Employee self-adaptive career path analysis method, device, equipment and medium

Abstract

The present application relates to the technical field of data analysis, which can be applied to the business system platform of financial technology, medical health and the like, and discloses an employee self-adaptive career path analysis method, device, equipment and medium. The method comprises the following steps: obtaining multi-modal work data of a target employee, performing data labeling and mode fusion on the multi-modal work data to obtain target work data; constructing a work knowledge graph according to the target work data and extracting corresponding multi-dimensional work features; performing multi-dimensional feature analysis on the target employee according to the multi-dimensional work features to obtain multi-dimensional work capacity quantitative characteristic values; generating structured employee capacity portrait data according to the multi-dimensional work capacity quantitative characteristic values and the target work data, performing Markov career decision analysis, obtaining a career development full connection graph, and generating a self-adaptive career graph path of the target employee based on the career development full connection graph. The present application can improve the accuracy of enterprise employee career path analysis.

Description

Technical Field

[0001] This invention relates to the field of data analysis, and in particular to a method, apparatus, device, and medium for analyzing employee adaptive career paths. Background Technology

[0002] In the process of digital transformation of enterprise human resource management, career-related analysis based on employee work data has become an important foundation for supporting enterprise talent development decisions. Its core objective is to provide technical support for quantitative assessment of employee career development through data mining and analysis.

[0003] However, existing career path analysis methods for corporate employees have significant shortcomings in terms of data analysis accuracy, making it difficult for the generated career-related analysis results, such as career development paths, to meet practical application needs.

[0004] On the one hand, traditional methods often rely on limited and single data sources, such as basic information like employees' years of service and job level, which makes it difficult to comprehensively and accurately reflect employees' actual work abilities and potential. If these are not fully explored and utilized, it can lead to a one-sided evaluation of employees. On the other hand, existing technologies lack effective integration and analysis methods when processing this complex data. Furthermore, traditional methods for analyzing employee work abilities often use simple qualitative evaluations and lack scientific quantitative standards, making it difficult to obtain accurate multi-dimensional quantitative scores of work abilities. Employee ability profiles generated based on inaccurate quantitative scores cannot truly present employees' strengths and weaknesses, and the resulting adaptive career paths are unlikely to align with employees' actual needs and corporate development goals.

[0005] For example, in the fintech field, employee career path analysis mostly relies on single job data such as professional qualification certificates and the number of projects participated in. It is difficult to accurately measure employees' abilities and potential in complex tasks such as algorithm development and risk modeling. As a result, the analysis of employees' career paths is not accurate enough, which makes it difficult to meet the needs of enterprises for innovative fintech talents and hinders the improvement of enterprise business innovation and competitiveness.

[0006] For example, in the healthcare field, physician career path analysis mostly relies on single work data such as job title and clinical work hours, which cannot comprehensively assess physicians' abilities and potential in complex disease diagnosis and treatment, multidisciplinary collaboration, etc. This results in inaccurate physician career path analysis, which in turn causes physicians with potential for scientific research to be limited to routine clinical work and unable to adapt to the needs of the healthcare industry's transformation towards intelligence and precision.

[0007] Therefore, improving the accuracy of career path analysis for corporate employees has become an urgent problem to be solved. Summary of the Invention

[0008] This invention provides a method, apparatus, equipment, and medium for employee adaptive career path analysis, the main purpose of which is to solve the problem of low accuracy in career path analysis of enterprise employees.

