Smart scene type marketing system and method

Abstract

The invention discloses a smart scene type marketing system and method, and the system comprises an ecological library module which is used for providing a user ecological settlement function supporting information input, autonomously positioning the industry ecology to which a settled user belongs, and searching an ecological partner; the knowledge base module is used for storing industry knowledge assets in a scene mode and supporting an intelligent retrieval function to achieve scene query of the industry knowledge assets; the industry policy module is used for collecting industry policy information and constructing scene associated policy information; the data center module is used for providing a multi-dimensional data fusion analysis model constructed based on a deep learning algorithm so as to obtain a product or a scheme meeting user scene requirements; the demand analysis module is used for mining user scene demands and training to obtain a user demand prediction model, and predicting current user scene demands; and by using the multi-dimensional data fusion analysis model, obtaining a product or a scheme meeting the user scene requirement and pushing the product or the scheme. According to the invention, the user can effectively participate in user precision marketing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent marketing, in particular to a smart scene marketing system and method. BACKGROUND

[0002] With the rapid development of Internet technology, online marketing has occupied a dominant position in promoting products and services of enterprises. Online marketing can break through geographical restrictions and quickly reach a large number of potential users with its wide coverage and strong dissemination capacity. It saves a lot of marketing costs for enterprises, enabling them to obtain higher returns with lower investment.

[0003] In the traditional online marketing mode, it mainly relies on a variety of single-dimensional promotion methods. Commonly seen are based on user behavior data analysis, collecting user browsing records, purchase history and other data to infer user interests and needs, and then pushing advertisements; using search engine optimization (SEO) technology to improve the ranking of enterprise websites in search engines and increase website traffic; placing advertisements on social media platforms to expand the brand influence of enterprises with the large user base of social media; and through email marketing, sending product and service information to potential users. These means can help enterprises promote products and services to a certain extent, but they are independent of each other and lack systematicness and comprehensiveness.

[0004] However, the traditional online marketing mode has obvious defects. On the one hand, when building user portraits, it is mostly based on static historical data such as browsing records and purchase history, lacking real-time scene perception ability of users, resulting in low matching degree between advertisement pushing and actual needs of users, making the conversion efficiency of marketing very low; on the other hand, in the information input process, users need to face complex information input work, and cannot distinguish their own field and specific needs, and there is no effective reference case to refer to. SUMMARY

[0005] In order to realize effective participation of users in user precision marketing, the present application provides a smart scene marketing system and method.

[0006] In the first aspect, the present application provides a smart scene marketing system, comprising: an ecological library module, a knowledge base module, an industry policy module, a data center module and a demand analysis module; The ecological library module is used to provide user ecological settlement function supporting information input and independently locate the industry ecology of the settled user, access third-party enterprise management information platform, and find user ecological partners matched with the user according to industry affiliation and industry preset scene; The knowledge base module is configured to manage and store industry knowledge assets in a structured manner and in a scenario, support intelligent retrieval function to realize scenario query of industry knowledge assets, and the industry knowledge assets include industry market pain points, product or scheme advantages, user cases and evaluations. The industry policy module is configured to collect industry policy information and build scenario-related policy information. The data center module is configured to provide a multi-dimensional data fusion analysis model built based on a deep learning algorithm to obtain a product or scheme meeting the user scenario demand, and the multi-dimensional data analysis model includes ecological library data, knowledge base data and industry policy data. The demand analysis module is configured to obtain historical user ecological library data and products or schemes applied by users in different scenarios, mine user scenario demand by using a deep learning algorithm, train a user demand prediction model, predict the current user scenario demand by using the user demand prediction model, and obtain a product or scheme meeting the user scenario demand by using the multi-dimensional data fusion analysis model and push the product or scheme.

[0007] By using the above scheme, the ecological library module helps users accurately match the industry ecology and find ecological partners, the knowledge base module structurally manages and scenario queries industry knowledge assets, the industry policy module collects and builds scenario-related policy information to provide industry insights for enterprises, the data center module builds an analysis model based on a deep learning algorithm to assist in obtaining a product or scheme meeting the user scenario demand, and the demand analysis module mines user scenario demand, trains a user demand prediction model, predicts scenario demand, and pushes a product or scheme meeting the user scenario demand to realize precise marketing.

