Dynamic label system construction method and device, terminal and storage medium
By automatically generating a tag system, the problems of error and inefficiency caused by differences in expert perspectives are solved, and an efficient and stable tag system can be built and dynamically updated.
Patent Information
- Application Number
- CN202510891148.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-30
- Publication Date
- 2025-10-31
AI Technical Summary
Existing methods for building tag systems suffer from large errors and low efficiency due to differences in expert perspectives, making it impossible to effectively build high-quality tag systems.
By identifying the analysis scenario, classifying entities, summarizing their features, filtering key features, generating and validating feature trees, and automatically generating a labeling system, the influence of human subjective factors is avoided.
It has achieved the construction of an efficient and stable labeling system, reduced human error, and can be dynamically updated to adapt to industry development trends.
Smart Images
Figure CN120873833A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of industrial label construction technology, and in particular to a method, apparatus, terminal and storage medium for constructing a dynamic label system. Background Technology
[0002] A tagging system is a structured classification system used to label, organize, and retrieve data (such as text, images, and industry entities). It can be a simple list of keywords or a hierarchical classification tree (such as "Industry > Technology > Sub-sector"). An effective tagging system can improve data discoverability, analytical efficiency, and the systematic nature of knowledge management.
[0003] In related technologies, the construction of existing labeling systems is generally based on expert experience. Experts define labeling rules based on their experience, and then manually label the labels, thereby iteratively optimizing the system.
[0004] However, in related technologies, due to subjective differences in the perspectives of different experts, there are subjective differences in the content that industry analysis focuses on, and the use of manually generated tags leads to a certain degree of error and inefficiency in the construction of the tag system; these issues urgently need to be improved. Summary of the Invention
[0005] This application provides a method, apparatus, terminal, and storage medium for constructing a dynamic tagging system, in order to solve the problems in related technologies, such as subjective differences in the perspectives of different experts, subjective differences in the content of industry analysis, and the use of manually generated tags, which lead to a certain degree of error and low efficiency in the construction of the tagging system.
[0006] The first aspect of this application provides a method for constructing a dynamic tagging system, including the following steps: determining an analysis scenario based on classification requirements, and obtaining the classification entity corresponding to the analysis scenario;
[0007] Obtain all features of the categorized entity, and determine at least one key feature corresponding to the categorized entity from all features; generate a feature tree corresponding to the categorized entity based on the at least one key feature, and verify the feature tree, so that if the feature tree passes the verification, construct a label system corresponding to the categorized entity based on the feature tree.
[0008] Through the above technical solution, the embodiments of this application can obtain the classification entities corresponding to the analysis scenario, and then determine the key features corresponding to the classification entities and generate a feature tree, thereby realizing the construction of the label system corresponding to the classification entities. The whole process analyzes the classification entities and automatically generates the label system without human intervention, effectively avoiding errors caused by human subjective factors.
[0009] Optionally, in one embodiment of this application, obtaining the classification entity corresponding to the analysis scenario includes: obtaining the business objective corresponding to the analysis scenario; extracting key entities in the analysis scenario based on the business objective as the classification entity, wherein, when the analysis scenario is an industry analysis scenario, the key entity includes at least one of enterprise, project, and talent.
[0010] Through the above technical solution, the embodiments of this application can extract key entities in the analysis scenario in combination with business objectives as classification entities, thereby more accurately determining the classification entities that match the indicators reflected in the business objectives in the analysis scenario.
[0011] Optionally, in one embodiment of this application, obtaining all features of the classified entity includes: if the analysis scenario is the industry analysis scenario, obtaining industry text data corresponding to the industry analysis, wherein the industry text data includes industry annual reports and / or industry research reports; and summarizing all features corresponding to the classified entity based on the industry text data according to a preset language model.
[0012] Through the above technical solution, the embodiments of this application can obtain text data corresponding to the analysis scenario, and then obtain all features of the classified entities based on the text data, thereby realizing the correspondence between the classification features and the analysis scenario, and thus more effectively extracting key features.
[0013] Optionally, in one embodiment of this application, determining at least one key feature corresponding to the classified entity from all the features includes: analyzing the attention information corresponding to each feature among all the features, wherein the attention information is used to reflect the frequency of mention of each feature in the industry text; and based on the attention information of all the features, selecting features whose attention information meets preset requirements as key features.
[0014] Through the above technical solution, the embodiments of this application can filter out key features among all features through attention analysis, so that key features can better represent classified entities and can serve as the iconic features of classified entities, which is more conducive to the generation of a labeling system based on key features in subsequent steps.
[0015] Optionally, in one embodiment of this application, determining at least one key feature corresponding to the classification entity from all the features includes: generating search terms corresponding to each feature; performing a search based on the search terms to obtain search results corresponding to each search term; determining the weight information corresponding to each feature in industry analysis based on the search results; and determining the key feature corresponding to the classification entity based on the weight information.