[0009] Firstly, to achieve the above objectives, the present invention provides an employee adaptive career path analysis method, comprising: Acquire multimodal work data of the target employee, and perform annotation and modality fusion on the multimodal work data to obtain the target work data; Construct a job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; Based on the multi-dimensional work characteristics, the target employee is analyzed for multi-dimensional work characteristics to obtain the corresponding multi-dimensional work ability quantitative characteristic value. Based on the multidimensional work ability quantitative feature values ​​and the target work data, a structured employee ability profile of the target employee is generated; Markov career decision analysis was performed on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development. Adaptive path generation is performed based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

[0010] Secondly, the present invention also provides an employee adaptive career path analysis device, comprising: The data annotation and fusion module is used to acquire multimodal work data of the target employee, and to annotate and fuse the multimodal work data to obtain the target work data. The job knowledge graph construction module is used to construct the job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; The job feature analysis module is used to perform multi-dimensional job feature analysis on the target employee based on the multi-dimensional job features, and obtain the corresponding multi-dimensional work ability quantitative feature value. The competency profile generation module is used to generate structured employee competency profile data of the target employee based on the multidimensional work competency quantitative feature values ​​and the target work data; The career decision analysis module is used to perform Markov career decision analysis on the structured employee competency profile data to obtain the full connectivity graph of the target employee's career development. The graph path generation module is used to generate adaptive paths based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

[0011] Thirdly, the present invention also provides an electronic device, the electronic device comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the employee adaptive career path analysis method described above.

[0012] Fourthly, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the employee adaptive career path analysis method described above.

[0013] In this embodiment of the invention, multimodal work data is collected in full coverage, breaking through the limitations of a single data dimension. Structured annotation provides standardized input for the computer, reducing the interference of data noise on subsequent processing. Modality fusion technology is used to achieve feature complementarity of heterogeneous data (such as text, voice, and behavior logs), enhancing the integrity and accuracy of data representation and ensuring the accuracy and real-time performance of subsequent computer analysis of employee work. A work knowledge graph is constructed through target work data, transforming unstructured data into a structured association network that can be parsed by the computer. With the help of graph visualization and association rule mining algorithms, the computer can automatically extract multi-dimensional features, overcoming the subjectivity and limitations of manual feature engineering and providing interpretable data support for employee work analysis.

[0014] This process transforms multi-dimensional job characteristics into quantitative feature values, avoiding the subjectivity and errors of manual assessment, improving the objectivity of data processing and the accuracy of capability analysis. The resulting structured employee capability profile data comprehensively and objectively reflects the capability characteristics of target employees, enhancing the accuracy of subsequent career path analysis. A Markov model is used to perform career decision analysis on the structured employee capability profile data, improving the computational efficiency of decision analysis and avoiding the subjectivity of manual decision-making. The generated career development fully connected graph is stored in a graphical data structure, improving data visualization and interpretation efficiency. Adaptive path generation is performed based on the career development fully connected graph. Computer algorithms automatically traverse the nodes and relationships in the graph, completing path search and filtering without manual intervention, improving the automation of career path analysis while avoiding the limitations of fixed algorithms, enhancing the accuracy of path matching, and thus improving the accuracy of career path analysis for target employees. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of an application environment for an employee adaptive career path analysis method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating an employee adaptive career path analysis method according to an embodiment of the present invention. Figure 3 This is a schematic diagram of an embodiment of the present invention for adaptive path generation based on a fully connected graph of career development; Figure 4 This is a functional block diagram of an employee adaptive career path analysis device provided in an embodiment of the present invention; Figure 5 A schematic diagram of an electronic device for implementing an employee adaptive career path analysis method according to an embodiment of the present invention; Figure 6 This is another structural schematic diagram of an electronic device that implements an employee adaptive career path analysis method according to an embodiment of the present invention.

[0017] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0019] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0020] This application provides an employee adaptive career path analysis method. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the device provided in this application: a server, a terminal, or other similar device. In other words, the employee adaptive career path analysis method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or 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 (CDNs), and big data and artificial intelligence platforms.

[0021] This invention discloses an employee adaptive career path analysis method, which can be applied to, for example... Figure 1 In this application environment, the client communicates with the server via a network. The server can obtain multimodal work data of the target employees through the client, breaking through the limitations of single data dimensions. Structured annotation provides standardized input to the computer, reducing the interference of data noise on subsequent processing. Modality fusion technology is used to achieve feature complementarity of heterogeneous data (such as text, voice, and behavioral logs), enhancing the completeness and accuracy of data representation and ensuring the accuracy and real-time nature of subsequent computer analysis of employee work. A work knowledge graph is constructed using the target work data, transforming unstructured data into a structured relational network that can be parsed by the computer. With the help of graph visualization and association rule mining algorithms, the computer can automatically extract multi-dimensional features, overcoming the subjectivity and limitations of manual feature engineering and providing interpretable data support for employee work analysis.

