Product promotion scheme generation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a product promotion scheme generation method and device, electronic equipment and a storage medium, belongs to the technical field of artificial intelligence, and is suitable for the field of financial science and technology. The method comprises the following steps: acquiring user information and user operation behavior data; determining product demand information of the user according to the user information and the user operation behavior data; determining target product information according to the product demand information; and performing product promotion period detection and product promotion triggering scene detection on the target product information, and generating a target product promotion scheme based on the target product information, the detected product promotion period and the product promotion triggering scene. According to the embodiment of the invention, the target product promotion scheme is generated based on the target product information, the product promotion period and the product promotion triggering scene, the demands of the target user and the product characteristics can be comprehensively considered, the personalized product promotion scheme is generated, and the promotion effect of the product promotion scheme is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology and is applicable to the field of financial technology, and in particular to a method and device for generating a product promotion plan, an electronic device, and a storage medium. Background Art

[0002] Traditional product promotion plan generation methods typically involve a company's operations staff selecting target product copy templates from a pre-set product copy template library to generate standardized product promotion copy. Taking the financial scenario as an example, to meet the need for health insurance product promotion copy, operations staff can select a health insurance product copy template from the product copy template library to generate product promotion copy that reads, "Our health insurance products offer coverage for a variety of diseases, providing you with comprehensive health protection and ensuring your family's health." However, because each customer has different product needs and preferences, this method only generates standardized product promotion copy and fails to meet the individual needs of customers, resulting in poor product promotion results. Therefore, improving the effectiveness of product promotion plans has become a pressing issue. Summary of the Invention

[0003] The main purpose of the embodiments of the present application is to propose a method and device for generating a product promotion plan, an electronic device and a storage medium, aiming to improve the promotion effect of the product promotion plan.

[0004] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a method for generating a product promotion plan, the method comprising:

[0005] In response to a product promotion request, obtaining user information of a target user and obtaining user operation behavior data of the target user;

[0006] Inferring product demand of the target user based on the user information and the user operation behavior data to obtain product demand information;

[0007] Based on the product demand information, product matching is performed from a pre-built product information database to obtain target product information;

[0008] Performing product promotion cycle detection on the target product information to obtain the product promotion cycle;

[0009] Performing product promotion triggering scenario detection on the target product information to obtain a product promotion triggering scenario;

[0010] A target product promotion plan is generated based on the target product information, the product promotion cycle, and the product promotion triggering scenario.

[0011] In some embodiments, generating a target product promotion plan based on the target product information, the product promotion cycle, and the product promotion triggering scenario includes:

[0012] Obtain product promotion influencing factors of the target product information, and determine a product promotion plan category based on the product promotion influencing factors; the product promotion plan category includes at least one of a copywriting promotion plan category, an image promotion plan category, and a video promotion plan category;

[0013] Constructing a plan based on the product promotion plan category, the product promotion cycle, and the product promotion triggering scenario to generate instruction information;

[0014] Based on the solution generation indication information, a pre-trained solution generation model is instructed to generate a promotion solution to obtain the target product promotion solution.

[0015] In some embodiments, the solution generation instruction information includes at least one of text solution generation instruction information, graphic solution generation instruction information, and video solution generation instruction information;

[0016] The step of instructing the pre-trained solution generation model based on the solution generation instruction information to generate a promotion solution to obtain the target product promotion solution includes:

[0017] If the solution generation instruction information is copy solution generation instruction information, then instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information, obtaining a product copy promotion solution, and determining the product copy promotion solution as the target product promotion solution;

[0018] If the solution generation instruction information is a graphic and text solution generation instruction information, then instructing the solution generation model to generate a product graphic and text promotion solution based on the graphic and text solution generation instruction information to obtain a product graphic and text promotion solution, and determining the product graphic and text promotion solution as the target product promotion solution;

[0019] If the solution generation indication information is video solution generation indication information, the solution generation model is instructed to generate a product video promotion solution based on the video solution generation indication information to obtain a product video promotion solution, and the product video promotion solution is determined as the target product promotion solution.

[0020] In some embodiments, instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information to obtain the product copy promotion solution includes:

[0021] Instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information to obtain a candidate product copy promotion solution;

[0022] Conducting promotion effect testing on the candidate product copy promotion plan to obtain promotion effect test data;

[0023] The candidate product copy promotion plans are screened according to the promotion effect test data to obtain the product copy promotion plan.

[0024] In some embodiments, inferring product demand of the target user based on the user information and the user operation behavior data to obtain product demand information includes:

[0025] Performing user level identification on the target user according to the user information to obtain the target user level;

[0026] Performing behavior preference detection on the user operation behavior data to obtain target user interest tags;

[0027] The product demand of the target user is analyzed according to the target user level and the target user interest tag to obtain the product demand information.

[0028] In some embodiments, performing behavior preference detection on the user operation behavior data to obtain target user interest tags includes:

[0029] Extracting features from the user operation behavior data to obtain user operation behavior features; the user operation behavior features are time series features;

[0030] Performing behavioral interest analysis on the user operation behavior characteristics to obtain the target user interest tag.

[0031] In some embodiments, before performing product matching from a pre-built product information database based on the product demand information to obtain target product information, the method further includes:

[0032] Acquire initial product information; the initial product information includes structured product information and unstructured product information;

[0033] Performing entity recognition on the structured product information to obtain structured entity information, and performing entity recognition on the unstructured product information to obtain unstructured entity information;

[0034] Performing entity relationship extraction on the structured entity information to obtain structured entity relationships, and performing entity relationship extraction on the unstructured entity information to obtain unstructured entity relationships;

[0035] Embedding the structured entity information and the structured entity relationship into a knowledge graph to obtain a structured knowledge base, and embedding the unstructured entity information and the structured entity relationship into a knowledge graph to obtain an unstructured knowledge base;

[0036] The product information base is determined based on the structured knowledge base and the unstructured knowledge base.