[0016] Through the above technical solution, the embodiments of this application can determine the weight information corresponding to each feature in industry analysis, and then determine the key features of the classified entity based on the weight, thereby effectively understanding the importance of each feature.
[0017] Optionally, in one embodiment of this application, generating a feature tree corresponding to the classified entity based on the at least one key feature and verifying the feature tree includes: establishing a feature tree corresponding to the classified entity by using the classified entity as a tree node and the at least one key feature as a branch node; obtaining a feature table corresponding to the industry analysis, and verifying the feature tree based on the feature table, wherein the feature table is used to reflect a feature set formulated based on expert experience for the industry analysis.
[0018] Through the above technical solution, the embodiments of this application can construct a feature tree by using key features and classified entities as branch nodes and tree nodes respectively, and then construct a label system corresponding to the classified entities based on the feature tree. The whole process does not involve the subjective consciousness of experts or manually generated labels, which is efficient, stable, and effectively avoids errors caused by human subjective factors.
[0019] Optionally, in one embodiment of this application, the method further includes: capturing industry development trend information at preset intervals, and updating the industry text data based on the industry development trend information to obtain new key features; updating the feature tree based on the new key features; and updating the label system corresponding to the classified entity based on the updated feature tree.
[0020] Through the above technical solution, the embodiments of this application can update the feature tree based on the industrial development trend information captured over a preset time period, and then update the label system corresponding to the classified entities, thereby realizing the dynamic updating of labels with the industrial development trend.
[0021] A second aspect of this application provides an apparatus for constructing a dynamic tagging system, comprising: an acquisition module, configured to determine an analysis scenario based on classification requirements and acquire a classification entity corresponding to the analysis scenario; a determination module, configured to acquire all features of the classification entity and determine at least one key feature corresponding to the classification entity from all features; and a construction module, configured to generate a feature tree corresponding to the classification entity based on the at least one key feature and verify the feature tree, so as to construct a tagging system corresponding to the classification entity based on the feature tree if the feature tree passes verification.
[0022] Through the above technical solution, the embodiments of this application can obtain the classification entities corresponding to the analysis scenario, and then determine the key features corresponding to the classification entities and generate a feature tree, thereby realizing the construction of the label system corresponding to the classification entities. The whole process analyzes the classification entities and automatically generates the label system without human intervention, effectively avoiding errors caused by human subjective factors.
[0023] Optionally, in one embodiment of this application, the acquisition module includes: an acquisition unit, configured to acquire business objectives corresponding to the analysis scenario; and an extraction unit, configured to extract key entities in the analysis scenario based on the business objectives as the classification entities, wherein, when the analysis scenario is an industry analysis scenario, the key entities include at least one of enterprises, projects, and talents.
[0024] Through the above technical solution, the embodiments of this application can extract key entities in the analysis scenario in combination with business objectives as classification entities, thereby more accurately determining the classification entities that match the indicators reflected in the business objectives in the analysis scenario.
[0025] Optionally, in one embodiment of this application, the determining module includes: an acquisition unit, configured to acquire industry text data corresponding to the industry analysis if the analysis scenario is the industry analysis scenario, wherein the industry text data includes industry annual reports and / or industry research reports; and a summarizing unit, configured to summarize all features corresponding to the classified entity based on the industry text data according to a preset language model.
[0026] Through the above technical solution, the embodiments of this application can obtain text data corresponding to the analysis scenario, and then obtain all features of the classified entities based on the text data, thereby realizing the correspondence between the classification features and the analysis scenario, and thus more effectively extracting key features.
[0027] Optionally, in one embodiment of this application, the determining module further includes: an analysis unit, configured to analyze the attention information corresponding to each of the features, wherein the attention information is used to reflect the frequency of each feature being mentioned in the industry text; and a filtering unit, configured to filter out features whose attention information meets preset requirements based on the attention information of all features, as key features.
[0028] Through the above technical solution, the embodiments of this application can filter out key features among all features through attention analysis, so that key features can better represent classified entities and can serve as the iconic features of classified entities, which is more conducive to the generation of a labeling system based on key features in subsequent steps.
[0029] Optionally, in one embodiment of this application, the determining module is further configured to: generate search terms corresponding to each feature; perform a search based on the search terms to obtain search results corresponding to each search term; determine the weight information corresponding to each feature in industry analysis based on the search results; and determine the key features corresponding to the classified entity based on the weight information.
[0030] Through the above technical solution, the embodiments of this application can determine the weight information corresponding to each feature in industry analysis, and then determine the key features of the classified entity based on the weight, thereby effectively understanding the importance of each feature.