[0022] This process involves transforming multi-dimensional job characteristics into quantitative feature values, avoiding the subjectivity and errors of manual assessment, improving the objectivity of data processing and the accuracy of capability analysis. The generated structured employee capability profile data comprehensively and objectively reflects the capability characteristics of target employees, improving the accuracy of subsequent career path analysis. A Markov model is used to perform career decision analysis on the structured employee capability profile data, improving the computational efficiency of decision analysis and avoiding the subjectivity of manual decision-making. The generated career development fully connected graph is stored in a graphical data structure, improving data visualization and interpretation efficiency. Adaptive path generation is performed based on the career development fully connected graph. Computer algorithms automatically traverse the nodes and relationships in the graph, completing path search and filtering without manual intervention, improving the automation of career path analysis, avoiding the limitations of fixed algorithms, enhancing the accuracy of path matching, and thus improving the accuracy of career path analysis for target employees. Finally, the adaptive career graph path output is fed back to the client.

[0023] The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will now be described in detail through specific embodiments.

[0024] Reference Figure 2 The diagram shown is a flowchart illustrating an employee adaptive career path analysis method according to an embodiment of the present invention. In this embodiment, the employee adaptive career path analysis method includes: S1. Obtain multimodal work data of the target employee, and label and fuse the multimodal work data to obtain target work data.

[0025] In this embodiment of the invention, the multimodal work data includes the target employee's basic personal data, business operation data, performance result data, learning behavior data, and market environment data. The basic personal data includes structured data such as the target employee's resume, qualification certificates, and work experience. The business operation data includes system logs such as business processing logs. The performance result data includes performance indicators and customer satisfaction. The learning behavior data includes learning completion rate and learning test scores. The market environment data includes industry trend data and regulatory policy data. Specifically, the automated acquisition and standardization of various modal data can be achieved through API interfaces, natural language processing, speech recognition, and other means.

[0026] In this embodiment of the invention, the step of labeling and modal fusion of the multimodal working data to obtain target working data includes: The multimodal working data is cleaned to generate cleaned multimodal working data; Extract the text data and image data from the cleaning multimodal working data; The text working data is segmented and part-of-speech tagged to obtain structured text annotation data; Semantic feature point annotation is performed on the image working data to obtain structured image annotation data; Based on preset modal alignment rules, the structured text annotation data and the structured image annotation data are aligned in both time and spatial dimensions to form aligned multimodal annotation data; Attention-weighted fusion is performed on the aligned multimodal labeled data to obtain the target working data.

[0027] In this embodiment of the invention, the data cleaning includes missing value processing, outlier detection and correction, data deduplication, and data format unification. Specifically, missing value processing involves traversing the multimodal working data and checking for missing values ​​in each item. For text-based working data, if some paragraphs in a work report are missing, they can be reasonably inferred and supplemented based on the context, or by referring to the content of other reports on similar topics. For image-based working data, if some pixels in a work scene photo are missing, image restoration algorithms can be used to recover the missing parts based on surrounding pixel information.

[0028] The data format unification includes converting multimodal working data from different sources and in different formats into a unified standard format; for example, converting various text working data in different formats into plain text format, and converting image working data with different resolutions and color modes into a standard image format with a specific resolution and color mode, so as to facilitate subsequent processing and analysis; after the above processing, cleaned multimodal working data is obtained.

[0029] Specifically, by examining the file extensions and header information of the multimodal working data, text and image data files are quickly identified to obtain the corresponding text and image working data. The text working data is segmented and tagged with parts of speech using word segmentation algorithms and part-of-speech tagging models from natural language processing. Specifically, a rule-based segmentation method is employed, which segments the text working data according to a pre-defined vocabulary dictionary and grammatical rules. For example, based on common Chinese word combination rules, a continuous sequence of Chinese characters is segmented into meaningful words to obtain the segmented text working data.