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

[0038] A user data acquisition module, configured to obtain user information of a target user and user operation behavior data of the target user in response to a product promotion request;

[0039] A product demand inference module is used to infer the product demand of the target user based on the user information and the user operation behavior data to obtain product demand information;

[0040] A product matching module is used to match products from a pre-built product information database based on the product demand information to obtain target product information;

[0041] A product promotion cycle detection module is used to perform product promotion cycle detection on the target product information to obtain a product promotion cycle; wherein the product promotion cycle is used to represent the period of the target product in the decision-making process of the target user;

[0042] A product promotion triggering scenario detection module is used to perform product promotion triggering scenario detection on the target product information to obtain a product promotion triggering scenario; wherein the product promotion triggering scenario is used to represent the promotion context associated with the target product;

[0043] The product promotion plan generation module is used to generate a target product promotion plan based on the target product information, the product promotion cycle and the product promotion triggering scenario.

[0044] To achieve the above-mentioned purpose, the 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 the processor implements the method described in the first aspect when executing the computer program.

[0045] To achieve the above-mentioned purpose, the fourth aspect of the embodiments of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method of the above-mentioned first aspect.

[0046] The product promotion plan generation method and device, electronic device and storage medium proposed in this application first obtain the user information of the target user and the user operation behavior data of the target user by responding to the product promotion request, which can fully understand the personal information and user behavior preferences of the target user, provide basic data support for the subsequent generation of personalized product promotion plans, and infer the product needs of the target user based on the user information and user operation behavior data. It can accurately infer the user's specific needs for the product based on the target user information, and further match products that meet the user's needs from the product information library, which can ensure that the content of the subsequently generated product promotion plan is highly relevant to the user's needs, so as to improve product promotion. The promotion effect of the plan; secondly, the product promotion cycle detection and product promotion trigger scenario detection are carried out on the target product information, which can identify the specific period of the target product in the target user's decision-making process, facilitate the subsequent dynamic adjustment of the promotion strategy, and can also associate external events to make the subsequent promotion plan more scenario-persuasive, which helps to further improve the effect of the subsequent product promotion plan generation; finally, based on the target product information, product promotion cycle and product promotion trigger scenario, the target product promotion plan is generated, which can comprehensively consider the needs of the target users and product characteristics, generate personalized product promotion plans, effectively solve the problem of not being able to meet the personalized needs of customers, and significantly improve the promotion effect of the product promotion plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flowchart of the method for generating a product promotion plan provided in an embodiment of the present application;

[0048] Figure 2 yes Figure 1 Flowchart of step S102 in FIG.

[0049] Figure 3 yes Figure 2 Flowchart of step S202 in FIG.

[0050] Figure 4 This is another flow chart of the method for generating a product promotion plan provided in an embodiment of the present application;

[0051] Figure 5 yes Figure 1 Flowchart of step S106 in FIG.

[0052] Figure 6 yes Figure 5 Flowchart of step S503 in FIG.

[0053] Figure 7 yes Figure 6 Flowchart of step S601 in FIG.

[0054] Figure 8This is a schematic diagram of the structure of a device for generating a product promotion plan provided in an embodiment of the present application;

[0055] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0057] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0059] First, let’s analyze some of the terms used in this application:

[0060] Artificial intelligence (AI) is a new technical discipline that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. A branch of computer science, AI seeks to understand the essence of intelligence and create new intelligent machines that can respond in a manner similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thinking. It also encompasses theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0061] The embodiments of the present application provide a method and device for generating a product promotion plan, an electronic device, and a storage medium, aiming to improve the promotion effect of the product promotion plan.

[0062] The product promotion plan generation 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 promotion plan generation method in the embodiments of the present application is described.

[0063] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. 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 achieve optimal results.

[0064] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0065] The product promotion scheme generation method provided in the embodiment of the present application relates to the field of artificial intelligence technology. The product promotion scheme generation method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a 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 can be configured as a server cluster or a distributed system composed of multiple physical servers, or can 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 promotion scheme generation method, etc., but is not limited to the above forms.

[0066] 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, and the like. 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, and the like 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 via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0067] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on user information, user operation behavior data or characteristic-related data, the user's permission or consent will be obtained first, and the collection, use and processing of such data will comply with relevant laws, regulations and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0068] Figure 1 This is an optional flowchart of the method for generating a product promotion plan provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S106.

[0069] Step S101 , in response to a product promotion request, obtaining user information of a target user and obtaining user operation behavior data of the target user.

[0070] Step S102: inferring product demand of target users based on user information and user operation behavior data to obtain product demand information.

[0071] Step S103: performing product matching from a pre-built product information database based on the product demand information to obtain target product information.

[0072] Step S104: Perform product promotion cycle detection on the target product information to obtain the product promotion cycle.

[0073] Step S105: Perform product promotion triggering scenario detection on the target product information to obtain the product promotion triggering scenario.

[0074] Step S106: Generate a target product promotion plan based on the target product information, product promotion cycle, and product promotion triggering scenario.

[0075] In the steps S101 to S106 shown in the embodiment of the present application, first, by responding to the product promotion request, the user information of the target user and the user operation behavior data of the target user are obtained, so that the personal information and behavior preferences of the target user can be fully understood, and basic data support can be provided for the subsequent generation of personalized product promotion plans. The product demand of the target user is inferred based on the user information and user operation behavior data, and the specific demand of the user for the product can be accurately inferred based on the target user information. The product that meets the user's demand can be further matched from the product information library, which can ensure that the content of the subsequently generated product promotion plan is highly relevant to the user's demand, so as to improve the effectiveness of the product promotion plan. Promotion effect; secondly, the product promotion cycle detection and product promotion trigger scenario detection are carried out on the target product information, which can identify the specific period of the target product in the target user's decision-making process, facilitate the subsequent dynamic adjustment of the promotion strategy, and can also be associated with external events, so that the subsequent promotion plan is more scenario-persuasive, which helps to further improve the effect of the subsequent product promotion plan generation; finally, based on the target product information, product promotion cycle and product promotion trigger scenario, the target product promotion plan is generated, which can comprehensively consider the needs of the target users and product characteristics, generate personalized product promotion plans, effectively solve the problem of not being able to meet the personalized needs of customers, and significantly improve the promotion effect of the product promotion plan.