[0031] Optionally, in one embodiment of this application, the construction module includes: a building unit, used to build a feature tree corresponding to the classified entity by taking the classified entity as a tree node and the at least one key feature as a branch node; and a verification unit, used to obtain a feature table corresponding to the industry analysis and verify the feature tree based on the feature table, wherein the feature table is used to reflect the feature set formulated for the industry analysis based on expert experience.
[0032] Through the above technical solution, the embodiments of this application can construct a feature tree by using key features and classified entities as branch nodes and tree nodes respectively, and then construct a label system corresponding to the classified entities based on the feature tree. The whole process does not involve the subjective consciousness of experts or manually generated labels, which is efficient, stable, and effectively avoids errors caused by human subjective factors.
[0033] Optionally, in one embodiment of this application, it further includes: a first update module, used to capture industry development trend information at preset intervals and update the industry text data based on the industry development trend information to obtain new key features; a second update module, used to update the feature tree based on the new key features; and a third update module, used to update the label system corresponding to the classified entity based on the updated feature tree.
[0034] Through the above technical solution, the embodiments of this application can update the feature tree based on the industrial development trend information captured over a preset time period, and then update the label system corresponding to the classified entities, thereby realizing the dynamic updating of labels with the industrial development trend.
[0035] A third aspect of this application provides a terminal, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for constructing a dynamic tagging system as described in the above embodiments.
[0036] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for constructing a dynamic tagging system.
[0037] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0038] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:
[0039] Figure 1 This is a schematic diagram illustrating the application environment of a method for constructing a dynamic tagging system according to a specific embodiment of this application;
[0040] Figure 2 This is a flowchart illustrating a method for constructing a dynamic tagging system according to an embodiment of this application;
[0041] Figure 3 This is a flowchart illustrating the actual application of a method for constructing a dynamic tagging system according to a specific embodiment of this application.
[0042] Figure 4 This is a flowchart illustrating a method for constructing a dynamic tagging system according to a specific embodiment of this application.
[0043] Figure 5 This is a schematic diagram of the structure of a device for constructing a dynamic tagging system according to an embodiment of this application;
[0044] Figure 6 A schematic diagram of the terminal provided in the embodiments of this application. Detailed Implementation
[0045] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0046] The following description, with reference to the accompanying drawings, outlines a method, apparatus, terminal, and storage medium for constructing a dynamic tagging system according to embodiments of this application. Addressing the issues raised in the background section regarding the related technologies, where subjective differences exist between expert perspectives and the content of industry analysis, coupled with the use of manually generated tags, lead to errors and inefficiency in tagging system construction, this application provides a method for constructing a dynamic tagging system. This method first determines the analysis scenario and obtains the corresponding classification entities. Next, it summarizes all features of the classification entities and identifies key features corresponding to the classification entities from all features. Then, it generates a feature tree corresponding to the classification entities based on the key features and verifies the feature tree. Once the feature tree is verified, a tagging system corresponding to the classification entities is generated based on the feature tree. This method analyzes classification entities and automatically generates a tagging system. The entire process is efficient and stable, effectively avoiding errors caused by human subjective factors. Therefore, it solves the problems of errors and inefficiency in tagging system construction caused by subjective differences between expert perspectives and the content of industry analysis, coupled with the use of manually generated tags, in the related technologies.
[0047] The method for constructing a dynamic tagging system provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown;
[0048] The terminal 102 communicates with the server 104 via a network; the data storage system can store the data that the server 104 needs to process; the data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers; the method for constructing the dynamic tag system can be executed by the terminal 102 or the server 104, or it can be executed collaboratively by the terminal 102 and the server 104.
[0049] The terminal 102 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, IoT device, or portable wearable device. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices, etc.
[0050] Server 104 can be an independent physical server or a service node in a blockchain system, where the service nodes form a peer-to-peer network.
[0051] In addition, server 104 can also be a server cluster consisting of multiple physical servers, which can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.
[0052] Terminal 102 and server 104 can be connected via Bluetooth, USB (Universal Serial Bus) or network communication, and this embodiment does not impose specific limitations.
[0053] Specifically, Figure 2 This is a flowchart illustrating a method for constructing a dynamic tagging system provided in an embodiment of this application.
[0054] like Figure 2 As shown, the construction method of this dynamic tagging system includes the following steps:
[0055] In step S101, the analysis scenario is determined based on the classification requirements, and the classification entities corresponding to the analysis scenario are obtained.
[0056] In this embodiment, the classification entity reflects the object for which a tag system needs to be built, and the analysis content corresponding to the analysis scenario is related to the classification entity. Therefore, the analysis scenario is associated with the object for which a tag system needs to be built, and the classification entity corresponding to different analysis scenarios will also be different. Therefore, after determining the analysis scenario, this embodiment can determine the classification entities that need to have a tag system built in that analysis scenario.
