Talent portrait construction method for mining triple based on deep semantic relationship

Abstract

The invention discloses a talent portrait construction method for mining triples based on a deep semantic relationship, and relates to the technical field of big data processing. Comprising the following steps: 1, collecting and preprocessing a data source: collecting related data of talents in different channels, and carrying out preprocessing operations of cleaning, denoising, word segmentation and part-of-speech tagging on the related data of the talents to form a data text; the triple information is name-skill-proficiency or project-role-contribution, (3) carrying out triple association analysis and integration, (4) constructing a talent portrait according to the triple, (5) evaluating and optimizing the talent portrait, and (6) applying the constructed talent portrait and obtaining application feedback. A multi-dimensional and accurate talent portrait is formed, and a scientific basis is provided for enterprise recruitment and talent evaluation and management.

Description

Technical Field

[0001] The invention discloses a talent portrait construction method based on deep semantic relationship mining triples, and relates to the technical field of big data processing. Background Art

[0002] As corporate competition intensifies, talent has become the core resource for corporate development. However, existing talent assessment methods are often limited by single data and limited analytical methods, making it difficult to fully and accurately reflect the true capabilities and potential of talent. Summary of the invention

[0003] In response to the problems of the prior art, the present invention provides a talent portrait construction method based on deep semantic relationship mining triples, and utilizes triple mining technology based on deep semantic relationships to construct talent portraits. The invention aims to automatically extract key information from massive data sources such as resumes, work results, social media content, etc., to form a multi-dimensional and accurate talent portrait, and provide a scientific basis for corporate recruitment, talent assessment and management.

[0004] The specific scheme proposed by the present invention is:

[0005] The present invention provides a talent portrait construction method based on deep semantic relationship mining triples, comprising:

[0006] Step 1: Collect and preprocess data sources: Collect talent-related data from different channels, perform preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text.

[0007] Step 2: Use deep learning models to mine talent triplet information, which is name-skill-proficiency or project-role-contribution.

[0008] Step 3: Perform triple association analysis and integration: Use graph neural network (GNN) to perform association analysis on the extracted triple information, build a semantic network of talent characteristics, and remove duplicates and merge the triples in the semantic network.

[0009] Step 4: Construct talent portraits based on triples: Construct talent portraits based on the extracted triple information and semantic network, combined with preset talent portrait templates and personalized needs. The talent portrait content includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations.

[0010] Step 5: Evaluate and optimize talent profiles,

[0011] Step 6: Build a talent profile for the application and obtain application feedback.

[0012] Furthermore, the talent-related data collected in step 1 of the talent portrait construction method based on deep semantic relationship mining triples include personal resumes, work results, social media content, and professional papers.

[0013] Furthermore, in step 2 of the talent portrait construction method based on deep semantic relationship mining triples, a deep learning model BERT or Transformer is used to perform deep semantic understanding on the preprocessed data text to identify entities, relationships and attributes in the data text.

[0014] Through semantic parsing and relationship extraction methods, triple information with clear semantic relations is automatically extracted from data text.

[0015] Furthermore, in step 5 of the talent portrait construction method based on deep semantic relationship mining triples, the constructed talent portrait is verified and optimized through expert evaluation and user feedback to ensure the accuracy and practicality of the portrait.

[0016] Iteratively optimize the talent profile construction method based on the evaluation results.

[0017] Furthermore, in step 6 of the talent portrait construction method based on deep semantic relationship mining triples, the constructed talent portrait is applied to the scenarios of corporate recruitment, talent assessment and management, and user feedback is collected to continuously optimize and improve the talent portrait.

[0018] The present invention also provides a talent portrait construction device based on deep semantic relationship mining triples, including a collection and preprocessing module, a data mining module, an analysis module, a talent portrait management module, an evaluation module and an application module.

[0019] The collection and preprocessing module collects and preprocesses data sources: collects talent-related data from different channels, performs preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text.

[0020] The data mining module uses deep learning models to mine talent triple information, which is name-skill-proficiency or project-role-contribution.

[0021] The analysis module performs triple association analysis and integration: the graph neural network GNN is used to perform association analysis on the extracted triple information, construct a semantic network of talent characteristics, and remove duplicates and merge triples in the semantic network.

[0022] The talent portrait management module constructs talent portraits based on triples: Based on the extracted triple information and semantic network, combined with the preset talent portrait template and personalized needs, the talent portrait is constructed. The content of the talent portrait includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations.

[0023] The assessment module evaluates and optimizes talent profiles.

[0024] The application module builds talent profiles and obtains application feedback.

