A method and system for accurately pushing information based on user portraits

Abstract

The present invention provides a method and system for accurately pushing information based on user profiles, relating to the field of data processing technology. The method comprises: obtaining enterprise behavior data and platform service data from an economic comprehensive service platform; classifying enterprises into industrial ecosystem clusters based on the enterprise behavior data, extracting dynamic demand characteristics of the industrial ecosystem clusters based on the platform service data, and constructing cluster user profiles for the industrial ecosystem clusters based on the dynamic demand characteristics; extracting service characteristics of different services based on the platform service data; matching the service characteristics with the cluster user profiles based on the technical standards corresponding to the services; and pushing service information to industrial ecosystem clusters whose matching degree meets a preset condition threshold. This method can accurately push service information based on the dynamic demand perception of industrial ecosystem groups, recommend services to specific industrial groups in a targeted manner, and improve the accuracy and coverage of service recommendations.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for accurately pushing information based on user portraits. Background Art

[0002] Intelligent recommendation technology is in a rapid development stage. With the advancement of big data, cloud computing and artificial intelligence (AI), especially the development of machine learning and deep learning, intelligent recommendation systems have been able to process more complex data sets and provide more accurate personalized services.

[0003] Intelligent recommendation technology based on industry ecosystem clusters and service sets is a complex but effective strategy for improving the accuracy of personalized service and product recommendations. This technology combines factors such as the interaction patterns of various participants in the industry ecosystem, business processes, and user preferences, and optimizes recommendation results through data analysis and machine learning algorithms. Services based on industry ecosystem clusters are often highly context-dependent. For example, recommended content may vary depending on factors such as seasonal changes, holiday promotions, or the user's geographic location. Therefore, the recommendation system must have strong contextual awareness and be able to quickly adjust its recommendation strategy.

[0004] Therefore, how to provide a method for accurately pushing information based on a full understanding of the needs of industrial ecological clusters is an urgent problem that needs to be solved. Summary of the Invention

[0005] In order to improve the above problems, the present invention provides a method and system for accurately pushing information based on user portraits.

[0006] A first aspect of an embodiment of the present invention provides a method for accurately pushing information based on user profiles, the method comprising:

[0007] Obtain enterprise behavior data and platform service data from the comprehensive economic service platform;

[0008] Classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range, and build cluster user profiles of industrial ecological clusters based on the dynamic demand characteristics;

[0009] Extract service characteristics of different services based on platform service data;

[0010] Matching the service characteristics with the cluster user profile based on the technical standards corresponding to the service;

[0011] Push service information to industrial ecological clusters whose matching degree reaches the preset condition threshold.

[0012] Optionally, the step of classifying enterprises into industrial ecological clusters based on the enterprise behavior data specifically includes:

[0013] Extracting basic information, business information, product and service information, and activity information of the enterprise from the enterprise behavior data;

[0014] Input the extracted information into the pre-trained industrial ecological cluster classification model;

[0015] Obtain the results of industrial ecological cluster classification.

[0016] Optionally, the step of extracting dynamic demand characteristics of industrial ecological clusters based on platform service data of various industrial ecological clusters within a certain time range specifically includes:

[0017] For each industrial ecological cluster, extract service demand information, service demand time, and service demand goals within a certain time range from the platform service data;

[0018] Obtain the industry event information and industry event time corresponding to each industrial ecological cluster within the same time range from the comprehensive economic service platform;

[0019] Correlate service demand information and industry event information according to the chronological order of service demand time and industry event time;

[0020] The dynamic demand characteristics of each industrial ecological cluster are extracted based on the correlation results.

[0021] Optionally, the step of constructing a cluster user profile of the industrial ecological cluster according to the dynamic demand characteristics specifically includes:

[0022] Obtaining cluster characteristics of each industrial ecological cluster classification based on the industrial ecological cluster classification model;

[0023] Extract the common characteristics of enterprises belonging to the same industrial ecological cluster classification based on the enterprise classification results;

[0024] A cluster user profile of each industrial ecological cluster is constructed by combining the cluster characteristics, the common characteristics of the enterprises and the dynamic demand characteristics.