[0030] The part-of-speech tagging is performed using a pre-trained part-of-speech tagging model. This model is trained on a large corpus of text with pre-tagged parts of speech. The segmented text data is input into the part-of-speech tagging model, and the model assigns corresponding part-of-speech tags, such as nouns, verbs, and adjectives, to each word based on its semantic and grammatical features in the context. For example, for the word "work," in the sentence "He is working hard," the model will tag it as a verb based on the context; in the sentence "This job is very important," the model will tag it as a noun. After word segmentation and part-of-speech tagging, structured text tagging data is obtained.

[0031] In detail, semantic feature point annotation is performed on the image working data, that is, the convolutional neural network (CNN) in deep learning is used to extract features from the image working data. The CNN model automatically learns the hierarchical features in the image working data through multiple convolutional layers, pooling layers and other structures. Based on the extracted features, combined with semantic understanding technology, the semantically meaningful feature points in the image are determined.

[0032] These feature points represent key information in the image data, such as the performance completion rate of a target employee in a performance completion chart. By manually annotating these feature points or using a pre-trained semantic annotation model, corresponding semantic labels are added, resulting in structured image annotation data.

[0033] Furthermore, based on the time information of the work events recorded in the text work data and image work data, corresponding alignment rules are formulated; for example, if the text records the start and end time of a certain work task, while the image records a scene of a key moment in the execution of the task, alignment can be performed according to the chronological order and logical relationship.

[0034] Specifically, the time information in the text and images is organized into a time series. For the time information in the text, key time points are extracted. For the images, they can be sorted according to the acquisition time or the time sequence reflected by the image content. By analyzing the similarity and correlation of the time series, the time points in the text and images corresponding to the same work stage or event are aligned in the time dimension. Using feature matching algorithms in image processing, such as SIFT (Scale Invariant Feature Transform) and SURF (Speeded Robust Feature Transform), feature points in the images are extracted, and the similarity between feature points is calculated. By pre-matching the spatial features described in the text and the feature points in the images, the text and images are aligned in the spatial dimension. After double alignment in the time and spatial dimensions, aligned multimodal labeled data is obtained.

[0035] In detail, in aligned multimodal labeled data, the importance of data from different modalities and different parts of the same modal data in reflecting the work situation may vary. An attention mechanism model is constructed. This model learns the correlation and importance between different data by learning a large amount of labeled multimodal work data. The aligned multimodal labeled data is input into the trained attention mechanism model, and the model will assign an attention weight to each data item according to the content and context information of the data.

[0036] The higher the weight, the greater the impact of the data item on the final result during the fusion process. Then, according to the assigned attention weight, the aligned multimodal labeled data is weighted and fused to organically combine the data of different modalities according to their importance to obtain the target working data.

[0037] For example, when fusing text and image data, if a key piece of information in the text has a higher importance weight, the related parts in the image will also be more fully reflected in the fusion process, thereby generating more comprehensive and accurate target work data that reflects the target employee's work situation.

[0038] In this embodiment of the invention, multimodal work data is collected in full coverage, breaking through the limitations of a single data dimension. Structured annotation provides standardized input for the computer, reducing the interference of data noise on subsequent processing and improving data usability. Modality fusion technology is used to achieve feature complementarity of heterogeneous data (such as text, voice, and behavior logs), enhancing the integrity and accuracy of data representation and ensuring the accuracy and real-time performance of subsequent computer analysis of employee work.

[0039] S2. Construct a work knowledge graph of the target employee based on the target work data, and extract multi-dimensional work features from the work knowledge graph.