[0025] Furthermore, the talent-related data collected by the collection and preprocessing module of the talent portrait construction device based on deep semantic relationship mining triples include personal resumes, work results, social media content, and professional papers.

[0026] Furthermore, the data mining module of the talent portrait construction device based on deep semantic relationship mining triples uses the deep learning model BERT or Transformer to perform deep semantic understanding on the preprocessed data text, identify entities, relationships and attributes in the data text,

[0027] Through semantic parsing and relationship extraction methods, triple information with clear semantic relations is automatically extracted from data text.

[0028] Furthermore, the evaluation module of the talent portrait construction device based on deep semantic relationship mining triples verifies and optimizes the constructed talent portrait through expert evaluation and user feedback to ensure the accuracy and practicality of the portrait.

[0029] Iteratively optimize the talent profile construction method based on the evaluation results.

[0030] Furthermore, the application module of the talent portrait construction device based on deep semantic relationship mining triples applies the constructed talent portrait to the scenarios of enterprise recruitment, talent evaluation and management, and collects user feedback to continuously optimize and improve the talent portrait.

[0031] The benefits of the present invention are:

[0032] It can integrate and clean personnel information in multiple dimensions and mine deep semantic relationships to build a comprehensive, multi-dimensional talent portrait and provide enterprises with rich talent information.

[0033] Based on the unique characteristics and needs of talents, personalized portrait templates are constructed to achieve accurate portraits; the automated processing flow reduces the cost and time of manual review and improves the efficiency of talent assessment; deep learning and semantic analysis technology are used to achieve in-depth understanding and analysis of complex text data, thereby improving the intelligence level of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0035] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0036] Example 1

[0037] The present invention provides a talent portrait construction method based on deep semantic relationship mining triples, comprising:

[0038] Step 1: Collect and preprocess data sources: Collect talent-related data from different channels, and perform preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text. The collected talent-related data may include personal resumes, work results, social media content, professional papers, etc.

[0039] Step 2: Use deep learning models to mine talent triplet information, which is name-skill-proficiency or project-role-contribution.

[0040] The deep learning model BERT or Transformer can be further used to perform deep semantic understanding on the preprocessed data text to identify entities, relationships and attributes in the data text.

[0041] Through semantic parsing and relationship extraction methods, triple information with clear semantic relations is automatically extracted from data text.

[0042] Step 3: Perform triple association analysis and integration: Use graph neural network (GNN) to perform association analysis on the extracted triple information, build a semantic network of talent characteristics, and deduplicate and merge the triples in the semantic network.

[0043] Step 4: Construct talent portraits based on triples: Based on the extracted triple information and semantic network, combined with the preset talent portrait template and personalized needs, construct talent portraits. The content of the talent portrait includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations.

[0044] Step 5: Evaluate and optimize talent portraits, verify and optimize the constructed talent portraits through expert evaluation and user feedback to ensure the accuracy and practicality of the portraits.

[0045] Iteratively optimize the talent profile construction method based on the evaluation results.

[0046] Step 6: Apply the constructed talent portrait and obtain application feedback. The constructed talent portrait can be applied to the scenarios of enterprise recruitment, talent assessment and management, and user feedback can be collected to continuously optimize and improve the talent portrait.

[0047] Example 2

[0048] The present invention also provides a talent portrait construction device based on deep semantic relationship mining triples, including a collection and preprocessing module, a data mining module, an analysis module, a talent portrait management module, an evaluation module and an application module.

[0049] The collection and preprocessing module collects and preprocesses data sources: collects talent-related data from different channels, performs preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text.

[0050] The data mining module uses deep learning models to mine talent triple information, which is name-skill-proficiency or project-role-contribution.

[0051] The analysis module performs triple association analysis and integration: the graph neural network GNN is used to perform association analysis on the extracted triple information, construct a semantic network of talent characteristics, and remove duplicates and merge triples in the semantic network.

[0052] The talent portrait management module constructs talent portraits based on triples: Based on the extracted triple information and semantic network, combined with the preset talent portrait template and personalized needs, the talent portrait is constructed. The content of the talent portrait includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations.

[0053] The assessment module evaluates and optimizes talent profiles.

[0054] The application module builds talent profiles and obtains application feedback.

[0055] As the information interaction and execution process between the modules of the above-mentioned device are based on the same concept as the embodiment of the method of the present invention, the specific contents can be found in the description of the embodiment of the method of the present invention and will not be repeated here.

[0056] Similarly, the device of the present invention can perform multi-dimensional integration and cleaning of personnel information and deep semantic relationship mining to construct a comprehensive, multi-dimensional talent portrait and provide enterprises with rich talent information.