[0025] Optionally, the step of extracting service features of different services based on the platform service data specifically includes:

[0026] Obtain the service source, service type, service content and service timeliness of different services;

[0027] Perform feature extraction on the acquired information to obtain service features.

[0028] Optionally, the step of matching the service characteristics with the cluster user profile based on the technical standards corresponding to the service specifically includes:

[0029] Determining whether there are technical standards associated with the service based on the service content;

[0030] If it exists, extract the standard parameters of the technical standard associated with the service;

[0031] Matching the standard parameters with the service demand targets and screening matching industrial ecological clusters;

[0032] The service characteristics of the successfully matched services are further matched with the cluster user portraits of the screened industrial ecological clusters.

[0033] Optionally, the step of matching the service characteristics with the cluster user profile based on the technical standards corresponding to the service further includes:

[0034] Establish binding relationships between services based on the associations between the extracted technical standards;

[0035] The service features of other services bound to the successfully matched service will also be further matched with the cluster user portraits of the screened industrial ecological clusters.

[0036] Optionally, the step of further matching the service characteristics with the cluster user profiles of the screened industrial ecological clusters specifically includes:

[0037] Matching the service demand information with the service type and service content;

[0038] Match the service demand time with the service effectiveness time.

[0039] Optionally, the step of pushing the service information to the industrial ecological cluster whose matching degree reaches a preset condition threshold specifically includes:

[0040] For industrial ecological clusters whose matching degree reaches the preset condition threshold but does not reach the preset optimization threshold, the service information of multiple bound services will be pushed;

[0041] For industrial ecological clusters whose matching degree reaches the preset optimal threshold, only service information of the services involved in the matching will be pushed.

[0042] A second aspect of an embodiment of the present invention provides a user profile-based accurate information push system, including:

[0043] A data acquisition unit, used to obtain enterprise behavior data and platform service data of the economic comprehensive service platform;

[0044] A user profiling unit is configured to classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range, and construct cluster user profiles of industrial ecological clusters based on the dynamic demand characteristics;

[0045] A feature extraction unit, used to extract service features of different services based on platform service data;

[0046] A feature matching unit, configured to match the service feature with the cluster user profile based on a technical standard corresponding to the service;

[0047] The information push unit is used to push service information to industrial ecological clusters whose matching degree reaches a preset condition threshold.

[0048] In summary, the present invention provides a method and system for accurately pushing information based on user portraits, which can construct industrial ecological clusters, service sets and service feature libraries based on basic data such as enterprise data and platform service data, establish a mapping relationship between industrial ecological clusters and service sets based on service-related technical standards, and recommend the service content used in the service set to all enterprises in the industrial ecological group, thereby realizing accurate push of service information based on the dynamic demand perception of the industrial ecological group, and recommending services to specific industrial groups in a targeted manner, thereby improving the accuracy and coverage of service recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 This is a flow chart of a method for accurately pushing information based on user portraits according to an embodiment of the present invention;

[0051] Figure 2 This is a functional module block diagram of the user portrait-based information precision push system according to an embodiment of the present invention.

[0052] Reference numerals:

[0053] Data acquisition unit 110; user profiling unit 120; feature extraction unit 130; feature matching unit 140; information push unit 150. DETAILED DESCRIPTION

[0054] Intelligent recommendation technology based on industry ecosystem clusters and service sets is a complex but effective strategy for improving the accuracy of personalized service and product recommendations. This technology combines factors such as the interaction patterns of various participants in the industry ecosystem, business processes, and user preferences, and optimizes recommendation results through data analysis and machine learning algorithms. Services based on industry ecosystem clusters are often highly context-dependent. For example, recommended content may vary depending on factors such as seasonal changes, holiday promotions, or the user's geographic location. Therefore, the recommendation system must have strong contextual awareness and be able to quickly adjust its recommendation strategy.

[0055] Therefore, how to provide a method for accurately pushing information based on a full understanding of the needs of industrial ecological clusters is an urgent problem that needs to be solved.

[0056] In view of this, the designers of the present invention have designed a method and system for accurately pushing information based on user portraits.