[0057] The embodiments of this application can determine the classification scenario based on the requirements, and then obtain the classification entity corresponding to the classification scenario, so as to realize the correspondence between the classification entity and the classification scenario.
[0058] In step S102, all features of the classified entity are obtained, and at least one key feature corresponding to the classified entity is determined from all features.
[0059] It is understood that the embodiments of this application can summarize all the features of the classified entities obtained in step S101, and determine the key features corresponding to the classified entities from all the features.
[0060] In actual implementation, after determining the classification entities, the embodiments of this application can summarize all the features of these classification entities and identify the key features from all the summarized features. These key features are the features with high importance in the classification entities.
[0061] This application embodiment can determine the key features corresponding to the classified entities, and then use them as the data basis to generate the feature tree corresponding to the classified entities, thereby providing favorable data support for constructing the label system of classified entities.
[0062] In step S103, a feature tree corresponding to the classified entity is generated based on at least one key feature, and the feature tree is verified. If the feature tree passes the verification, a label system corresponding to the classified entity is constructed based on the feature tree.
[0063] As is understandable, a feature tree is a hierarchical structure used to represent the hierarchical, associative, or dependent relationships between categorical entities (such as products, users, and industry objects) and their features (attributes or tags). It organizes the multi-dimensional features of entities into a tree or graph structure, facilitating systematic analysis, retrieval, or decision-making.
[0064] After determining the key features of the classified entity, this embodiment of the application can generate a feature tree of the classified entity based on the key features. In the feature tree, all the key features of the classified entity are reflected. Since each key feature has its own corresponding weight information or attention information, the weight information or attention information corresponding to the key features can also be reflected in the feature tree when constructing the feature tree.
[0065] Once the feature tree is constructed, this embodiment of the application can verify the feature tree. After the feature tree is verified, a label system corresponding to the classified entity is generated based on the feature tree.
[0066] This application embodiment can generate a feature tree corresponding to a classified entity based on key features, thereby using the feature tree to reflect the structural relationship between the classified entity and its features, so that the label system corresponding to the classified entity can better integrate various key features.
[0067] Optionally, in one embodiment of this application, obtaining the classification entity corresponding to the analysis scenario includes: obtaining the business objective corresponding to the analysis scenario; extracting key entities in the analysis scenario based on the business objective as classification entities, wherein, when the analysis scenario is an industry analysis scenario, the key entities include at least one of enterprises, projects, and talents.
[0068] Specifically, this application embodiment can determine classification requirements and then determine the analysis scenario based on these requirements. In this application embodiment, classification requirements are related to user roles, and different user roles have different classification requirements. For example, when the user role is a researcher in industry research, the corresponding classification requirement is to construct a tag system for the project content in industry research, and the corresponding analysis scenario is industry analysis. As another example, when the user role is a person who summarizes and organizes technical documents, the corresponding classification requirement is to construct a tag system for the field, author, and other content of the technical documents, and the corresponding analysis scenario is technical document classification. Therefore, it can be seen that this application embodiment can determine classification requirements based on user roles. Once the user role is determined, the research content or analysis content of that user role can be determined, and then the analysis scenario corresponding to that user role can be determined based on the research content or analysis content.
[0069] After determining the analysis scenario, this embodiment of the application can identify the objects for which the user needs to construct a tagging system based on that scenario, thus obtaining the classification entities. For example, in the above example, if the analysis scenario is industry analysis, the objects for which the tagging system needs to be constructed are enterprises, projects, talents, etc., therefore the classification entities can be determined to be any one or more of enterprises, projects, and talents. In other implementations, if the analysis scenario is technical data classification, the objects for which the tagging system needs to be constructed are fields, authors, etc., therefore the classification entities are fields or authors.
[0070] As a specific implementation method, in order to better identify classified entities, such as Figure 3 As shown, the embodiments of this application can be combined with business objectives for comprehensive analysis. Taking industry analysis as an example, the business objective can be an industry ranking list, which clearly defines the listed companies, key projects, and key talent qualifications, among other indicators. Therefore, this business objective can more accurately reflect which indicators are of greater concern in the context of industry analysis, thus more accurately determining whether the corresponding classification entity for industry analysis is any one or more of the following: company, project, or talent.
[0071] Through the above technical solution, the embodiments of this application can extract key entities in the analysis scenario in combination with business objectives as classification entities, thereby more accurately determining the classification entities that match the indicators reflected in the business objectives in the analysis scenario.
[0072] Optionally, in one embodiment of this application, obtaining all features of the classified entity includes: if the analysis scenario is an industry analysis scenario, obtaining industry text data corresponding to the industry analysis, wherein the industry text data includes industry annual reports and / or industry research reports; and summarizing all features corresponding to the classified entity based on a preset language model and the industry text data.