[0057] Based on the unique characteristics and needs of talents, personalized portrait templates are constructed to achieve accurate portraits; the automated processing flow reduces the cost and time of manual review and improves the efficiency of talent assessment; deep learning and semantic analysis technology are used to achieve in-depth understanding and analysis of complex text data, thereby improving the intelligence level of the system.

[0058] It should be noted that not all steps and modules in the above-mentioned processes and device structures are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.

[0059] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A talent portrait construction method based on deep semantic relationship mining triples, which is characterized by include: Step 1: Collect and preprocess data sources: Collect talent-related data from different channels, perform preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text. Step 2: Use deep learning models to mine talent triplet information, which is name-skill-proficiency or project-role-contribution. Step 3: Perform triple association analysis and integration: Use graph neural network (GNN) to perform association analysis on the extracted triple information, build a semantic network of talent characteristics, and remove duplicates and merge the triples in the semantic network. Step 4: Construct talent portraits based on triples: Construct talent portraits based on the extracted triple information and semantic network, combined with preset talent portrait templates and personalized needs. The talent portrait content includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations. Step 5: Evaluate and optimize talent profiles, Step 6: Build a talent profile for the application and obtain application feedback.

2. The talent portrait construction method based on deep semantic relationship mining triples according to claim 1 is characterized by: The talent-related data collected in step 1 includes personal resumes, work achievements, social media content, and professional papers.

3. The talent portrait construction method based on deep semantic relationship mining triples according to claim 1 is characterized by: In step 2, the deep learning model BERT or Transformer is used to perform deep semantic understanding on the preprocessed data text to identify entities, relationships, and attributes in the data text. Through semantic parsing and relationship extraction methods, triple information with clear semantic relations is automatically extracted from data text.

4. The talent portrait construction method based on deep semantic relationship mining triples according to claim 1 is characterized by: In step 5, the constructed talent profile is verified and optimized through expert evaluation and user feedback to ensure the accuracy and practicality of the profile. Iteratively optimize the talent profile construction method based on the evaluation results.

5. According to the talent portrait construction method based on deep semantic relationship mining triples as described in claim 1, it is characterized in that the constructed talent portrait in step 6 is applied to the scenarios of corporate recruitment, talent evaluation and management, and user feedback is collected to continuously optimize and improve the talent portrait.

6. A talent portrait construction device based on deep semantic relationship mining triples, characterized by: It includes collection and preprocessing module, data mining module, analysis module, talent profile management module, evaluation module and application module. The collection and preprocessing module collects and preprocesses data sources: collects talent-related data from different channels, performs preprocessing operations such as cleaning, denoising, word segmentation, and part-of-speech tagging on the talent-related data to form data text. The data mining module uses deep learning models to mine talent triple information, which is name-skill-proficiency or project-role-contribution. The analysis module performs triple association analysis and integration: the graph neural network GNN is used to perform association analysis on the extracted triple information, construct a semantic network of talent characteristics, and remove duplicates and merge triples in the semantic network. The talent portrait management module constructs talent portraits based on triples: Based on the extracted triple information and semantic network, combined with the preset talent portrait template and personalized needs, the talent portrait is constructed. The content of the talent portrait includes basic information, professional skills, work experience, project results, personal characteristics, and career inclinations. The assessment module evaluates and optimizes talent profiles. The application module builds talent profiles and obtains application feedback.

7. The talent portrait construction device based on deep semantic relationship mining triples according to claim 6 is characterized by: The talent-related data collected by the collection and preprocessing module includes personal resumes, work results, social media content, and professional papers.

8. The talent portrait construction device based on deep semantic relationship mining triples according to claim 6 is characterized by: The data mining module uses the deep learning model BERT or Transformer to perform deep semantic understanding on the preprocessed data text and identify the entities, relationships and attributes in the data text. Through semantic parsing and relationship extraction methods, triple information with clear semantic relations is automatically extracted from data text.

9. The talent portrait construction device based on deep semantic relationship mining triples according to claim 6, characterized in that The evaluation module verifies and optimizes the constructed talent portrait through expert evaluation and user feedback to ensure the accuracy and practicality of the portrait. Iteratively optimize the talent profile construction method based on the evaluation results.

10. The talent portrait construction device based on deep semantic relationship mining triples according to claim 6, characterized in that The application module will apply the constructed talent portrait to the scenarios of corporate recruitment, talent assessment and management, and collect user feedback to continuously optimize and improve the talent portrait.

Images