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0058] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0059] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0060] In the description of the present invention, it should be noted that the terms "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the inventive product is typically placed when in use. These terms are intended solely to facilitate the description of the present invention and simplify the description, and are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," etc., etc., are used solely for distinction and should not be construed as indicating or implying relative importance.

[0061] In the description of the present invention, it should also be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0062] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0063] The following is a detailed description of the method for accurately pushing information based on user portraits provided by this embodiment.

[0064] See also Figure 1 This embodiment provides a method for accurately pushing information based on user portraits, the method comprising:

[0065] Step S101: Obtain enterprise behavior data and platform service data of the comprehensive economic service platform.

[0066] An integrated economic service platform is a digital platform that integrates multiple functions and services, aiming to provide comprehensive economic information and decision-making support to governments, businesses, and the public. Such platforms typically integrate macroeconomic data, industry trends, corporate information, policies and regulations, and offer a range of tools and services, including data analysis, market forecasts, and consulting services.

[0067] The enterprise behavior data on the comprehensive economic service platform reflects the various interactions between enterprises on the platform and during their business activities. This data provides an important basis for enterprise profiling, demand analysis, and personalized service delivery. Enterprise behavior data includes basic information, transaction activities, financial performance, partnership networks, innovation and development, marketing and promotion, and other information generated during the production and operation of enterprises.

[0068] Platform service data covers the range of services provided by the platform to businesses and other users, as well as their usage. This helps evaluate the platform's service quality and efficiency, and also provides a reference for improving existing services and developing new ones. Platform service data includes information services, transaction support, training and education, technical support, community interaction, and other aspects.

[0069] It can be seen that the comprehensive economic service platform not only provides enterprises with rich resources and services, but also accumulates a large amount of valuable data. By obtaining enterprise behavior data and platform service data from the comprehensive economic service platform and effectively utilizing this data, it can play a huge role in accurately pushing information.

[0070] As a preferred implementation method, external information such as macroeconomic data, policy and regulatory changes, etc. can also be obtained on the comprehensive economic service platform, providing more dimensional support for the subsequent dynamic perception of industrial ecological cluster service needs.

[0071] It should be noted that although the comprehensive economic service platform also provides data analysis tools and services, its functions are usually more universal and not customized for specific application scenarios. Therefore, the analytical effects that can be achieved are limited.

[0072] Step S102: classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on platform service data of various industrial ecological clusters within a certain time range, and construct cluster user portraits of industrial ecological clusters based on the dynamic demand characteristics.

[0073] An industrial ecosystem cluster refers to a network of interconnected enterprises, institutions, and individuals within a specific region or virtual environment, collaborating to foster innovation and enhance competitiveness. This ecosystem emphasizes the cooperative and competitive relationships among multiple enterprises and service providers. By establishing a network comprised of manufacturers, suppliers, distributors, retailers, and technical service providers, resource sharing and innovation capabilities can be fostered. In this scenario, analyzing the interaction patterns among these entities can optimize recommendation logic and provide services that better meet their needs.

[0074] To achieve accurate information push for industrial ecological clusters, the prerequisite is to classify the industrial ecological clusters. Specifically, as a preferred embodiment of the present invention, the method for classifying industrial ecological clusters in step S102 specifically includes:

[0075] Extracting basic information, business information, product and service information, and activity information of the enterprise from the enterprise behavior data;

[0076] Input the extracted information into the pre-trained industrial ecological cluster classification model;

[0077] Obtain the results of industrial ecological cluster classification.

[0078] For specific classification, it can be completed with the help of existing industrial ecological cluster classification models, such as K-means clustering, DBSCAN (Density-Based Spatial Clustering of Applications with Noise), spectral clustering, hierarchical clustering, etc. The basic information, business information, product and service information, and activity behavior information of the enterprise will be extracted from the enterprise behavior data and input into the existing industrial ecological cluster classification model to realize the industrial ecological cluster classification of enterprises on the current economic comprehensive service platform.

[0079] In addition, you can also build a classification model based on customized classification requirements. The specific implementation method is as follows:

[0080] 1. Data collection and preprocessing:

[0081] Multi-source data integration:

[0082] Internal operational data: Obtain detailed operating data from the company's ERP system, CRM system, financial statements, etc.