[0076] In step S101 of some embodiments, specifically, the product promotion request refers to an instruction signal actively triggered by a user to initiate a product promotion process. The product promotion request includes but is not limited to fields such as a user identity and a request timestamp.

[0077] Specifically, user information refers to basic user data related to the target user, including but not limited to the user's age, gender, occupation, and income level.

[0078] Specifically, in insurance application scenarios, user information may include the user's health status, the number of family members, and whether there is a mortgage.

[0079] Specifically, user operation behavior data refers to the user's operation records on the network platform or mini program.

[0080] For example, in insurance application scenarios, user operation behavior data includes but is not limited to current data such as browsing history, click behavior, purchase records, page dwell time, and search keywords on insurance web pages or insurance apps, in order to understand the types of insurance that users have inquired about on the insurance platform, the users' repeated comparisons of the coverage terms of different insurance products, the users' reading time for the insurance terms, and the modification of the parameters of the insurance amount calculator.

[0081] See also Figure 2 In some embodiments, step S102 includes but is not limited to steps S201 to S203:

[0082] Step S201: Identify the target user's user level based on the user information to obtain the target user level.

[0083] Step S202: Perform behavior preference detection on the user operation behavior data to obtain the target user interest tag.

[0084] Step S203 , analyzing the product requirements of the target user according to the target user level and the target user interest tags to obtain product requirement information.

[0085] In step S201 of some embodiments, specifically, the target user level refers to the user's level of product consumption ability, and the target user level may include ordinary level, gold card level, and diamond level.

[0086] Specifically, users can be evaluated and graded using predefined classification rules based on their annual income level, historical insurance policy amounts, occupation type, and other comprehensive values. Classification rules are determined based on specific application scenarios and are not limited here.

[0087] For example, in an insurance application scenario, if the user is a senior executive of a technology company with an annual income of more than 1 million yuan and a lifetime insurance policy of tens of millions, the user can be classified as a diamond-level user; if the user is a chief physician of a tertiary hospital with an annual income of 500,000 yuan, and holds a critical illness insurance with a coverage of 3 million yuan and a dividend-type annuity insurance with an annual premium of 50,000 yuan, the user can be classified as a gold-card-level user with strong insurance needs; if the user is a white-collar employee with an annual income of less than 200,000 yuan, who has only purchased a basic million-yuan medical insurance and whose historical average annual premium expenditure does not exceed 5,000 yuan, the user can be classified as an ordinary-level user.

[0088] In this embodiment, target users are identified by level based on user information, which can achieve accurate user classification, facilitate the subsequent provision of differentiated product promotion plans for users of different levels, and help improve user satisfaction and enhance user stickiness.

[0089] See also Figure 3In some embodiments, step S202 includes but is not limited to steps S301 to S303:

[0090] Step S301 , extracting features from user operation behavior data to obtain user operation behavior features; user operation behavior features are time series features.

[0091] Step S302: Perform behavioral interest analysis on the user's operation behavior characteristics to obtain target user interest tags.

[0092] In step S301 of some embodiments, specifically, the user operation behavior feature refers to a structured feature vector with time sequence characteristics, which is used to characterize the user's interactive behavior on the digital platform.

[0093] Specifically, to obtain the behavior sequence of user operation behavior data, the input behavior features of the behavior sequence at the current moment can be controlled through the input gate of the long short-term memory network (LSTM), and the forget gate of the LSTM can be used to decide whether to retain or discard the input behavior features of the previous moment to output the current memory state, and the output gate of the LSTM can be used to activate the current memory state and input behavior features to output the user operation behavior features at the current moment.

[0094] Specifically, user operation behavior characteristics may include the basic behavior type of each time step (such as browsing, clicking, sliding, etc.), behavior intensity (such as click frequency, sliding speed, etc.) and operation heat map characteristics (such as grid density statistics of the current page focus area).

[0095] For example, in an insurance product browsing scenario, user operation behavior characteristics can include the user's behavior characteristics every minute for 30 consecutive minutes, such as the rapid sliding characteristics on the corporate property insurance homepage in the first minute, the high-density click characteristics on the coverage terms area on the auto insurance product page in the fifth minute, and the time series characteristics of repeated parameter adjustment characteristics on the premium calculator in the 15th minute.

[0096] In this embodiment, by extracting features from user operation behavior data, it is possible to capture the changing patterns of user behavior over time in real time, which helps to subsequently identify user interest preferences for products in real time and facilitates the subsequent provision of personalized product promotions to users.

[0097] In step S302 of some embodiments, specifically, the target user interest tag refers to a user behavior preference identifier, which is used to reflect the user's product demand.

[0098] Specifically, the cosine similarity calculation can be performed between the predefined product interest pattern library (including insurance research, price comparison, quick decision-making and other patterns) and the user operation behavior characteristics to calculate the product interest matching score between the user operation behavior characteristics and the product interest pattern, and combined with the Transformer-based temporal attention model to identify the behavioral interest identifier in the user operation behavior characteristics. If it is detected that the user has switched from general browsing of the web page to focusing on the auto insurance terms, the auto insurance interest identifier is triggered, and the behavioral interest identifier score is weightedly fused with the product interest matching score to generate the target interest tag.

[0099] For example, in the insurance application scenario, the user's operation behavior characteristics indicate that multiple insurance products are evenly browsed in the first 15 minutes (matching the broad comparison mode score of 0.7), and the next 20 minutes focus on the auto insurance liability clauses of auto insurance products (triggering the clause in-depth analysis indicator), and finally generate a main label focusing on auto insurance protection, while retaining the secondary label of product comparison tendency.

[0100] In this embodiment, by performing behavioral interest analysis on user operation behavior characteristics, the user's product interest preferences can be accurately identified, providing data support for subsequent personalized and product promotion.