[0008] Preferably, the method further comprises: The scene interaction module is configured to create a scene and a user portrait in the scene by using 3D modeling and VR technology, create a relationship between the scene and the product or scheme based on the ecological library data, the knowledge base data and the industry policy data, receive a user-selected scene, automatically push a product or scheme according to the selected scene, and mine user scenario demand by using a deep learning algorithm according to a scene interaction process of the pushed product or scheme selected by the user.

[0009] By using the above scheme, considering that the prior art lacks deep interaction with the user scenario and is difficult to stimulate the willingness of the user to actively participate in the marketing activity, the 3D modeling and VR technology are used to create a scene and a user portrait in the scene, which can provide an immersive experience for the user and facilitate mining of user demand according to the industry scenario and improvement of purchase or project decision-making.

[0010] Preferably, the demand analysis module is further configured to directly receive a scenario demand published by a user, query a matching user ecological partner, generate an invitation corresponding thereto, and push the invitation to the user ecological partner through multiple channels; and receive an ecological response including a product or a scheme sent by the user ecological partner and push the ecological response back to the user.

[0011] By using the above scheme, the demand published by the user is directly received, the ecological recommendation is generated by querying and matching to the user ecological partner, the ecological recommendation is pushed to the user ecological partner through multiple channels, the ecological response of the ecological partner is received, and the product or the scheme is pushed back to the user, so that the communication between the user and the ecological partner is more efficient, the demand transmission is more direct and accurate, the possibility of the product or the scheme satisfying the user demand is improved, and the practicability and effectiveness of the marketing system are enhanced.

[0012] Preferably, the demand analysis module is further configured to perform full-process management of demand investigation, demand review, project tracking, and single generation process on the predicted current user scenario demand.

[0013] By using the above scheme, the full-process management from the demand investigation to the single generation process is performed on the predicted current user scenario demand, the user demand and the business process are finely controlled, and the marketing effect and the success rate are improved.

[0014] Preferably, the data center module is further configured to monitor an accuracy rate of demand review corresponding to the predicted scenario demand and a success rate of single generation in the full-process management of each user, determine whether the accuracy rate of demand review corresponding to the predicted scenario demand is greater than a preset review accuracy rate, when the accuracy rate is determined to be not greater than the preset review accuracy rate, optimize and train the user demand prediction model in an incremental training manner until the preset review accuracy rate is reached, determine whether the success rate of single generation is greater than a preset single generation success rate, when the success rate is determined to be not greater than the preset single generation success rate, optimize and train the multi-dimensional data fusion analysis model in the incremental training manner until the preset single generation success rate is reached.

[0015] By using the above scheme, the accuracy rate of demand review corresponding to the predicted scenario demand and the success rate of single generation in the full-process management of each user are monitored, for the case that the demand review accuracy rate or the single generation success rate is not up to standard, the user demand prediction model or the multi-dimensional data fusion analysis model is optimized and trained in the incremental training manner, and the prediction accuracy of the system and the success rate of marketing are improved.

[0016] Preferably, the scene interaction module is configured to create a user customized scene and a user portrait in the customized scene by using 3D modeling and VR technology, create a relationship between the customized scene and a product or a scheme, automatically push the product or the scheme according to the customized scene, mine user scene demand by using a deep learning algorithm according to a scene interaction process of the pushed product or scheme selected by the user, and the demand analysis module is further configured to obtain historical user ecological database data and a product or a scheme applied by the user in the customized scene, mine user scene demand by using a deep learning algorithm, and supplement training to obtain a user demand prediction model.

[0017] By using the above scheme, the user can customize a scene, create a user customized scene and a portrait, establish a relationship between the customized scene and a product or a scheme, automatically push the product or the scheme, and mine user scene demand through a scene interaction process, thereby providing personalized experience for the user and improving marketing accuracy. The historical user and the product or the scheme applied in the customized scene are used for supplement training to mine user demand, predict current user scene demand, obtain and push a product or a scheme meeting the demand, and further improve the matching degree of the product or the scheme and the user demand.

[0018] Preferably, the AI assistant module is configured with a user input unit and an intelligent semantic analysis unit, configured to analyze user input information by using the intelligent semantic analysis unit, and find user ecological partners, knowledge assets, and industry policies based on an industry preset scene. Meanwhile, a visual display interface is set to display the user ecological partners, the knowledge assets, and the industry policies corresponding to the industry preset scene found. The smart exhibition hall module is configured to integrate and uniformly manage the results of the predicted scene demand of each user and the whole-process management of demand investigation, demand review, project tracking, and order forming, and display the results.