[0040] In this embodiment of the invention, the work knowledge graph is a structured semantic network that organizes and represents the work-related knowledge of the target employee in the form of a graph. It is presented in the form of nodes and edges, where nodes represent a set of standardized work entities and edges refer to the weighted association relationships between standardized work entities. The multi-dimensional work features include task-dimensional features, skill-dimensional features, result-dimensional features, relationship-dimensional features, etc., specifically including features such as work type, work difficulty, and work completion time.

[0041] In this embodiment of the invention, constructing the job knowledge graph of the target employee based on the target job data includes: The target working data is parsed to obtain information on multiple working entities; The information of multiple work entities is normalized to generate a standardized set of work entities; Identify the relationships between standardized work entities in the set of standardized work entities; Configure weight parameters for each of the aforementioned relationships to obtain weighted relationships; Construct the initial graph topology of the target employee based on the standardized set of working entities and the weighted association relationships; The initial graph topology is subjected to conflict detection and graph enhancement to generate the target employee's work knowledge graph.

[0042] In this embodiment of the invention, for text content in target work data, such as work reports and emails, named entity recognition technology is used. This technology is based on a pre-trained model, which learns from a large amount of labeled text data to master the characteristics of different types of entities. When processing text, the model analyzes word by word or sentence by sentence to identify work entities such as names of people (e.g., employee names, colleague names), place names (work locations), organization names (departments, partner companies), work task names (e.g., project names, specific task titles), and work tool names (e.g., software, equipment used).

[0043] For target work data that already has a certain structure, such as work time and work output recorded in tabular form, data structure analysis methods are used to parse the data according to the field definitions and logical relationships. For example, in a table recording employees' daily work time, the work entity information such as employee name, date, and work time can be extracted by analyzing the row and column structure of the table. By combining the parsing results of text and structured data, multiple work entity information can be obtained.

[0044] In detail, for the identified work entity information, there may be cases where the expressions are different but the semantics are the same or similar; for example, "laptop" and "notebook" may refer to the same entity in a work scenario. Through semantic similarity calculation methods, which are based on word vector models, each work entity is converted into vector form, and then the similarity between these vectors is calculated. When the similarity between the vectors of two work entities exceeds a certain threshold, they are determined to be semantically similar, and semantically similar work entities are mapped to a unified standard name. Through semantic similarity calculation and rule mapping, multiple work entity information is normalized into a standardized set of work entities, eliminating the differences and ambiguities in entity expressions.

[0045] Specifically, the semantic logical relationships between work entities are analyzed. For example, there is a semantic relationship of "undertaking" between "Employee A" and "XX Project", and a semantic relationship of "used for" between "XX Software" and "XX Project". Through semantic role labeling and semantic relationship analysis techniques in natural language processing, the semantic associations between work entities are identified, and then the associations between standardized work entities in the standardized work entity set are identified.

[0046] In this process, weights are assigned based on factors such as the frequency of occurrence, importance, and semantic strength of the associations. Specifically, a combination of expert evaluation and data statistics is used to configure reasonable weight parameters for each association, resulting in weighted associations.

[0047] Furthermore, using graph construction methods from graph theory, standardized work entities are used as nodes in the graph, and weighted relationships are used as edges. Each work entity in the set of standardized work entities is treated as a node, and corresponding attributes are set for each node, such as node name and node type (person name, task name, tool name, etc.). Based on the weighted relationships, edges are created between the corresponding nodes, and the direction of the edges is determined according to the semantic logic of the relationship. For example, in the relationship "employees undertake tasks", the edge points from the "employee" node to the "task" node. At the same time, the weight parameter is used as an attribute of the edge to represent the strength of the relationship.

[0048] For example, if the association between "Employee A" and "XX Core Project" has a high weight, then the weight attribute value of the edge connecting these two nodes will also be high. By creating nodes and edges, an initial graph topology structure of the target employee is constructed. This structure intuitively displays the association and relative importance between work entities in the form of a graph.