[0101] Through steps S301 to S302, by performing feature extraction and behavioral interest analysis on user operation behavior data, the user's interest preferences can be accurately identified, and target user interest tags can be generated. The user's operation behavior can be labeled to generate corresponding interest tags for different users, and it is convenient to better understand user needs in the future to generate personalized product promotion plans.

[0102] In step S203 of some embodiments, specifically, the product demand information refers to the user's specific demand for the product, and the product demand information is used to filter the user's products of interest.

[0103] For example, in an insurance scenario, for user A who is a gold card user and has a health insurance interest tag, it can be inferred that user A has a demand for health insurance products, and that such health insurance products generally have more comprehensive protection and higher insured amounts.

[0104] In this embodiment, the product needs of the target user are analyzed based on the target user level and the target user interest tags to obtain product demand information. The user level and user interest tags can be combined to accurately infer the user's specific needs for the product, which is helpful for subsequent matching of products that meet the user's needs.

[0105] Through steps S201 to S203, through user level identification, behavior preference detection and product demand analysis, the user's basic characteristics, interest preferences and specific needs can be fully understood, which not only improves the accuracy of product demand inference, but also helps the subsequently generated product promotion plan to be more in line with the user's actual needs, thereby improving user satisfaction and product market competitiveness.

[0106] See also Figure 4 In some embodiments, before step S103, the product promotion plan generation method further includes but is not limited to steps S401 to S405:

[0107] Step S401: Acquire initial product information; the initial product information includes structured product information and unstructured product information.

[0108] Step S402: Entity recognition is performed on the structured product information to obtain structured entity information, and entity recognition is performed on the unstructured product information to obtain unstructured entity information.

[0109] Step S403: extracting entity relationships from the structured entity information to obtain structured entity relationships, and extracting entity relationships from the unstructured entity information to obtain unstructured entity relationships.

[0110] In step S404, the structured entity information and the structured entity relationship are embedded in the knowledge graph to obtain a structured knowledge base, and the unstructured entity information and the structured entity relationship are embedded in the knowledge graph to obtain an unstructured knowledge base.

[0111] Step S405: determining a product information base based on the structured knowledge base and the unstructured knowledge base.

[0112] In step S401 of some embodiments, specifically, initial product information refers to a set of original data of various products, including structured product information and unstructured product information; structured product information refers to standardized data stored in formats such as tables and databases, such as the product's coverage, insured amount, insurance information, price range, and applicable population; unstructured product information refers to data that has not been standardized, such as the product's reimbursement conditions, product advantages, and user reviews.

[0113] In step S402 of some embodiments, for structured product information, field mapping is performed using a predefined data dictionary to convert column names in the database table into business entities, such as mapping the "premium_rate" field to the "rate" entity. For unstructured product information, product entities are identified using a domain-pretrained named entity recognition model (such as a BERT model). Taking the accident insurance exemption clause as an example, from the text "No compensation will be paid for accidents incurred by the insured while rock climbing," "rock climbing" can be extracted as a specific instance of the "exemption clause" entity.

[0114] In this embodiment, by performing entity recognition on structured product information and unstructured product information respectively, a unified entity naming standard can be established to ensure that product elements from different sources can be consistently identified and processed, providing a data basis for subsequent relationship extraction.

[0115] In step S403 of some embodiments, specifically, the structured entity relationship refers to an association relationship between structured product entities, such as the relationship between the name of an insurance product and the premium amount in a database.

[0116] Specifically, unstructured entity relationships refer to the association relationships between unstructured entities, such as the relationship between the behavior provisions clause and the exemption clause in the text of insurance terms.

[0117] Specifically, for structured entity information, the relationship between fields can be analyzed through predefined business rules (such as establishing a multiplication relationship between the "basic premium" entity and the "age coefficient" entity); for unstructured product entities, semantic dependency segmentation can be performed on the unstructured entity information to extract the semantic relationship between entities (such as establishing a positive correlation between "pension amount" and "years of payment").

[0118] In this embodiment, by extracting entity relationships from structured entity information and unstructured entity information respectively, it is possible to understand the intrinsic connections and logical relationships between entities, providing richer semantic information for subsequent knowledge graph construction.

[0119] In step S404 of some embodiments, specifically, a structured knowledge base refers to a knowledge storage collection in which structured entity information and its relationships are organized in the form of a graph; an unstructured knowledge base refers to a knowledge network formed by embedding unstructured entity information and its relationships into a knowledge graph.

[0120] Specifically, for structured knowledge, the TransE algorithm can be used to embed structured entities to represent them as vectors, and the vector distance between entities reflects the correlation; for unstructured knowledge, the attention mechanism can be used to capture the deep semantic associations of product texts. For example, the "deductible" entity of an auto insurance product is associated with the specific numerical value in the structured data, and is linked to the applicable circumstances in the auto insurance clause interpretation text to form a complete knowledge network representation.

[0121] In this embodiment, by embedding structured entity information and structured entity relationships into the knowledge graph, and embedding unstructured entity information and unstructured entity relationships into the knowledge graph, not only is the intelligent processing of product knowledge at the semantic level achieved, but also the scattered product information can be integrated into a unified knowledge graph, thereby improving the efficiency of subsequent product information queries.

[0122] In step S405 of some embodiments, the product information database is a knowledge graph that stores product-related information, and the product information database integrates structured and unstructured product data.

[0123] Specifically, the structured knowledge entities in the structured knowledge base can be associated with the unstructured knowledge entities in the unstructured knowledge base to construct a product knowledge graph with a hierarchical structure.

[0124] For example, in an insurance application scenario, the numerical entities in the structured knowledge base (such as "deductible is 10,000 yuan") are logically associated with the knowledge entities in the unstructured knowledge base (such as "deductible refers to the cumulative amount of a single visit") to store hierarchical product knowledge in the product information library.