[0019] By using the above scheme, the AI assistant module can analyze user input content, find related information in a scene mode, and input an analysis model to assist in obtaining and pushing a product or a scheme meeting user scene demand, while displaying the found information for the user to intuitively understand. The smart exhibition hall module can integrate and uniformly manage the predicted scene demand of the user and the whole-process management results and display them, thereby realizing centralized processing and display of information.

[0020] In a second aspect, the present application provides a smart scene marketing method, comprising: predicting current user scene demand by using a user demand prediction model; monitoring user ecological entry and positioning an industry ecology to which an entered user belongs by using the ecological library module, accessing a third-party enterprise management information platform, finding user ecological partners matched with the user through industry attribution and an industry preset scene, querying scene-based industry knowledge assets by using an intelligent search function of the knowledge base module, and collecting scene-related policy information by using the industry policy module. In combination with the user ecological partner, the scenario-based industry knowledge asset, and the scenario-related policy information, a multi-dimensional data fusion analysis model is used to obtain a product or a scheme meeting the user scenario demand and to push the product or the scheme.

[0021] By using the above scheme, the user demand prediction model can be used to predict the user's current scenario demand, and the matching degree and conversion efficiency of the advertisement pushing and the user's current demand can be improved. The ecological library module can be used to locate the industry ecology of the user and to match the ecological partner. The knowledge base module can be used to query the scenario-based industry knowledge asset. The industry policy module can be used to collect the scenario-related policy information. In combination with the multi-party information, the multi-dimensional data fusion analysis model is used to obtain a product or a scheme meeting the user scenario demand and to push the product or the scheme, so as to realize accurate marketing.

[0022] In a third aspect, a computer readable storage medium is provided, which includes a stored computer program, wherein the computer program, when executed, controls a device where the computer readable storage medium is located to perform the method described above.

[0023] In a fourth aspect, a computer device is provided, which includes a memory, a processor, and a program stored in the memory and executable by the processor, and the program, when executed by the processor, implements the steps of the method described above.

[0024] In summary, the present application has the following advantages: 1. The ecological library module is designed to match the user ecology, and the information filling and the scenario-based user ecological partner matching are realized by accessing the three-party enterprise management information platform. The knowledge base module is designed to structurally manage the industry knowledge asset, so that the user can understand the corresponding industry market pain points, product or scheme advantages, and the like in different scenarios, to assist subsequent accurate marketing in combination with the knowledge asset. The demand analysis module is designed to predict and manage the demand based on the user demand prediction model, and to push a product or a scheme meeting the user scenario demand in combination with the multi-dimensional data fusion analysis model, so as to improve the marketing conversion efficiency. 2. Considering that the existing technology lacks deep interaction with the user scenario and is difficult to stimulate the user's willingness to actively participate in the marketing activity, the scenario interaction module is designed to provide immersive experience by using 3D modeling and VR technology, to carry out scenario marketing, to mine the user scenario demand, and to solve the problems of lack of deep interaction with the user scenario in the traditional mode and low matching degree of the advertisement pushing and the user's current demand. 3. The demand analysis model is designed to directly receive the user's published scenario demand, to query the matched user ecological partner, to push to the user ecological partner through multiple channels, to receive the ecological response of the ecological partner and to push the product or the scheme back to the user, so that the communication between the user and the ecological partner is more efficient, the demand transmission is more direct and accurate, the possibility of the product or the scheme meeting the user demand is improved, and the practicality and effectiveness of the marketing system are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the structure of the smart scenario-based marketing system described in a specific embodiment; Figure 2 This is a flowchart of the smart scenario-based marketing method described in a specific embodiment. DETAILED DESCRIPTION

[0026] 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.

[0027] This application mainly adopts the construction of a multi-module intelligent scenario-based marketing system and application method, which achieves the effects of accurate matching, predicting user needs, and improving marketing efficiency. The following is a further detailed description of this application.

[0028] Example 1 like Figure 1 As shown, the embodiment of the present application discloses an intelligent scenario-based marketing system, including an ecological library module 1, a knowledge base module 2, an industry policy module 3, a data center module 4 and a demand analysis module 5; wherein, each module cooperates with each other and communicates data, and through the integration of multiple aspects of information, it is possible to accurately predict user scenario needs and effectively push products or solutions, thereby improving the pertinence and conversion rate of marketing under different scenario conditions, and solving the problems of low demand matching and low conversion efficiency in traditional marketing models.