[0083] Transaction records: including purchase orders, sales invoices, logistics information, and other data reflecting interactions between businesses.

[0084] External data: Integrate macro-environmental factors such as industry reports, market research results, and changes in policies and regulations.

[0085] Data cleaning:

[0086] Remove duplicates, fill in missing values, and correct erroneous data to ensure data quality.

[0087] Standardize data formats from different sources to facilitate subsequent analysis.

[0088] 2. Feature Engineering:

[0089] Define the base features:

[0090] Enterprise attributes: such as size, years of establishment, location, industry, etc.

[0091] Behavioral patterns: such as procurement frequency, number of partners, investment in technological innovation, etc.

[0092] Performance indicators: such as sales growth rate, profit margin, market share change, etc.

[0093] Advanced Feature Extraction:

[0094] Use NLP technology to parse text content (such as announcements and news reports) and extract semantic features from it.

[0095] Apply time series data analysis methods to discover patterns in time series, such as seasonal fluctuations or long-term trends.

[0096] 3. Choose a suitable clustering algorithm:

[0097] Unsupervised learning methods:

[0098] K-means clustering: Applicable to situations with a clear number of clusters, it assigns data points to the nearest centroid through iterative optimization.

[0099] DBSCAN (Density-Based Spatial Clustering): Suitable for processing noisy data sets and can find clusters of arbitrary shapes.

[0100] HDBSCAN (Hierarchical Density Based Clustering): extends the functionality of DBSCAN and can automatically determine the optimal number of clusters.

[0101] Spectral Clustering: It uses graph theory principles to segment data, and is particularly suitable for data with non-convex distributions.

[0102] Semi-supervised learning methods:

[0103] If there are some enterprise samples with known labels, a semi-supervised learning algorithm with label propagation can be used to improve classification accuracy.

[0104] 4. Constructing an industrial ecological cluster classification model:

[0105] Training the model:

[0106] The preprocessed data is input into the selected clustering algorithm for training to generate preliminary industrial ecological clusters.

[0107] Model Evaluation:

[0108] Model performance was evaluated using internal metrics (e.g., silhouette coefficient, Calinski-Harabasz index) and external metrics (e.g., mutual information metric, F1 score).

[0109] Compare the results of different algorithms and select the model that best suits the current dataset and application scenario.

[0110] Whether using a pre-used classification model or temporarily constructing a classification model, after completing the classification of the enterprise's industrial ecological clusters through the classification model, the dynamic demand perception of the industrial ecological cluster can be carried out.

[0111] It should be noted that although the industrial ecological cluster classification model also has characteristic definitions for different industrial ecological clusters, for enterprise groups on different platforms, even if they belong to the same industrial ecological cluster according to the classification results, the cluster characteristics they embody are different. Correspondingly, the service demands of industrial ecological clusters are also different. At the same time, due to the influence of external environmental factors such as time and industry events, the service demands of industrial ecological clusters are also dynamically changing. To achieve dynamic demand perception of industrial ecological clusters, the influence of time factors and industry events must be considered. Specifically, as a preferred method of an embodiment of the present invention, the method of extracting dynamic demand characteristics of each industrial ecological cluster in step S102 specifically includes:

[0112] For each industrial ecological cluster, extract service demand information, service demand time, and service demand goals within a certain time range from the platform service data;

[0113] Obtain the industry event information and industry event time corresponding to each industrial ecological cluster within the same time range from the comprehensive economic service platform;

[0114] Correlate service demand information and industry event information according to the chronological order of service demand time and industry event time;

[0115] The dynamic demand characteristics of each industrial ecological cluster are extracted based on the correlation results.

[0116] It's important to note that the choice of time range depends on the accuracy and timeliness of service information push. The larger the time range, the more accurate the feature extraction results will be. The closer the time range is to the current time point, the more timely the feature extraction results will be. Many business activities have highly time-sensitive requirements, often closely tied to holidays and dates.