[0073] Specifically, taking industry analysis as an example, this application embodiment can first obtain industry text data corresponding to the industry analysis, including industry annual reports and / or industry research reports; then, analyze the industry text data to determine all features corresponding to the classified entity. For example, if the classified entity is talent, all relevant talent features can be summarized from industry annual reports and / or industry research reports, including but not limited to: name, place of origin, education, graduating institution, and position.
[0074] As one possible approach, this application embodiment can extract features from the aforementioned industry text data based on a Large Language Model (LLM), thereby summarizing all features corresponding to the classified entities. The entire feature extraction process is efficient and fast.
[0075] Through the above technical solution, the embodiments of this application can obtain text data corresponding to the analysis scenario, and then obtain all features of the classified entities based on the text data, thereby realizing the correspondence between the classification features and the analysis scenario, and thus more effectively extracting key features.
[0076] Optionally, in one embodiment of this application, determining at least one key feature corresponding to the classified entity from all features includes: analyzing the attention information corresponding to each feature among all features, wherein the attention information is used to reflect the frequency of each feature being mentioned in industry text materials; and based on the attention information of all features, selecting features whose attention information meets preset requirements as key features.
[0077] Understandably, in tagging systems or data analysis, "attention information" refers to metrics used to quantify the importance, popularity, or degree of user interest in a particular feature (such as tags, keywords, entities, etc.). It helps identify which features are more worthy of attention, thereby optimizing resource allocation or decision-making.
[0078] In actual implementation, the embodiments of this application can first analyze the attention information corresponding to each feature, and then, based on the attention information of all features, select features whose attention information meets the preset requirements as key features.
[0079] Specifically, as an achievable approach, this embodiment can perform attention analysis on each feature to determine the attention information corresponding to each feature. This attention information reflects the frequency with which each feature is mentioned in industry textual materials. In other words, after determining the features corresponding to a categorized entity, this embodiment can further determine the frequency of these features appearing in industry textual materials, and then use the frequency as the attention information for that feature. The higher the frequency, the higher the attention information for that feature, indicating that this feature plays a more important role for the categorized entity. For example, when the categorized entity is a company, its corresponding features include: registration time, registered capital, company size, honors and qualifications, etc. After analyzing all features, if the feature with the highest attention information is honors and qualifications, it indicates that honors and qualifications play a more important role for the company. After obtaining the attention information for all features, this embodiment can select features whose attention information meets preset requirements as key features. During the selection process, this embodiment can select the feature with the highest attention information as the key feature, and can also select features with attention information higher than a preset value as key features.
[0080] Through the above technical solution, the embodiments of this application can filter out key features among all features through attention analysis, so that key features can better represent classified entities and can serve as the iconic features of classified entities, which is more conducive to the generation of a labeling system based on key features in subsequent steps.
[0081] Optionally, in one embodiment of this application, determining at least one key feature corresponding to the classified entity from all features includes: generating search terms corresponding to each feature; performing a search based on the search terms to obtain search results corresponding to each search term; determining the weight information corresponding to each feature in industry analysis based on the search results; and determining the key feature corresponding to the classified entity based on the weight information.
[0082] In addition to filtering key features based on attention information, as another possible approach, this application embodiment can also use weight analysis to determine the key features among all features when determining key features.
[0083] Specifically, this application embodiment can generate search terms corresponding to each feature based on all features. Then, a search is performed based on the search terms to obtain the search results corresponding to each search term. Furthermore, based on the search results, the weight information corresponding to each feature in the industry analysis is determined. This weight information can be directly determined based on the number of search results for each feature; the more search results, the higher the weight information. Finally, based on the weight information, the key features corresponding to the classified entity are determined. This application embodiment can use the feature with the highest weight information as the key feature, or the feature with a weight information higher than a preset value as the key feature.
[0084] Through the above technical solution, the embodiments of this application can determine the weight information corresponding to each feature in industry analysis, and then determine the key features of the classified entity based on the weight, thereby effectively understanding the importance of each feature.
[0085] Optionally, in one embodiment of this application, generating a feature tree corresponding to a classified entity based on at least one key feature and verifying the feature tree includes: establishing a feature tree corresponding to a classified entity by using the classified entity as a tree node and at least one key feature as a branch node; obtaining a feature table corresponding to the industry analysis; and verifying the feature tree based on the feature table, wherein the feature table is used to reflect the feature set formulated based on expert experience for the industry analysis.
[0086] As one possible approach, embodiments of this application can use classified entities as tree nodes and determined key features as branch nodes to establish a feature tree corresponding to the classified entities; then obtain a preset feature table corresponding to industry analysis, and verify the feature tree based on the preset feature table.