[0029] Specifically, the ecological library module 1 includes a user settlement submodule 11 and an ecological partner search submodule 12. Among them, the user settlement submodule 11 is used to provide a user ecological settlement function that supports information entry, which is specifically presented in the form of a registration page. In order to facilitate user operation and reference, an information classification guidance input page is designed, which is divided into basic registration information, investment information, bidding data, social media behavior data, device logs and other user ecological data; and a template is provided for each type of information classification for user reference. Among them, the filling of settlement information can also adopt the interface docking method to directly obtain user information from other existing data sources to avoid repeated entry by users.

[0030] The user registration submodule 11 is used to automatically locate the industry ecosystem to which the registered user belongs and perform labeling processing using natural language analysis technology after the user fills in relevant information on the registration page. For example, for a technology company, it can accurately identify that it belongs to the technology industry ecosystem; The ecological partner finding submodule 12 is configured to access a third-party enterprise management information platform, the platform stores a large amount of enterprise related information, and finds a user ecological partner matched with the user according to industry attribution and industry preset scenes. Specifically, based on the industry attribution and the industry preset scene (i.e., the industry conventional scene setting for the registered user) of the registered user, the graph technology is used to mine the ecological partners associated with the user, including the upstream and downstream associated enterprises; for example, if the registered user is in the environmental protection industry, in the preset environmental protection project scene, the upstream and downstream enterprises matched with the registered user are quickly found as ecological partners, the ecological partner query capability is set to be visualized, and the "my ecology" common management is added.

[0031] The knowledge base module 2 includes a knowledge management submodule 21 and an intelligent retrieval submodule 22, etc.; the knowledge management submodule 21 is configured to manage industry knowledge assets in a structured manner (data is structured), specifically including storing the pre-collected industry market pain point information, the advantages of existing products / schemes, technical architecture, the written associated list (such as the association between products and different industry preset scene demands), user cases and evaluations, case application values, etc. in a scene classification manner, so as to facilitate subsequent query and use; and a knowledge updating mechanism is established to update the knowledge base content regularly; The intelligent retrieval submodule 22 is configured to provide a retrieval column and receive a scene keyword input by a user, to quickly realize scene-based query of industry knowledge assets; in addition, in order to provide better industry knowledge assets information for the user to assist in obtaining more accurate marketing decisions, an associated intercommunication function is provided to recommend similar knowledge base content and application scenes, and to obtain knowledge assets associated with similar industries and scenes.

[0032] The industry policy module 3 includes an information collection submodule 31 and an association construction submodule 32, etc.; the information collection submodule 31 is configured to collect industry policy information, and automatically crawls the latest industry policy from authoritative channels such as government websites and industry association official websites through network crawler technology.

[0033] The association construction submodule 32 is configured to associate the collected policy information with industry preset scenes, and construct scene-associated policy information, for example, the subsidy policy for the new energy automobile industry is associated with specific scenes such as new energy automobile promotion and research and development. In addition, the industry policy module 3 further includes a policy pushing submodule 33 configured to receive a user policy information content subscription pushing service.

[0034] The data center module 4 comprises a model construction sub-module and the like for data intercommunication between the module itself and other modules; the model construction sub-module is configured to provide a plurality of single-dimension data analysis sub-models constructed based on a deep learning algorithm to perform single-dimension data analysis and support marketing decision-making; the model construction sub-module is also configured to provide a multi-dimension data fusion analysis model constructed based on deep learning to obtain a product or a scheme meeting a user scenario requirement; the multi-dimension data analysis model comprises ecological library data, knowledge base data and industry policy data; the single-dimension data analysis model refers to data analysis of any dimension data in the ecological library data, the knowledge base data and the industry policy data; The plurality of single-dimension data analysis sub-models constructed include a user analysis sub-model and a knowledge asset analysis sub-model and the like; the user analysis sub-model is mainly based on user data in the ecological library data to analyze user preference information for a product or a scheme in different scenarios; the knowledge asset analysis sub-model is mainly based on knowledge asset content in the knowledge base data to analyze a user preference group of different products or schemes and determine a user preference group ecology; the multi-dimension data fusion analysis model is mainly to obtain a product or a scheme meeting a user scenario requirement according to a user scenario requirement, in combination with user ecological partners, scenario-based industry knowledge assets and scenario-related policy information queried based on the ecological library data, the knowledge base data and the industry policy data; input of the model is a user scenario requirement (including user direct input or predicted user scenario requirement), user ecological partners, scenario-based industry knowledge assets and scenario-related policy information queried based on the scenario requirement; output is a product or a scheme meeting the user scenario requirement; the model is trained by historical user scenario requirement, user ecological partners, scenario-based industry knowledge assets, scenario-related policy information and the product or the scheme meeting the user scenario requirement.