[0117] Therefore, after obtaining the service demand information and industry event information of each industrial ecological cluster, the two are linked according to the chronological order. Among them, the service demand information corresponds to the demand information of the service content required by the enterprise. The service demand time refers to the time point or time period when the enterprise has this demand. The service demand target refers to the goal or result that the enterprise hopes to achieve based on this demand. For example, a company hopes to complete the preparation of 5 million pieces of its best-selling products in early November to prepare for the shopping street event on November 11. The service demand information it proposes is "Is XX product produced?", the service demand time is before November 11, and the service demand target is to complete the production of 5 million pieces. The corresponding industry event information is the shopping festival, and the industry event time is November 11.

[0118] Extracting demand characteristics based on time correlation can more accurately reflect the actual needs of the industrial ecological cluster within a time range.

[0119] After the dynamic demand feature extraction is completed, combined with the cluster characteristics of the industrial ecological cluster classification and the common characteristics of enterprises on the current economic comprehensive service platform, a cluster user portrait for each industrial ecological cluster classification can be obtained. Specifically, as a preferred embodiment of the present invention, the method of constructing the cluster user portrait in step S102 specifically includes:

[0120] Obtaining cluster characteristics of each industrial ecological cluster classification based on the industrial ecological cluster classification model;

[0121] Extract the common characteristics of enterprises belonging to the same industrial ecological cluster classification based on the enterprise classification results;

[0122] A cluster user profile of each industrial ecological cluster is constructed by combining the cluster characteristics, the common characteristics of the enterprises and the dynamic demand characteristics.

[0123] Cluster characteristics are universal features constructed within the industrial ecosystem cluster classification model that reflect the classification of that industrial ecosystem cluster. Enterprise common characteristics reflect the common characteristics of enterprises belonging to that industrial ecosystem cluster on the current comprehensive economic service platform. These characteristics share common elements with cluster characteristics, but also include some characteristics related to the current comprehensive economic service platform that are not reflected in the cluster characteristics. When constructing cluster user profiles, common elements can be merged and processed.

[0124] By combining the cluster characteristics, the common enterprise characteristics and the dynamic demand characteristics, it is possible to obtain a cluster user portrait for a certain industrial ecological cluster on the current economic comprehensive service platform.

[0125] On this basis, as a preferred method of an embodiment of the present invention, in order to achieve more refined service information push, enterprise user portraits can be further constructed for specific enterprises, and enterprise characteristics extracted based on enterprise information can be used to replace common enterprise characteristics, and then the enterprise user portraits can be constructed together with cluster characteristics and dynamic demand characteristics. The reason why cluster characteristics and dynamic demand characteristics are used here is that, on the one hand, it is hoped that the constructed enterprise user portrait can reflect the characteristics of the industrial ecological cluster where the enterprise is located, and on the other hand, it is considered that many enterprises cannot clearly understand their current real needs. By using dynamic demand characteristics and the common needs of the industrial ecological cluster where they are located, the hidden needs of the enterprise can be excavated and then the service information can be matched.

[0126] Step S103: extracting service features of different services based on the platform service data.

[0127] Service characteristics reflect the basic attributes of the services provided. By matching service characteristics with cluster user profiles, service information can be pushed to specific industrial ecological clusters.

[0128] Specifically, as a preferred embodiment of the present invention, step S103 specifically includes:

[0129] Obtain the service source, service type, service content and service timeliness of different services;

[0130] Perform feature extraction on the acquired information to obtain service features.

[0131] As mentioned in the above description, the needs of enterprises or industrial ecological clusters are strongly correlated with time. Therefore, when extracting service features, in addition to considering the matching of service content and demand content, the matching of time must also be considered.

[0132] On the basis of extracting service features, as a preferred method of an embodiment of the present invention, a service set can also be constructed based on the extracted service features. In order to improve the efficiency and accuracy of service information push, the service features are decomposed and aggregated from the aspects of service source, service type, service content and service timeliness to form a multi-granularity and multi-level service set, and the service set is automatically labeled. Due to the characteristics of services such as many types, large numbers, and overlapping content, it is difficult to accurately reveal the various cluster structures presented within the service using a traditional single clustering method. In this embodiment, an integrated clustering method is used to solve this problem. First, feature elements are extracted from the original data such as the content, type, and service objects of multi-source services, and feature screening is performed to construct a feature database for all services; then, a variety of traditional classical clustering algorithms or subspace clustering algorithms are used, using random parameters, random sampling, etc. to quickly generate base clusters and perform cluster difference analysis; finally, a weighted method is used to integrate the base clusters.