[0087] Specifically, in this embodiment, the categorized entity can be used as a tree node, and the determined key features as branch nodes to establish a feature tree corresponding to the categorized entity. When reflecting the attention information or weight information of each key feature, this embodiment can use branch nodes of different lengths to represent key features with different attention information or weight information. For example, if the attention information of key feature A is 40% and the attention information of key feature B is 15%, then the branch node of key feature A is longer than the branch node of key feature B. In other possible implementations, a feature list generated based on the key features and their corresponding attention information or weight information can also reflect the key features of the categorized entity.
[0088] Furthermore, when verifying the aforementioned feature tree, this embodiment of the application can obtain a preset feature table corresponding to the industry analysis, and verify the feature tree based on the preset feature table. This preset feature table can be set based on the prior knowledge of industry experts. If the preset feature table matches the key features reflected in the feature tree, the feature tree verification is successful, and a tag system can be generated based on the key features reflected in the feature tree. The tag system in this embodiment of the application maps the analysis scenario, the classification entity, and the key features. For example, if the analysis scenario is industry analysis, the corresponding classification entity is enterprise, and the corresponding key feature is honors and qualifications, then a tag system of industry analysis -- enterprise -- honors and qualifications is generated. If the preset feature table does not match the key features reflected in the feature tree, the feature tree verification fails. In this case, key features in the feature tree that are inconsistent with the preset feature table can be obtained, and analysis can be performed based on these inconsistent key features. The inconsistent key features can be optimized, or all features of the classification entity can be re-analyzed, and key features can be re-selected until the selected key features can pass the verification of the preset feature table.
[0089] Furthermore, in order to make the labeling system of classified entities more accurate, this application embodiment can summarize the contextual information of the industry text data corresponding to the classified entities based on a large language model, and perform feature search based on the summarized contextual information, so as to find more features and be more accurate when analyzing key features in subsequent steps.
[0090] Optionally, in one embodiment of this application, generating a feature tree corresponding to a classified entity based on at least one key feature and verifying the feature tree includes: establishing a feature tree corresponding to a classified entity by using the classified entity as a tree node and at least one key feature as a branch node; obtaining a feature table corresponding to the industry analysis; and verifying the feature tree based on the feature table, wherein the feature table is used to reflect the feature set formulated based on expert experience for the industry analysis.
[0091] Through the above technical solution, the embodiments of this application can construct a feature tree by using key features and classified entities as branch nodes and tree nodes respectively, and then construct a label system corresponding to the classified entities based on the feature tree. The whole process does not involve the subjective consciousness of experts or manually generated labels, which is efficient, stable, and effectively avoids errors caused by human subjective factors.
[0092] Optionally, in one embodiment of this application, the method further includes: capturing industry development trend information at preset intervals, and updating industry text data based on the industry development trend information to obtain new key features; updating the feature tree based on the new key features; and updating the label system corresponding to the classified entities based on the updated feature tree.
[0093] It is understood that the duration for capturing industry development trend information can be set according to actual circumstances in the embodiments of this application.
[0094] Specifically, this application embodiment can capture industry development trend information at preset time intervals, update industry text data based on industry development trend information, extract new features based on the updated industry text data, update the key features corresponding to the classified entities, and update the feature tree of the classified entities based on the updated key features. Finally, the label system corresponding to the classified entities is updated based on the updated feature tree, realizing the dynamic updating of the label system.
[0095] Through the above technical solution, the embodiments of this application can update the feature tree based on the industrial development trend information captured over a preset time period, and then update the label system corresponding to the classified entities, thereby realizing the dynamic updating of labels with the industrial development trend.
[0096] To enable those skilled in the art to more clearly understand the method for constructing the dynamic tagging system proposed in this application, the method will be further illustrated below with a specific embodiment.
[0097] like Figure 4 As shown, this method can be derived from... Figure 2 It can be executed by a server or terminal in the system, or by... Figure 2 The server and terminal in the process work together to execute the method. Figure 2 Taking the terminal execution in the example, the following steps are included:
[0098] Step S401: Determine the analysis scenario and obtain the classification entities corresponding to the analysis scenario;
[0099] Step S402: Summarize all features of the classified entities and determine the key features corresponding to the classified entities from all features;
[0100] Step S403: Generate a feature tree corresponding to the classified entity based on the key features, and verify the feature tree. After the feature tree is verified, generate a label system corresponding to the classified entity based on the feature tree.