[0035] The demand analysis module 5 includes a historical data acquisition submodule 51, a model training submodule 52, and a demand prediction pushing submodule 53, and a product or scheme acquisition submodule 54; the historical data acquisition submodule 51 is used to acquire historical user ecological library data and products or schemes applied by users in different scenarios, and these data are derived from historical records of the system and other related data sources; the model training submodule 52 is used to mine the acquired data, analyze user scenario demand, and train a user demand prediction model; the user demand prediction model is a neural network model constructed by using deep learning technology, the input is historical user ecological library data, products or schemes applied by historical users in different scenarios, and the output is user scenario demand; the demand prediction pushing submodule 53 is used to predict the current user scenario demand by using the trained user demand prediction model, that is, to predict the direction and time node of future scenario demand of the user by using current user ecological data, products or schemes corresponding to the historical user in the corresponding scenario; the product or scheme acquisition submodule 54 is used to combine the predicted user scenario demand, query user ecological partners, scenario-based industry knowledge assets, and scenario-related policy information from ecological library data, knowledge base data, and industry policy data, acquire products or schemes meeting the demand by using a multi-dimensional data fusion analysis model, and push the products or schemes to the user.

[0036] The implementation principle of the embodiment is that: through the cooperative work of various modules, the system integrates data resources in multiple aspects such as ecology, knowledge, and policy; the ecological library module provides extensive ecological partner information, the knowledge base module accumulates rich industry knowledge, the industry policy module introduces policy guidance, the data center module performs deep analysis on these data through advanced algorithms, and the demand analysis module focuses on mining user demand; the modules cooperate with each other to form an organic whole, analyze and predict user scenario demand from different angles, and finally realize accurate pushing of products or schemes, greatly improve the effect and conversion rate of marketing, and make up for the shortcomings of the traditional marketing mode.

[0037] Embodiment 2 The embodiment is different from the above-mentioned embodiment 1 in that, in order to enhance the interactivity of users and marketing decisions, further improve the accuracy of product or scheme pushing, and improve the participation and satisfaction of users, the system further includes a scenario interaction module 6.

[0038] Specifically, the scene interaction module 6 includes a scene creation submodule 61 and an interaction analysis submodule 62, etc. The scene interaction module 61 is configured to create a scene and a user portrait in the scene by using 3D modeling and VR technology; analyze the correlation between the scene and a product or a scheme under the user portrait of the scene based on ecological library data, knowledge base data, and industry policy data; create a relationship between the scene and the product or the scheme; and automatically push the product or the scheme according to the selected scene and the relationship between the scene and the product or the scheme after the user selects the scene in the virtual scene.

[0039] The interaction analysis submodule 62 is configured to mine user scene demand by using a deep learning algorithm according to a scene interaction process of the pushed product or scheme selected by the user. For example, the user's preferences and demands are obtained by analyzing the user's stay time and operation behavior in the virtual scene.

[0040] The implementation principle of the embodiment is that the scene interaction module provides a more real and intuitive marketing experience for the user by introducing 3D modeling and VR technology, and enhances the interactivity between the user and the marketing scene. By analyzing the interaction behavior of the user in the virtual scene, the potential demand of the user can be more deeply mined, more accurate data support can be provided for the demand analysis module, and the accuracy of product or scheme pushing can be further improved.

[0041] In addition, considering that some users have customization needs for scene interaction, to meet the user needs, the scene interaction module 6 is also configured to receive user customization scene demand, create a user customization scene and a user portrait in the customization scene by using 3D modeling and VR technology, create a relationship between the customization scene and a product or a scheme based on ecological library data, knowledge base data, and industry policy data; automatically push the product or the scheme according to the customization scene, mine user scene demand by using a deep learning algorithm according to a scene interaction process of the pushed product or scheme selected by the user. Correspondingly, the demand analysis module 5 is also configured to obtain historical user ecological data and a product or a scheme applied by the user in the customization scene, mine user scene demand by using a deep learning algorithm, and supplement training to obtain a user demand prediction model; then predict the current user scene demand by using the user demand prediction model, obtain a product or a scheme satisfying the user scene demand by using a multi-dimensional data fusion analysis model, and push the product or the scheme.