[0125] In this embodiment, the product information base is determined based on the structured knowledge base and the unstructured knowledge base, which can integrate standardized data and unstructured data into multi-source information to form a comprehensive and semantically rich knowledge system, providing data support for subsequent product query and promotion.

[0126] Through steps S401 to S405, the scattered and heterogeneous product data can be transformed into a unified knowledge system, covering comprehensive content from standardized data to unstructured text descriptions. It not only enriches the dimensions of product information, but also reveals the relationship between different entity information of the product in the form of a knowledge graph, constructing a comprehensive and semantically rich product information database, which helps provide multi-dimensional product knowledge support for subsequent intelligent product matching.

[0127] In step S103 of some embodiments, specifically, the target product information may be product information that the user has purchased, or product information that the user is interested in, which is specifically determined based on the product demand information.

[0128] For example, in an insurance application scenario, if it is inferred that the user has a demand for health insurance, health insurance products will be searched in the product information database, and the target health insurance products will be screened out based on the user's demand details (such as insurance coverage, premium budget, etc.); if it is inferred that the user has purchased vehicle insurance products, the vehicle insurance products that have been purchased will be searched in the product information database as target product information.

[0129] In this embodiment, product matching is performed from a pre-built product information library based on product demand information to obtain target product information, which can quickly filter out products that meet user needs from a large number of products, thereby helping to improve the accuracy of subsequent product promotion plan generation.

[0130] In step S104 of some embodiments, specifically, the product promotion cycle refers to the specific period of the target product in the target user's purchase process, including but not limited to the customer operation cycle, the product sales cycle and the customer service cycle, wherein the customer operation cycle is used to reflect the period before the user purchases the product; the product sales cycle is used to reflect the period during which the user purchases the product; and the customer service cycle is used to reflect the period after the user purchases the product.

[0131] For example, in an insurance application scenario, the user has purchased auto insurance product B, and the expiration date of auto insurance product B is October 9. In this case, the entire September can be used as the customer operation cycle, October as the product sales cycle, and November as the customer service cycle.

[0132] In this embodiment, the product promotion cycle detection is performed on the target product information, which can identify the specific period of the target product in the target user's purchase process, set a reasonable time range for product promotion plan planning, ensure that the product promotion plan is highly correlated with user needs, and help improve the effectiveness of product promotion.

[0133] In step S105 of some embodiments, specifically, the product promotion triggering scenario refers to a specific scenario or event that triggers the product promotion activity. The product promotion triggering scenario may include but is not limited to real-time hot news (such as weather forecasts, natural disasters, social hot spots, etc.), holidays and user generation, etc.

[0134] Specifically, during the product promotion trigger scenario detection process, the insurance company's internal information aggregation platform can be connected in real time through the API interface to obtain multi-dimensional scenario data streams. For weather forecast scenario data, if a typhoon red alert is detected in the target area, the "Natural Disaster Emergency Response" scenario tag is triggered. For holiday and social hot spots, news keywords are analyzed through natural language processing technology. If reports related to the "summer travel peak" are identified, the "travel insurance demand" scenario tag is activated.

[0135] In this embodiment, by performing product promotion trigger scenario detection on the target product information, the specific scenarios that match the insurance product promotion can be accurately identified to ensure that subsequent product promotion plans and product information can be closely integrated with the current specific scenarios and user needs, thereby enhancing users' attention to and willingness to purchase insurance products, and achieving product promotion effects that meet personalized needs.

[0136] See also Figure 5 In some embodiments, step S106 includes but is not limited to steps S501 to S503:

[0137] Step S501 , obtaining product promotion influencing factors of target product information, and determining a product promotion plan category based on the product promotion influencing factors; the product promotion plan category includes at least one of a copy promotion plan category, an image promotion plan category, and a video promotion plan category.

[0138] Step S502 : constructing a plan to generate instruction information based on the product promotion plan category, product promotion cycle, and product promotion triggering scenario.

[0139] Step S503: Based on the plan generation instruction information, the pre-trained plan generation model is instructed to generate a promotion plan to obtain a promotion plan for the target product.

[0140] In step S501 of some embodiments, specifically, product promotion influencing factors refer to various factors that affect the product promotion effect, including but not limited to user information, user operation behavior data, product promotion cycle, product promotion triggering scenario, product type, product price, product guarantee content and product audience preferences, etc.

[0141] For example, in an insurance application scenario, factors influencing product promotion may include the coverage of the insurance product, premium level, user age, income level, and insurance products that the user is interested in.

[0142] Specifically, the product promotion plan category refers to the promotion form selected based on product characteristics, product value, plan production cost and product display needs. The product promotion plan category includes at least one of the copywriting promotion plan category, image promotion plan category and video promotion plan category.

[0143] Specifically, product features, product value, program production costs, and product display requirements are converted into evaluation matrices of four different dimensions, and the evaluation matrices of the four different dimensions are weighted and integrated to determine the integrated evaluation matrix, and the product promotion program category is determined based on the integrated evaluation matrix.

[0144] For example, for accident insurance products with relatively simple product coverage, the copywriting promotion plan category is given priority; for critical illness insurance products with more complex coverage, you can choose the image or video promotion plan category, so as to visually display the disease coverage through infographics, and combine it with short video situational dramas to help users understand the details of the terms; for product display needs, for young user groups who are accustomed to quick browsing, choose the copywriting promotion plan category; for users in the middle and old age groups, choose the graphic and text promotion plan category; for the plan production cost, you can determine the promotion plan category based on the expected product returns and the production costs of different forms of materials, and intelligently balance the promotion effect and input-output ratio; for insurance products with high premium value, such as whole life insurance, you can choose a high-cost video promotion plan with 3D animation commentary, and for low-profit products such as short-term accident insurance, you can use a copywriting product promotion plan.