[0101] According to the method for constructing a dynamic tagging system proposed in this application, the analysis scenario is first determined, and the corresponding classification entities are obtained. Then, all features of the classification entities are summarized, and key features corresponding to the classification entities are determined from all features. Next, a feature tree corresponding to the classification entities is generated based on the key features, and the feature tree is verified. Once the feature tree is verified, a tagging system corresponding to the classification entities is generated based on the feature tree. This method analyzes classification entities and automatically generates a tagging system. The entire process is efficient and stable, effectively avoiding errors caused by human subjective factors. Therefore, it solves the problems in related technologies, such as subjective differences in the perspectives of different experts, subjective differences in the content focused on in industry analysis, and the use of manually generated tags, which lead to errors and low efficiency in the construction of the tagging system.
[0102] Next, refer to the appendix. Figure 5 This application describes an apparatus for constructing a dynamic tagging system according to embodiments thereof.
[0103] Figure 5 This is a block diagram of a device for constructing a dynamic tagging system according to an embodiment of this application.
[0104] like Figure 5 As shown, the construction device 10 of the dynamic tag system includes: an acquisition module 100, a determination module 200, and a construction module 300.
[0105] The acquisition module 100 is used to determine the analysis scenario based on the classification requirements and acquire the classification entities corresponding to the analysis scenario.
[0106] The determination module 200 is used to obtain all features of the classified entity and determine at least one key feature corresponding to the classified entity from all features.
[0107] The construction module 300 is used to generate a feature tree corresponding to a classified entity based on at least one key feature, and to verify the feature tree so that if the feature tree passes the verification, a label system corresponding to the classified entity is constructed based on the feature tree.
[0108] Optionally, in one embodiment of this application, the acquisition module 100 includes: an acquisition unit and an extraction unit; the acquisition unit is used to acquire business objectives corresponding to the analysis scenario; the extraction unit is used to extract key entities in the analysis scenario based on the business objectives as classification entities, wherein, in the case of an industry analysis scenario, the key entities include at least one of enterprises, projects, and talents.
[0109] Optionally, in one embodiment of this application, the determining module 200 includes: an acquisition unit and a summarizing unit; the acquisition unit is used to acquire industry text data corresponding to the industry analysis if the analysis scenario is an industry analysis scenario, wherein the industry text data includes industry annual reports and / or industry research reports; the summarizing unit is used to summarize all features corresponding to the classified entities based on a preset language model and the industry text data.
[0110] Optionally, in one embodiment of this application, the determining module 200 further includes: an analysis unit, configured to analyze the attention information corresponding to each feature among all features, wherein the attention information is used to reflect the frequency of each feature being mentioned in industry text materials; and a filtering unit, configured to filter out features whose attention information meets preset requirements based on the attention information of all features, as key features.
[0111] Optionally, in one embodiment of this application, the determining module 200 is further configured to: generate search terms corresponding to each feature; perform a search based on the search terms to obtain search results corresponding to each search term; determine the weight information corresponding to each feature in the industry analysis based on the search results; and determine the key features corresponding to the classified entity based on the weight information.
[0112] Optionally, in one embodiment of this application, the construction module 300 includes: a building unit and a verification unit; the building unit is used to build a feature tree corresponding to the classified entity by taking the classified entity as a tree node and taking at least one key feature as a branch node; the verification unit is used to obtain a feature table corresponding to the industry analysis and verify the feature tree based on the feature table, wherein the feature table is used to reflect the feature set formulated based on expert experience for the industry analysis.
[0113] Optionally, in one embodiment of this application, the dynamic tagging system construction device 10 further includes: a first update module, a second update module, and a third update module; the first update module is used to capture industry development trend information every preset time interval, and update industry text data based on industry development trend information to obtain new key features; the second update module is used to update the feature tree based on the new key features; and the third update module is used to update the tagging system corresponding to the classified entity based on the updated feature tree.
[0114] It should be noted that the foregoing explanation of the embodiment of the method for constructing the dynamic tag system also applies to the apparatus for constructing the dynamic tag system in this embodiment, and will not be repeated here.
[0115] Each module in the aforementioned dynamic tagging system construction device can be implemented entirely or partially through software, hardware, or a combination thereof.
[0116] The above modules can be embedded in the processor of the terminal in hardware form or independent of it, or they can be stored in the memory of the terminal in software form so that the processor can call and execute the corresponding operations of the above modules.
[0117] Each module in the aforementioned dynamic tagging system construction device can be implemented entirely or partially through software, hardware, or a combination thereof.
[0118] The above modules can be embedded in the processor of the terminal in hardware form or independent of it, or they can be stored in the memory of the terminal in software form so that the processor can call and execute the corresponding operations of the above modules.