[0133] When pushing information later, you can push individual service information or push service sets to corresponding industrial ecological clusters or enterprises.

[0134] Step S104: Match the service characteristics with the cluster user profile based on the technical standards corresponding to the service.

[0135] Due to the complexity of the sources and content, as well as the heterogeneity of the data, the enterprise behavior data and platform service data obtained on the Economic Comprehensive Service Platform are bound to be inaccurate or impossible to match if the demand and service are directly matched after feature extraction. Technical standards are chosen as the intermediate bridge for matching because the descriptions of standardized terms, technical requirements, performance indicators, etc. in technical standards are fixed and unified. Non-standard content in demand or service can be mapped to the descriptions in technical standards, thereby achieving alignment between the two content. Further matching on this basis can make the matching results more consistent.

[0136] Specifically, as a preferred embodiment of the present invention, step S104 specifically includes:

[0137] Determining whether there are technical standards associated with the service based on the service content;

[0138] If it exists, extract the standard parameters of the technical standard associated with the service;

[0139] Matching the standard parameters with the service demand targets and screening matching industrial ecological clusters;

[0140] The service characteristics of the successfully matched services are further matched with the cluster user portraits of the screened industrial ecological clusters.

[0141] In the above steps, matching standard parameters with service requirements specifically refers to whether the standardized descriptions of specific parameters are consistent, that is, whether the two can be expressed using the same standardized description. If they do match, it means that they can be benchmarked to the same technical standard. After this stage of matching, further matching is carried out on the specific content.

[0142] The further matching process specifically includes:

[0143] Matching the service demand information with the service type and service content;

[0144] Match the service demand time with the service efficiency:

[0145] The matching content in this stage is aimed at matching specific content and time. Through the matching of two stages, the mapping relationship between industrial ecological clusters and services or service sets can be effectively determined, and then the service information can be accurately pushed based on the mapping relationship.

[0146] It should be noted that in many industries, technical standards are ubiquitous and contain a large amount of data. Therefore, service content can be associated with multiple technical standards at the same time, and these technical standards naturally have correlations. At the same time, multiple different service contents may also correspond to the same or related technical standards. At this time, in order to improve the effectiveness of information push, as a preferred method of the embodiment of the present invention, step S104 specifically includes:

[0147] Establish binding relationships between services based on the associations between the extracted technical standards;

[0148] The service features of other services bound to the successfully matched service will also be further matched with the cluster user portraits of the screened industrial ecological clusters.

[0149] That is, in addition to the services and industrial ecological clusters that are successfully matched directly through technical standards, the service characteristics of other services that will be bound to the successfully matched services are also further matched with the cluster user portraits of the screened industrial ecological clusters, thereby effectively expanding the scope of industrial ecological clusters that have demand for services.

[0150] The specific content of the further matching performed at this time is the same as described above.

[0151] Step S105: Push the service information to the industrial ecological cluster whose matching degree reaches the preset condition threshold.

[0152] If the matching degree reaches the preset condition threshold, it is considered that a mapping relationship between the industrial ecological cluster and the service or service set has been established through the above matching process. At this time, the service information can be pushed to the industrial ecological cluster that has established a mapping relationship with the service or service set.

[0153] On this basis, as a preferred embodiment, the scope of the push service information can be adjusted more flexibly based on the specific matching degree during further matching. Specifically, the above step S105 specifically includes:

[0154] For industrial ecological clusters whose matching degree reaches the preset condition threshold but does not reach the preset optimization threshold, the service information of multiple bound services will be pushed;

[0155] For industrial ecological clusters whose matching degree reaches the preset optimal threshold, only service information of the services involved in the matching will be pushed.

[0156] The value set for the preferred threshold is greater than the conditional threshold. For industrial ecological clusters whose matching degree reaches the preferred threshold, it is considered that their demand and service are highly matched. At this time, further pushing other service information in the service set is also more likely to achieve good feedback effects.