[0145] In an optional embodiment of the present application, when evaluating the promotion scheme category, the product characteristics (such as product protection complexity), product value (such as product premium scale), scheme production cost (such as material production cost) and product display requirements (such as user preference) can also be quantified into a scoring matrix from 0 to 1. For example, the four dimensions of the target critical illness insurance product are scored as 0.8 (high insurance complexity), 0.9 (high premium value), 0.4 (medium to high material production cost), and 0.7 (image display required). After weighted fusion, the weights corresponding to these four dimensions are 0.4, 0.3, 0.2, and 0.1, respectively. The final comprehensive score of the product is 0.74, which is greater than 0.5 and less than 0.8, and belongs to the image promotion scheme threshold. Therefore, the product promotion scheme category is determined to be image. Among them, the threshold for the copywriting promotion scheme can be that the final comprehensive score of the product is less than 0.5, and the threshold for the video promotion scheme can be that the final comprehensive score of the product is greater than 0.8.

[0146] In this embodiment, by obtaining the product promotion influencing factors of the target product information and determining the product promotion plan category based on the product promotion influencing factors, it is possible to match the optimal promotion plan form according to different product characteristics and the needs of the target audience, and provide a variety of promotion plan options, which not only ensures the full display of high-value products, but also realizes the effective allocation of product promotion resources, helps to improve the targeted nature of product promotion content, so as to improve the promotion effect of subsequent product promotion plan generation.

[0147] In step S502 of some embodiments, specifically, the plan generation instruction information refers to specific instructions for guiding the generation of the product promotion plan based on the product promotion plan category, promotion cycle, and triggering scenario.

[0148] For example, if there are discussions on social media about health problems caused by high temperatures in summer and the Dragon Boat Festival is approaching, please write an advertising plan for promoting health insurance products with the title "Travel during the Dragon Boat Festival, health insurance will protect your journey."

[0149] See also Figure 6 In some embodiments, the solution generation instruction information includes at least one of text solution generation instruction information, graphic solution generation instruction information, and video solution generation instruction information. Step S503 includes but is not limited to steps S601 to S603:

[0150] Step S601: If the solution generation instruction information is a copy solution generation instruction information, then based on the copy solution generation instruction information, the solution generation model is instructed to generate a product copy promotion solution, obtain a product copy promotion solution, and determine the product copy promotion solution as a target product promotion solution.

[0151] Step S602: If the solution generation instruction information is a graphic and text solution generation instruction information, then the graphic and text solution generation instruction information instructs the solution generation model to generate a product graphic and text promotion solution, obtain a product graphic and text promotion solution, and determine the product graphic and text promotion solution as the target product promotion solution.

[0152] Step S603: If the solution generation instruction information is video solution generation instruction information, then generate a product video promotion solution based on the video solution generation instruction information instructing the solution generation model to obtain a product video promotion solution, and determine the product video promotion solution as the target product promotion solution.

[0153] See also Figure 7 In some embodiments, step S601 includes but is not limited to steps S701 to S703:

[0154] Step S701 : instructing a solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information to obtain a candidate product copy promotion solution.

[0155] Step S702: testing the promotion effect of the candidate product copy promotion plan to obtain promotion effect test data.

[0156] Step S703 , screening candidate product copy promotion plans based on the promotion effect test data to obtain a product copy promotion plan.

[0157] In step S701 of some embodiments, specifically, the copywriting plan generation instruction information refers to specific instructions for guiding the generation of a copywriting promotion plan, including but not limited to information such as the theme, style, and target audience of the copywriting.

[0158] Specifically, the candidate product copywriting promotion plan refers to multiple versions of preliminary copywriting promotion plans generated according to the copywriting plan generation instruction information.

[0159] Specifically, the copywriting plan generation instruction information can be input into a plan generation model (such as a Transformer model) to generate a product copywriting promotion plan in an autoregressive manner.

[0160] For example, in the insurance business scenario, when generating promotional copy for critical illness insurance products, the solution generation model outputs three candidate solutions: the first solution focuses on the coverage of critical illness insurance that users are interested in (such as covering 120 diseases), the second solution emphasizes the convenience of critical illness insurance (such as insurance can be completed in 3 minutes), and the third solution focuses on the after-sales service of claims for car insurance products that users have purchased (such as immediate compensation upon diagnosis, without the need for advance payment).

[0161] In this embodiment, by generating a product copy promotion plan based on the copy plan generation indication information indicating the plan generation model, not only can the automatic generation of product promotion plans be achieved, thereby improving the generation efficiency of product promotion plans, but also, based on the needs and preferences of different users for products, different promotion plan options can be provided to different users through different expression angles and emphases, thereby avoiding the expression limitations that may exist in a single plan.

[0162] In step S702 of some embodiments, specifically, the generated candidate copy promotion plan is actually tested in a small range, such as pushing the candidate copy promotion plan to the target audience through social media advertising, email marketing, etc., and collecting user feedback data (such as click-through rate, reading rate, conversion rate, etc.) as promotion effect test data.

[0163] For example, the generated candidate health insurance promotion copy can be published on the insurance mini program, and the user's click-through rate (click-through rate = number of clicks / number of publications) and conversion rate (conversion rate = product transaction volume / number of clicks) indicators can be counted.

[0164] In this embodiment, by testing the promotion effect of the candidate product copy promotion plans, it is possible to objectively reflect the impact of different expression methods on actual user decisions, which is helpful to subsequently improve the promotion effect of the product copy promotion plan.

[0165] In step S703 of some embodiments, specifically, the copy with the highest click-through rate and conversion rate is selected as the final product copy promotion plan.

[0166] In this embodiment, candidate product copy promotion plans are screened according to the promotion effect test data, so that the most effective product copy promotion plan can be screened out, thereby ensuring the effectiveness of the product promotion activities.

[0167] Through steps S701 to S703, by generating candidate copy promotion plans, conducting promotion effect tests, and screening copies based on the test data, it is possible to ensure that the final product copy promotion plan has the best promotion effect, help improve the quality of the product copy promotion plan, and ensure the promotion effect of this product promotion activity.

[0168] In step S602 of some embodiments, specifically, the graphic and text plan generation instruction information refers to specific instructions for guiding the generation of a graphic and text promotion plan, including but not limited to the theme of the image, the combination of the text and the image, etc.