[0042] Embodiment 3 The embodiment is different from the above-mentioned embodiment 1 in that, considering that there is a case where a user directly publishes demand, to better assist marketing decision-making, the demand analysis module 5 in the system further adds a demand publishing processing submodule 55 and a process management submodule 56.

[0043] The demand publishing processing submodule 55 is configured to receive the user published scene demand based on the set user scene demand input interface, obtain the user ecological partner matched with the user from the ecological library data, generate an invitation corresponding to the user ecological partner, and push the invitation to the user ecological partner through multiple channels; for example, a pharmaceutical enterprise user belongs to the medical industry, receives the user published scene demand, and obtains the allowed selling object in the seasonal disease outbreak scene. According to the user published scene demand, the downstream selling drug object matched is queried, the cooperation invitation is generated corresponding to the selling drug object, and the cooperation invitation is pushed to the downstream selling drug object through multiple channels such as short message and video recommendation. The demand publishing processing submodule 55 is configured to receive the ecological response including the product or scheme sent by the user ecological partner and push the ecological response back to the user; that is, when the user ecological partner receives the cooperation invitation, the scene demand of the cooperation user is determined according to the cooperation invitation, the product or scheme that the opposite party may accept is generated as the ecological response, and the ecological response is pushed back to the original user, so that the whole marketing process is improved, from the publishing of the user demand to the recommendation of the ecological partner, to the whole process processing of the subsequent cooperation, forming a closed service system.

[0044] Correspondingly, considering that there is a whole process processing from the user demand publishing to the cooperation invitation to the subsequent cooperation, the whole process needs to be managed, and therefore the process management submodule 56 is designed. The process management submodule 56 is configured to perform whole process management of demand investigation, demand review, project tracking and single flow process on the predicted current user scene demand. Specifically, in the demand investigation stage, considering the existence of multiple scene demands published or predicted, the knowledge base data is combined, the scene demand obviously not meeting the scene is analyzed and screened, and the scene demand is fed back to the user input interface; in the demand review stage, when the scene demand obviously not meeting the scene is fed back to the user input interface, the user is reminded to confirm the scene demand, and the scene demand fed back by the user after the review is received; in the project tracking stage, the progress of the user cooperation invitation is monitored, whether the ecological response fed back by the ecological partner is received and the user is timely reminded; in the single flow process stage, the whole process of the user and the ecological partner reaching cooperation and signing the cooperation contract for the product or scheme is monitored in real time, and the progress is fed back in time to ensure that the transaction is completed smoothly.

[0045] In addition, the demand analysis module 5 is also used for judging whether the current scenario marketing better meets the user demand according to the accuracy rate of corresponding demand review of the predicted scenario demand and the success rate of order forming in the whole process management, as feedback data to correspondingly adjust the system to generate more accurate marketing strategy (product or scheme).

[0046] Embodiment 4 The difference between this embodiment and the above-mentioned embodiments is that, in order to further provide the user with a more convenient and efficient information query and acquisition method, and realize effective management and optimization of marketing activities, the system Further comprises an AI assistant module 7 and a smart exhibition hall module 8.

[0047] The AI assistant module 7 comprises an input analysis sub-module 71 and an information display sub-module 72.

[0048] The input analysis sub-module 71 is provided with a user input unit and an intelligent semantic analysis unit, which analyzes the user input information by using the intelligent semantic analysis unit, such as a series of ecological data or demand description input by the user, accurately understands the meaning by using the intelligent semantic analysis unit, and finds the user's ecological partners, knowledge assets and industry policies based on the industry preset scene, to assist subsequent use of the multi-dimensional data fusion analysis model to obtain products or schemes meeting the user's scene demand and push them.

[0049] The information display sub-module 72 is provided with a visual display interface, which visually displays the found ecological partners, knowledge assets and industry policies to the user, facilitating the user to view and select.

[0050] The smart exhibition hall module 8 comprises a result integration sub-module 81 and a display sub-module 82.