[0049] The process involves conflict detection and graph enhancement of the initial graph topology. Conflict detection employs rule checking and logical reasoning methods, while graph enhancement utilizes data supplementation and relationship expansion methods. The process checks whether there are any rule violations in the initial graph topology. For example, if a rule stipulates that an employee cannot simultaneously undertake two mutually exclusive tasks, and such a situation occurs in the graph, it is considered a conflict. By traversing the nodes and edges in the graph, the process checks whether these rules are satisfied and identifies potential conflicts.

[0050] The graph enhancement refers to obtaining more information related to the target employee's work from external data sources or other relevant internal data and adding it to the initial graph topology; for example, obtaining the employee's skill certificate information from the company's human resources system, adding it to the graph and associating it with the employee node, thereby obtaining the target employee's work knowledge graph.

[0051] In this embodiment of the invention, the job knowledge graph is preprocessed, including removing noisy data, standardizing the attribute information of nodes and edges, and using a feature extraction model based on graph embedding algorithm to map the nodes and edges in the job knowledge graph to a low-dimensional vector space to obtain an initial feature vector. According to a preset multi-dimensional partitioning rule, multi-dimensional job features related to job content, job skills, job relationships, job environment, and career development are extracted from the initial feature vector. The extracted multi-dimensional job features are then fused and optimized to obtain the final multi-dimensional job feature set.

[0052] In this embodiment of the invention, a work knowledge graph is constructed by using target work data, transforming unstructured data into a structured association network that can be parsed by computers, reducing the complexity of data storage and retrieval, and improving retrieval efficiency; with the help of graph visualization and association rule mining algorithms, computers can automatically extract multi-dimensional features, overcoming the subjectivity and limitations of manual feature engineering, and providing interpretable data support for employee work analysis.

[0053] S3. Perform multidimensional work characteristic analysis on the target employee based on the multidimensional work characteristics to obtain the corresponding multidimensional work ability quantitative characteristic value.

[0054] In this embodiment of the invention, the multidimensional job feature analysis refers to the process of in-depth exploration of multidimensional job features, uncovering the inherent connections and patterns between multidimensional job features, and revealing the characteristics, strengths, and weaknesses of the target employee's work; the multidimensional job capability quantification feature value refers to the result of representing the target employee's job capability obtained through multidimensional job feature analysis in a quantified numerical form. These quantified feature values ​​can more intuitively and accurately reflect the level and degree of the target employee in different job capabilities.

[0055] In this embodiment of the invention, the step of performing multidimensional work characteristic analysis on the target employee based on the multidimensional work characteristics to obtain corresponding multidimensional work ability quantitative feature values ​​includes: Analyze the multi-dimensional work characteristics and determine the capability feature dimensions corresponding to the multi-dimensional work characteristics based on the analysis results; Based on the aforementioned capability characteristic dimensions, a corresponding quantitative analysis indicator system is constructed, and quantifiable characteristic indicators, unstructured characteristic correlation indicators, and potential characteristic indicators in the quantitative analysis indicator system are identified. The first quantitative feature value is obtained by performing time-series trend analysis on the multi-dimensional work characteristics using the quantifiable feature indicators. The unstructured feature correlation index is used to perform work behavior feature analysis on the multi-dimensional work features to obtain the second quantitative feature value; The potential capability characteristics of the multi-dimensional work characteristics are analyzed using the potential characteristic indicators to obtain the third quantitative characteristic value; The first quantitative feature value, the second quantitative feature value, and the third quantitative feature value are weighted and summed according to the weight coefficients of the quantitative analysis indicator system to obtain the multidimensional work capability quantitative feature value.

[0056] In this embodiment of the invention, for structured or semi-structured multi-dimensional work characteristic data, such as data on the type, quantity, and time of tasks completed by employees, the data is classified and summarized according to its attributes and logical relationships, grouping similar types of data into one category. For example, all data related to task completion is grouped into one category, from which characteristics related to task execution ability are analyzed; data involving employee collaboration with others is grouped into another category, from which characteristics related to teamwork ability are analyzed. Based on the comprehensive results of data classification and summarization, the corresponding capability characteristic dimensions for the multi-dimensional work characteristics are determined, such as leadership ability, innovation ability, marketing ability, task execution ability, and teamwork ability.