[0169] Specifically, the core sales points of the copy (such as critical illness insurance covering 120 diseases) in the graphic and text scheme indication information are parsed through a solution generation model (such as a Transformer model). The image theme is a diagram showing the coverage of different insurance products of the same type of insurance. The cosine similarity between the image or chart and the core sales points is calculated from the preset corporate insurance industry visual library to find the target image that matches the core sales point. The target visual template is selected from the company's product visual model library and the target image is automatically typeset with the product copy promotion plan to generate a product graphic and text promotion plan.

[0170] For example, in an insurance application scenario, if the solution generation model (Transformer) recognizes that the graphic solution generation instruction information includes core requirements such as visualizing the coverage of 120 diseases and comparing the claims differences of insurance products under different insurance amounts, the solution generation model will extract key data from the product database (such as the incidence rate of each age group, the claim amount, etc.), and select the disease types and claims standard charts corresponding to the basic, mid-range and high-end products of critical illness insurance from the company's insurance industry visual library, and select the critical illness insurance visual template from the company's product visual template library. The product copy promotion plan containing the visualization of the coverage of 120 diseases and the comparison of the claims differences of insurance products under different insurance amounts will be automatically laid out with the chart through the critical illness insurance visual template to generate the final graphic and text promotion plan for critical illness insurance.

[0171] In this embodiment, the product graphic and text promotion plan is generated based on the graphic and text plan generation indication information indication plan generation model, which can not only organically combine images and text to generate a promotion plan with greater visual appeal and information communication effect, but also effectively improve the readability of product content, thereby improving the promotion effect of the product promotion plan generation.

[0172] In step S603 of some embodiments, specifically, the video plan generation instruction information refers to specific instructions for guiding the generation of a video promotion plan, including but not limited to the theme, script, visual effects, etc. of the video.

[0173] Specifically, the solution generation model may also be a visual LLM (Large Language Model) generation model.

[0174] Specifically, the video plan generation instruction information including the video theme, script, and visual effects is input into the visual LLM to generate a product storyboard script (such as converting the "cash flow advantage of annuity insurance" into a visual narrative of "dynamic line chart comparing pension savings and insurance returns"). The corresponding scene images, animation effects, and music styles are matched to the storyboard script. Combined with the core elements of the product extracted by the visual LLM, product promotion subtitles are generated, and each frame of the storyboard, audio track, and subtitles are synchronized to output the product video promotion plan.

[0175] In this embodiment, the product video promotion plan is generated based on the video plan generation indication information indication plan generation model, which can generate vivid and infectious product promotion videos, realize product promotion visualization, and more intuitively and generate product content and advantages, thereby improving the promotion effect of the product promotion plan generation.

[0176] Through steps S601 to S603, the scheme generation model is instructed to automatically generate personalized promotion schemes according to different categories of scheme generation indication information, which significantly improves the generation efficiency and quality of product promotion schemes, and provides a variety of corresponding text, graphic or video promotion schemes for different users and different product levels, significantly improving the promotion effect of the product promotion scheme.

[0177] Through steps S501 to S503, by comprehensively considering product characteristics, promotion cycles and triggering scenarios, accurate and diversified promotion plans can be generated. Not only can the most appropriate promotion form be selected according to product characteristics and the needs of the target audience, but also more user-targeted promotion content can be generated in combination with the promotion cycle and triggering scenarios, effectively solving the problem of not being able to meet the personalized needs of customers and significantly improving the promotion effect of the product promotion plan.

[0178] The embodiment of the present application first obtains the user information of the target user and the user operation behavior data of the target user in response to a product promotion request, thereby comprehensively understanding the personal information and behavioral preferences of the target user, providing basic data support for the subsequent generation of a personalized product promotion plan, and inferring the product needs of the target user based on the user information and the user operation behavior data. Based on the target user information, the embodiment can accurately infer the user's specific product needs, further match products that meet the user's needs from the product information library, and ensure that the content of the subsequently generated product promotion plan is highly relevant to the user's needs, thereby improving the promotion effect of the product promotion plan; secondly, the product promotion cycle and product promotion trigger scenario detection are performed on the target product information, and the specific period of the target product in the target user's decision-making process can be identified, facilitating the subsequent dynamic adjustment of the promotion strategy, and can also be associated with external events, making the subsequent promotion plan more scenario-based and persuasive, which helps to further improve the effect of the subsequent product promotion plan generation; finally, the target product promotion plan is generated based on the target product information, the product promotion cycle, and the product promotion trigger scenario, and can comprehensively consider the needs of the target user and the product characteristics to generate a personalized product promotion plan, effectively solving the problem of failing to meet the personalized needs of customers and significantly improving the promotion effect of the product promotion plan.

[0179] See also Figure 8 The present application also provides a product promotion plan generation device that can implement the above-mentioned product promotion plan generation method. The device includes:

[0180] A user data acquisition module is used to obtain user information of target users and user operation behavior data of target users in response to product promotion requests;

[0181] The product demand inference module is used to infer the product demand of target users based on user information and user operation behavior data to obtain product demand information;

[0182] The product matching module is used to match products from a pre-built product information database based on product demand information to obtain target product information;

[0183] A product promotion cycle detection module is used to detect the product promotion cycle of the target product information and obtain the product promotion cycle; wherein the product promotion cycle is used to represent the period of the target product in the decision-making process of the target user;

[0184] A product promotion trigger scenario detection module is used to detect product promotion trigger scenarios for target product information and obtain product promotion trigger scenarios; wherein the product promotion trigger scenario is used to represent the promotion context associated with the target product;

[0185] The product promotion plan generation module is used to generate a target product promotion plan based on target product information, product promotion cycle and product promotion trigger scenario.

[0186] The specific implementation of the product promotion plan generating device is basically the same as the specific embodiment of the above-mentioned product promotion plan generating method, and will not be repeated here.

[0187] The present application also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned product promotion plan generation method. The electronic device can be any smart terminal including a tablet computer, an in-vehicle computer, or the like.