[0051] The result integration sub-module 81 is used for integrating and uniformly managing the results of the whole process management of the predicted scenario demand and the corresponding demand investigation, demand review, project tracking and order forming process of each user. The display sub-module 82 displays the integrated results in a clear and easy-to-understand way (such as structured data form), which provides strong support for the marketing decision of enterprises.

[0052] For example,Figure 2 As shown, the embodiment provides a smart scene marketing method, and the specific steps include: S1, predicting the current user scene demand by using a pre-trained user demand prediction model.

[0053] S2, monitoring the user's entry by using an ecological library module, positioning the industry ecology to which the user belongs, accessing a third-party enterprise management information platform, finding a user ecological partner matched with the user through industry attribution and industry preset scene, completing the query of scene-based industry knowledge assets by using the intelligent retrieval function of the knowledge base module, and collecting scene-related policy information by using the industry policy module.

[0054] S3, combining the user ecological partner, scene-based industry knowledge assets, and scene-related policy information, and obtaining a product or scheme meeting the user scene demand by using a multi-dimensional data fusion analysis model and pushing it.

[0055] In addition, S1 in the method further includes: directly receiving the user's published scene demand; S4, querying the matched user ecological partner, generating an invitation corresponding to the user ecological partner and pushing it to the user ecological partner through multiple channels, and receiving the ecological response including the product or scheme sent by the user ecological partner and pushing it back to the user.

[0056] In one specific embodiment, the method further includes: receiving the user's selected scene by using a scene interaction module, automatically pushing the product or scheme according to the selected scene, and mining the user's scene demand by using a deep learning algorithm according to the scene interaction process including the pushed product or scheme selected by the user.

[0057] In one specific embodiment, the method further includes: performing full-process management of demand investigation, demand review, project tracking, and single formation process on the predicted current user scene demand.

[0058] In one specific embodiment, the method further includes: monitoring the accuracy rate of the predicted scene demand corresponding to the demand review and the success rate of the single formation in the full-process management process of each user; determining whether the accuracy rate of the predicted scene demand corresponding to the demand review is greater than a preset review accuracy rate, and when the determination is not greater than the preset review accuracy rate, optimizing and training the user demand prediction model in an incremental training manner until the preset review accuracy rate is reached; determining whether the success rate of the single formation is greater than a preset single formation success rate, and when the determination is not greater than the preset single formation success rate, optimizing and training the multi-dimensional data fusion analysis model in an incremental training manner until the preset single formation success rate is reached.

[0059] In a specific embodiment, the method further comprises: automatically pushing a product or a solution according to the customized scenario by using the scenario interaction module, mining the user scenario demand by using a deep learning algorithm according to a scenario interaction process including the pushed product or solution selected by the user, and obtaining a user demand prediction model by mining the user scenario demand and supplementing training by using a deep learning algorithm based on historical user ecological data and the product or solution applied by the user in the customized scenario.

[0060] In a specific embodiment, the method further comprises: analyzing user input ecological data by using an AI assistant module, and displaying a user ecological partner, knowledge asset, and industry policy based on a preset industry scenario.

[0061] Embodiments of the present application also disclose a computer readable storage medium.

[0062] Specifically, the computer readable storage medium stores a computer program capable of being loaded and executed by a processor to implement the above-mentioned intelligent scenario marketing method, and the computer readable storage medium includes various program code storage media such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0063] Embodiments of the present application also disclose a computer device.

[0064] Specifically, the computer device includes a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to implement the above-mentioned intelligent scenario marketing method.

[0065] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in the specification (including the abstract and the drawings) can be replaced by other equivalent or similar features unless specifically described. That is, each feature is only an example of a series of equivalent or similar features unless specifically described.

Claims

1. A smart scenario marketing system, characterized by: include: Ecological library module, knowledge base module, industry policy module, data center module and demand analysis module; The ecosystem library module is used to provide a user ecosystem entry function that supports information entry and independently locates the industry ecosystem to which the settled user belongs; Access third-party enterprise management information platforms to find user ecosystem partners that match users based on industry affiliation and industry preset scenarios; The knowledge base module is used to manage and store industry knowledge assets in a structured manner and in a scenario-based manner, supporting intelligent retrieval functions to achieve scenario-based queries on industry knowledge assets; the industry knowledge assets include: industry market pain points, product or solution advantages, user cases and evaluations; The industry policy module is used to collect industry policy information and construct scenario-related policy information; The data center module communicates data with other modules outside of itself to provide a multi-dimensional data fusion analysis model built based on deep learning algorithms to obtain products or solutions that meet user scenario needs; the multi-dimensional data analysis model includes ecological library data, knowledge base data, and industry policy data; The demand analysis module is used to obtain historical user ecological library data and products or solutions used by users in different scenarios, use deep learning algorithms to explore user scenario needs and train to obtain a user demand prediction model; use the user demand prediction model to predict current user scenario needs; use a multi-dimensional data fusion analysis model to obtain products or solutions that meet user scenario needs and push them.