[0119] According to the dynamic tagging system construction apparatus proposed in this application, the analysis scenario is first determined, and the corresponding classification entities are obtained. Then, all features of the classification entities are summarized, and key features corresponding to the classification entities are determined from all features. Next, a feature tree corresponding to the classification entities is generated based on the key features, and the feature tree is verified. Once the feature tree is verified, a tagging system corresponding to the classification entities is generated based on the feature tree. This method analyzes classification entities and automatically generates a tagging system. The entire process is efficient and stable, effectively avoiding errors caused by human subjective factors. Therefore, it solves the problems in related technologies, such as subjective differences in the perspectives of different experts, subjective differences in the content focused on in industry analysis, and the use of manually generated tags, which lead to errors and low efficiency in the construction of the tagging system.
[0120] Figure 6 This is an internal structural diagram of a terminal provided in an embodiment of this application. The terminal may include: a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device.
[0121] The processor, memory, and input / output interfaces are connected via a system bus. The communication interface, display unit, and input devices are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for constructing a dynamic tag system. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the terminal casing, or an external keyboard, touchpad, or mouse.
[0122] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the embodiments of this application, and does not constitute a limitation on the terminal to which the embodiments of this application are applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0123] In some embodiments, the terminal may include a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the following steps:
[0124] Determine the analysis scenario and obtain the corresponding classification entities;
[0125] Summarize all features of the classified entities and determine the key features corresponding to the classified entities from all features;
[0126] Based on key features, generate feature trees corresponding to the classified entities, and verify the feature trees. Once the feature trees are verified, generate a label system corresponding to the classified entities based on the feature trees.
[0127] It should be noted that if the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this invention are all information and data authorized by the user or fully authorized by all parties, the collection, use and processing of the relevant data shall comply with relevant regulations.
[0128] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for constructing a dynamic tagging system.
[0129] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0130] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0131] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0132] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0133] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0134] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
[0135] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0136] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A method for constructing a dynamic tagging system, characterized in that, Includes the following steps: The analysis scenario is determined based on the classification requirements, and the classification entities corresponding to the analysis scenario are obtained; Obtain all features of the classified entity, and determine at least one key feature corresponding to the classified entity from all features; Based on the at least one key feature, a feature tree corresponding to the classified entity is generated, and the feature tree is verified. If the feature tree passes the verification, a label system corresponding to the classified entity is constructed based on the feature tree.
2. The method according to claim 1, characterized in that, The acquisition of the classification entity corresponding to the analysis scenario includes: Obtain the business objectives corresponding to the analysis scenario; Based on the business objectives, key entities in the analysis scenario are extracted as the classification entities. In the case that the analysis scenario is an industry analysis scenario, the key entities include at least one of enterprises, projects, and talents.
3. The method according to claim 2, characterized in that, The process of obtaining all features of the classified entity includes: If the analysis scenario is the industry analysis scenario, then obtain the industry text data corresponding to the industry analysis, wherein the industry text data includes industry annual reports and / or industry research reports; Based on a preset language model, all features corresponding to the classified entities are summarized from the industry text data.
4. The method according to claim 1, characterized in that, Determining at least one key feature corresponding to the classified entity from all the features includes: Analyze the attention information corresponding to each feature among all the features, wherein the attention information is used to reflect the frequency of each feature being mentioned in the industry text materials; Based on the attention information of all the features, features whose attention information meets the preset requirements are selected as key features.
5. The method according to claim 4, characterized in that, Determining at least one key feature corresponding to the classified entity from all the features includes: Generate search terms corresponding to each of the aforementioned features; Based on the search terms, perform a search and obtain the search results corresponding to each search term; Based on the search results, determine the weight information corresponding to each feature in the industry analysis; Based on the weight information, the key features corresponding to the classified entity are determined.
6. The method according to claim 1, characterized in that, The step of generating a feature tree corresponding to the classified entity based on the at least one key feature, and verifying the feature tree, includes: By using the classified entity as a tree node and the at least one key feature as a branch node, a feature tree corresponding to the classified entity is established. Obtain a feature table corresponding to the industry analysis, and verify the feature tree based on the feature table, wherein the feature table is used to reflect the feature set formulated based on expert experience for the industry analysis.
7. The method according to claim 1, characterized in that, Also includes: At preset intervals, industry development trend information is captured, and the industry text data is updated based on the industry development trend information to obtain new key features; The feature tree is updated based on the new key features; Based on the updated feature tree, the label system corresponding to the classified entity is updated.
8. A device for constructing a dynamic tagging system, characterized in that, include: The acquisition module is used to determine the analysis scenario based on the classification requirements and acquire the classification entities corresponding to the analysis scenario; A determination module is used to acquire all features of the classified entity and determine at least one key feature corresponding to the classified entity from all features; A construction module is used to generate a feature tree corresponding to the classified entity based on the at least one key feature, and to verify the feature tree so as to construct a label system corresponding to the classified entity based on the feature tree if the feature tree passes the verification.
9. A terminal, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the method for constructing a dynamic tagging system as described in any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the method for constructing the dynamic tagging system as described in any one of claims 1-7.