[0157] Through the above process, accurate service information push to the industrial ecological cluster is completed.

[0158] It's important to note that during the execution of this method, the companies and services on the Economic Comprehensive Service Platform evolve. For newly added companies, we use the industrial ecosystem cluster classification model to classify them and determine the industrial ecosystem group they belong to. We then use group recommendation methods to recommend services or service sets to these companies. For newly added services, we use natural language processing, machine learning, and other methods to determine the service set they belong to. We then use group recommendation methods to recommend these services to the most compatible industrial ecosystem clusters or companies.

[0159] To sum up, the method for accurate information push based on user portraits provided by the present invention can construct industrial ecological clusters, service sets and service feature libraries based on basic data such as enterprise data and platform service data, establish a mapping relationship between industrial ecological clusters and service sets based on service-related technical standards, and recommend the service content used in the service set to all enterprises in the industrial ecological group, thereby realizing accurate push of service information based on the dynamic demand perception of the industrial ecological group, recommending services to specific industrial groups in a targeted manner, and improving the accuracy and coverage of service recommendations.

[0160] like Figure 2 As shown, the present invention implements a user portrait-based accurate information push system, which includes:

[0161] The data acquisition unit 110 is used to acquire enterprise behavior data and platform service data of the economic comprehensive service platform;

[0162] The user profiling unit 120 is configured to classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range, and construct cluster user profiles of industrial ecological clusters based on the dynamic demand characteristics;

[0163] A feature extraction unit 130 is configured to extract service features of different services based on the platform service data;

[0164] A feature matching unit 140 is configured to match the service feature with the cluster user profile based on a technical standard corresponding to the service;

[0165] The information push unit 150 is used to push service information to the industrial ecological clusters whose matching degree reaches a preset condition threshold.

[0166] The user portrait-based information precision push system provided in an embodiment of the present invention is used to implement the above-mentioned user portrait-based information precision push method. Therefore, the specific implementation method is the same as the above-mentioned method and will not be repeated here.

[0167] In summary, the present invention provides a method and system for accurately pushing information based on user portraits, which can construct industrial ecological clusters, service sets and service feature libraries based on basic data such as enterprise data and platform service data, establish a mapping relationship between industrial ecological clusters and service sets based on service-related technical standards, and recommend the service content used in the service set to all enterprises in the industrial ecological group, thereby realizing accurate push of service information based on the dynamic demand perception of the industrial ecological group, and recommending services to specific industrial groups in a targeted manner, thereby improving the accuracy and coverage of service recommendations.

[0168] In the several embodiments disclosed in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.

[0169] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0170] If the functions are implemented in the form of software function modules and sold or used as independent products, they 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 the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, 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.

Claims

1. A method for accurately pushing information based on user portraits, characterized in that: The method comprises: Obtain enterprise behavior data and platform service data from the comprehensive economic service platform; Classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range, and build cluster user profiles of industrial ecological clusters based on the dynamic demand characteristics; Extract service characteristics of different services based on platform service data; Matching the service characteristics with the cluster user profile based on the technical standards corresponding to the service; Push service information to industrial ecological clusters whose matching degree reaches the preset condition threshold; The step of extracting the dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range specifically includes: extracting service demand information, service demand time, and service demand target within a certain time range from the platform service data for each industrial ecological cluster; obtaining industrial event information and industrial event time corresponding to each industrial ecological cluster within the same time range from the comprehensive economic service platform; correlating the service demand information and industrial event information according to the chronological order of the service demand time and the industrial event time; and extracting the dynamic demand characteristics of each industrial ecological cluster based on the correlation result; The step of constructing a cluster user profile of the industrial ecological cluster based on the dynamic demand characteristics specifically includes: obtaining cluster characteristics of each industrial ecological cluster classification based on the industrial ecological cluster classification model; extracting common characteristics of enterprises belonging to the same industrial ecological cluster classification based on the classification results of the enterprises; and constructing a cluster user profile of each industrial ecological cluster based on the cluster characteristics, the common characteristics of the enterprises, and the dynamic demand characteristics; The step of matching the service characteristics with the cluster user portrait based on the technical standards corresponding to the service specifically includes: judging whether there are technical standards associated with the service according to the service content; if so, extracting the standard parameters of the technical standards associated with the service; matching the standard parameters with the service demand targets, judging whether the standard parameters are consistent with the standardized description of the service demand targets for specific parameters, and screening the matching industrial ecological clusters; further matching the service characteristics of the successfully matched services with the cluster user portraits of the screened industrial ecological clusters.