2. The intelligent scenario marketing system according to claim 1 is characterized in that: Also includes: The scenario interaction module is used to use 3D modeling and VR technology to create scenarios and user portraits in the scenarios, and to create relationships between scenarios and products or solutions based on ecological library data, knowledge base data, and industry policy data; receive user selections of scenarios, and automatically push products or solutions based on the selected scenarios; and use deep learning algorithms to explore user scenario needs based on the scenario interaction process including the pushed products or solutions selected by the user.

3. The intelligent scenario marketing system according to claim 1 is characterized in that: The demand analysis module is also used to directly receive scenario requirements published by users, query matching user ecological partners, generate corresponding invitations and push them to user ecological partners through multiple channels; receive ecological responses including products or solutions sent by user ecological partners and push them back to users.

4. The intelligent scenario-based marketing system according to claim 3 is characterized in that: The demand analysis module is also used to conduct full-process management of demand investigation, demand review, project tracking and order processing based on the predicted current user scenario needs.

5. The intelligent scenario-based marketing system according to claim 4 is characterized in that: The data center module is also used to monitor the accuracy of the demand review corresponding to the predicted scenario demand and the success rate of the order in the whole process management of each user; judge whether the accuracy of the demand review corresponding to the predicted scenario demand is greater than the preset review accuracy rate, and when it is judged that it is not greater than the preset review accuracy rate, use the incremental training method to optimize the training of the user demand prediction model until the preset review accuracy rate is reached; judge whether the success rate of the order is greater than the preset order success rate, and when it is judged that it is not greater than the preset order success rate, use the incremental training method to optimize the training of the multi-dimensional data fusion analysis model until the preset order success rate is reached.

6. The intelligent scenario marketing system according to claim 2 is characterized in that: The scenario interaction module is used to use 3D modeling and VR technology to create user-customized scenarios and user portraits in customized scenarios, and to establish the relationship between customized scenarios and products or solutions; automatically push products or solutions based on customized scenarios, and use deep learning algorithms to mine user scenario needs based on the scenario interaction process including the pushed products or solutions selected by the user; the demand analysis module is also used to obtain historical user ecological library data and the products or solutions applied by users in customized scenarios, use deep learning algorithms to mine user scenario needs, and supplement training to obtain a user demand prediction model.

7. The intelligent scenario marketing system according to claim 1 is characterized in that: Also includes: The AI ​​assistant module is equipped with a user input unit and an intelligent semantic analysis unit. It is used to use the intelligent semantic analysis unit to parse user input information and search for user ecological partners, knowledge assets, and industry policies based on industry preset scenarios. At the same time, a visual display interface is set up to display the user ecological partners, knowledge assets, and industry policies corresponding to the found industry preset scenarios. The smart exhibition hall module is used to integrate and uniformly manage the results of each user's corresponding predicted scenario needs and the corresponding demand investigation, demand review, project tracking and order-making process, and display them.

8. A marketing method using the smart scenario marketing system according to any one of claims 1 to 7, characterized in that: include: Use the user demand prediction model to predict the current user scenario needs; Utilize the ecological library module to monitor user ecosystem entry and locate the industry ecosystem to which the settled users belong, access third-party enterprise management information platforms, and find user ecosystem partners that match users based on industry affiliation and industry preset scenarios; utilize the intelligent search function of the knowledge base module to complete scenario-based industry knowledge asset queries; utilize the industry policy module to collect scenario-related policy information; By combining user ecosystem partners, scenario-based industry knowledge assets, and scenario-related policy information, and utilizing multi-dimensional data fusion analysis models, we can obtain and push products or solutions that meet user scenario needs.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to claim 8.

10. A computer device, characterized in that: The computer device includes a memory, a processor, and a program stored and executable on the memory, and the steps of the method according to claim 8 are implemented when the program is executed by the processor.

Images