text semantics; Based on the text semantics, the product recommendation text is generated.

5. The method according to claim 1, characterized in that The step of selecting a target promotional template from the promotional text templates based on the user portrait and the historical behavior data includes: Perform feature analysis on the user portrait to obtain the product preference tendency of the target object; Based on the product preference tendency, screening out products of interest to the target object from the candidate products; Determine the type of promotional template based on the historical behavior data; Based on the product of interest and the promotional template type, the promotional text template is screened to obtain the target promotional template.

6. The method according to any one of claims 1 to 5, characterized in that: The recommending a product to the target object based on the product recommendation text and the target promotion template includes: Performing a target query on the target promotion template to obtain a text filling position; Based on the product recommendation text and the text filling position, obtain the target product promotion text for the target promotion template; Based on the target product promotion text, the text is transmitted to the target object to obtain a user preferred product recommendation text.

7. The method according to claim 6, characterized in that After recommending the product to the target object based on the product recommendation text and the target promotion template, the method further includes: Acquire text feedback information of the target object based on the user's preferred product recommendation text feedback; Adjusting the text generation model based on the text feedback information to obtain a text generation update model; Based on the text generation update model and the target promotion template, product recommendations are made to the target object to obtain a target product recommendation text.

8. A product recommendation device, characterized in that: The device comprises: The information acquisition module is used to obtain the historical behavior data of the target object and obtain the promotional text template of the candidate product; A portrait building module, used to build a user portrait of the target object based on the historical behavior data; A copywriting generation module, used to generate a recommendation copy for the candidate product based on the user portrait, the historical behavior data and a preset text generation model to obtain a product recommendation text; A template screening module, used to screen out a target promotional template from the promotional text templates based on the user portrait and the historical behavior data; The text generation module is used to make product recommendations to the target object based on the product recommendation text and the target promotion template.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the product recommendation method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the product recommendation method according to any one of claims 1 to 7 is implemented.

Images