2. The method for accurately pushing information based on user portraits according to claim 1, characterized in that: The step of classifying enterprises into industrial ecological clusters based on the enterprise behavior data specifically includes: Extracting basic information, business information, product and service information, and activity information of the enterprise from the enterprise behavior data; Input the extracted information into the pre-trained industrial ecological cluster classification model; Obtain the results of industrial ecological cluster classification.

3. The method for accurately pushing information based on user portraits according to claim 2, characterized in that: The step of extracting service features of different services based on platform service data specifically includes: Obtain the service source, service type, service content and service timeliness of different services; Perform feature extraction on the acquired information to obtain service features.

4. The method for accurately pushing information based on user portraits according to claim 3 is characterized in that: The step of matching the service characteristics with the cluster user profile based on the technical standards corresponding to the service specifically includes: Establish binding relationships between services based on the associations between the extracted technical standards; The service features of other services bound to the successfully matched service will also be further matched with the cluster user portraits of the screened industrial ecological clusters.

5. The method for accurately pushing information based on user portraits according to claim 4 is characterized in that: The step of further matching the service characteristics with the cluster user profiles of the screened industrial ecological clusters specifically includes: Matching the service demand information with the service type and service content; Match the service demand time with the service effectiveness time.

6. The method for accurately pushing information based on user portraits according to claim 5, characterized in that: The step of pushing the service information to the industrial ecological cluster whose matching degree reaches the preset condition threshold specifically includes: For industrial ecological clusters whose matching degree reaches the preset condition threshold but does not reach the preset optimization threshold, the service information of multiple bound services will be pushed; For industrial ecological clusters whose matching degree reaches the preset optimal threshold, only service information of the services involved in the matching will be pushed.

7. A precise information push system based on user portraits, characterized in that: include: A data acquisition unit, used to obtain enterprise behavior data and platform service data of the economic comprehensive service platform; A user profiling unit is configured to classify enterprises into industrial ecological clusters based on the enterprise behavior data, extract dynamic demand characteristics of industrial ecological clusters based on the platform service data of various industrial ecological clusters within a certain time range, and construct cluster user profiles of industrial ecological clusters based on the dynamic demand characteristics; A feature extraction unit, used to extract service features of different services based on platform service data; A feature matching unit, configured to match the service feature with the cluster user profile based on a technical standard corresponding to the service; An information push unit is used to push service information to industrial ecological clusters whose matching degree reaches a preset condition threshold; The user portrait unit is specifically used for: For each industrial ecological cluster, extract service demand information, service demand time, and service demand target within a certain time range from the platform service data; obtain industrial event information and industrial event time corresponding to each industrial ecological cluster within the same time range from the comprehensive economic service platform; Correlate service demand information and industry event information according to the chronological order of service demand time and industry event time; Extract the dynamic demand characteristics of each industrial ecological cluster based on the correlation results; Obtaining cluster characteristics of each industrial ecological cluster classification based on the industrial ecological cluster classification model; extracting common characteristics of enterprises belonging to the same industrial ecological cluster classification based on the enterprise classification results; and constructing a cluster user profile of each industrial ecological cluster by combining the cluster characteristics, the common characteristics of the enterprises, and the dynamic demand characteristics; The feature matching unit is specifically used for: Determine whether there are technical standards associated with the service based on the service content; if so, extract the standard parameters of the technical standards associated with the service; match the standard parameters with the service demand targets, determine whether the standard parameters are consistent with the standardized description of the service demand targets for specific parameters, and screen the matching industrial ecological clusters; further match the service characteristics of the successfully matched services with the cluster user portraits of the screened industrial ecological clusters.

Images