[0188] See also Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0189] The processor 901 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an 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;

[0190] 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 the processing system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the product promotion plan generation method of the embodiments of this application;

[0191] Input / output interface 903, used to implement information input and output;

[0192] Communication interface 904, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0193] Bus 905 , which transmits information between various components of the device (e.g., processor 901 , memory 902 , input / output interface 903 , and communication interface 904 );

[0194] 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 .

[0195] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned product promotion plan generating method when executed by a processor.

[0196] 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 arranged 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.

[0197] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this 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 this application are also applicable to similar technical problems.

[0198] Those skilled in the art will understand 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.

[0199] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0200] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0201] 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 sequential order. 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 variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes 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.

[0202] It should be understood that in this 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 previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items 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.

[0203] In the several embodiments provided in this 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 merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, 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.

[0204] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0205] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0206] 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, which is stored in a storage medium and includes multiple instructions for enabling 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: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0207] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for generating a product promotion plan, characterized in that: The method comprises: In response to a product promotion request, obtaining user information of a target user and obtaining user operation behavior data of the target user; Inferring product demand of the target user based on the user information and the user operation behavior data to obtain product demand information; Based on the product demand information, product matching is performed from a pre-built product information database to obtain target product information; Performing product promotion cycle detection on the target product information to obtain the product promotion cycle; Performing product promotion triggering scenario detection on the target product information to obtain a product promotion triggering scenario; A target product promotion plan is generated based on the target product information, the product promotion cycle, and the product promotion triggering scenario.

2. The method according to claim 1, characterized in that The generating of the target product promotion plan based on the target product information, the product promotion cycle and the product promotion triggering scenario includes: Obtain product promotion influencing factors of the target product information, and determine a product promotion plan category based on the product promotion influencing factors; the product promotion plan category includes at least one of a copywriting promotion plan category, an image promotion plan category, and a video promotion plan category; Constructing a plan based on the product promotion plan category, the product promotion cycle, and the product promotion triggering scenario to generate instruction information; Based on the solution generation indication information, a pre-trained solution generation model is instructed to generate a promotion solution to obtain the target product promotion solution.

3. The method according to claim 2, characterized in that The scheme generation instruction information includes at least one of text scheme generation instruction information, graphic scheme generation instruction information, and video scheme generation instruction information; The step of instructing the pre-trained solution generation model based on the solution generation instruction information to generate a promotion solution to obtain the target product promotion solution includes: If the solution generation instruction information is copy solution generation instruction information, then instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information, obtaining a product copy promotion solution, and determining the product copy promotion solution as the target product promotion solution; If the solution generation instruction information is a graphic and text solution generation instruction information, then instructing the solution generation model to generate a product graphic and text promotion solution based on the graphic and text solution generation instruction information to obtain a product graphic and text promotion solution, and determining the product graphic and text promotion solution as the target product promotion solution; If the solution generation indication information is video solution generation indication information, the solution generation model is instructed to generate a product video promotion solution based on the video solution generation indication information to obtain a product video promotion solution, and the product video promotion solution is determined as the target product promotion solution.

4. The method according to claim 3, characterized in that The step of instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information to obtain the product copy promotion solution includes: Instructing the solution generation model to generate a product copy promotion solution based on the copy solution generation instruction information to obtain a candidate product copy promotion solution; Conducting promotion effect testing on the candidate product copy promotion plan to obtain promotion effect test data; The candidate product copy promotion plans are screened according to the promotion effect test data to obtain the product copy promotion plan.

5. The method according to claim 1, wherein The inferring product demand of the target user based on the user information and the user operation behavior data to obtain product demand information includes: Performing user level identification on the target user according to the user information to obtain the target user level; Performing behavior preference detection on the user operation behavior data to obtain target user interest tags; The product demand of the target user is analyzed according to the target user level and the target user interest tag to obtain the product demand information.

6. The method according to claim 5, characterized in that The performing behavior preference detection on the user operation behavior data to obtain the target user interest tag includes: Extracting features from the user operation behavior data to obtain user operation behavior features; the user operation behavior features are time series features; Performing behavioral interest analysis on the user operation behavior characteristics to obtain the target user interest tag.

7. The method according to any one of claims 1 to 6, characterized in that Before performing product matching from a pre-built product information database based on the product demand information to obtain target product information, the method further includes: Acquire initial product information; the initial product information includes structured product information and unstructured product information; Performing entity recognition on the structured product information to obtain structured entity information, and performing entity recognition on the unstructured product information to obtain unstructured entity information; Performing entity relationship extraction on the structured entity information to obtain structured entity relationships, and performing entity relationship extraction on the unstructured entity information to obtain unstructured entity relationships; Embedding the structured entity information and the structured entity relationship into a knowledge graph to obtain a structured knowledge base, and embedding the unstructured entity information and the structured entity relationship into a knowledge graph to obtain an unstructured knowledge base; The product information base is determined based on the structured knowledge base and the unstructured knowledge base.

8. A device for generating a product promotion plan, characterized in that: The device comprises: A user data acquisition module, configured to obtain user information of a target user and user operation behavior data of the target user in response to a product promotion request; A product demand inference module is used to infer the product demand of the target user based on the user information and the user operation behavior data to obtain product demand information; A product matching module is used to match products from a pre-built product information database based on the product demand information to obtain target product information; A product promotion cycle detection module is used to perform product promotion cycle detection on the target product information to obtain a product promotion cycle; wherein the product promotion cycle is used to represent the period of the target product in the decision-making process of the target user; A product promotion triggering scenario detection module is used to perform product promotion triggering scenario detection on the target product information to obtain a product promotion triggering scenario; wherein the product promotion triggering scenario is used to represent the promotion context associated with the target product; The product promotion plan generation module is used to generate a target product promotion plan based on the target product information, the product promotion cycle and the product promotion triggering scenario.

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 promotion plan generating 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 promotion plan generating method according to any one of claims 1 to 7 is implemented.

Images