[0057] In detail, based on the identified capability characteristics, and in conjunction with industry standards and actual work conditions, a quantitative analysis indicator system is constructed. For example, for the task execution capability dimension, experts may set indicators such as task completion rate, on-time task completion rate, and task quality score; for the team collaboration capability dimension, indicators such as the number of collaborative projects and team member evaluations may be set.

[0058] Among them, in the constructed quantitative analysis indicator system, indicators that can be directly measured with specific values ​​are selected to quantify the characteristic indicators. For example, the task completion rate can be calculated by the ratio of the number of completed tasks to the total number of tasks, and the on-time completion rate of tasks can be determined by the ratio of the number of on-time completed tasks to the total number of tasks, thus obtaining the first quantitative characteristic value.

[0059] The analysis and quantitative analysis index system is used to identify indicators associated with unstructured features such as text and images, and extract features such as sentiment and information richness to indirectly conduct quantitative evaluation and obtain the second quantitative feature value. Then, the latent feature indicators are used to discover the relationship between latent feature indicators and other work features through association rule mining and predictive analysis techniques. The predictive analysis technique is used to predict the potential abilities of target employees in the future based on existing data. Based on the predictive analysis results and combined with pre-set evaluation criteria, the potential ability features are quantitatively evaluated to obtain the third quantitative feature value.

[0060] Furthermore, the Analytic Hierarchy Process (AHP) is used to compare the relative importance of different types of indicators (quantifiable feature indicators, unstructured feature correlation indicators, and latent feature indicators) and specific indicators in the quantitative analysis indicator system pairwise, calculating the weight coefficients of each indicator. The first, second, and third quantitative feature values ​​are then multiplied by their corresponding weight coefficients, and the products are summed to obtain the multidimensional work ability quantitative feature value. This value comprehensively considers the target employee's performance in different types of indicators, providing a comprehensive and objective reflection of their multidimensional work ability level.

[0061] In this embodiment of the invention, multi-dimensional working characteristics are transformed into quantitative feature values, avoiding the subjectivity and error of manual evaluation, and improving the objectivity of data processing and the accuracy of capability analysis.

[0062] S4. Generate structured employee capability profile data of the target employee based on the multidimensional work capability quantitative feature values ​​and the target work data.

[0063] In this embodiment of the invention, the structured employee competency profile data refers to a data set presented in a structured form that can comprehensively and accurately describe the competency characteristics of the target employee. Specifically, it integrates multi-dimensional quantitative feature values ​​of work competency and target work data, providing a detailed description of the employee's competency from multiple perspectives (such as competency dimensions, work experience, performance, etc.). For example, the structured employee competency profile data may include the employee's basic information (name, age, position, etc.), quantitative scores of various competencies, key projects and achievements in work experience, areas of strength, and areas for improvement, presenting a clear and organized overview of the target employee's competency.

[0064] In this embodiment of the invention, generating structured employee competency profile data for the target employee based on the multidimensional work competency quantification feature values ​​and the target work data includes: Construct a mapping relationship between the target work data and the corresponding multidimensional work capability quantitative feature values; The target work data and the corresponding multidimensional work ability quantitative feature values ​​are filled into the preset employee ability profile data framework according to the association mapping relationship to obtain the initial employee ability profile data; The initial employee competency profile data is matched and verified, and the initial employee competency profile data is optimized and adjusted according to the matching and verification results to obtain structured employee competency profile data.

[0065] In this embodiment of the invention, data association analysis technology is used when constructing the correlation mapping relationship between target work data and multidimensional work ability quantitative feature values. First, the target work data is analyzed in detail to extract key information related to work ability, such as the difficulty of the work task, the quality of completion, and the required skills. At the same time, the ability dimensions and meanings represented by the multidimensional work ability quantitative feature values ​​are clarified. Then, by analyzing the inherent logical relationship between the two, such as the correlation between the completion status of a high-difficulty work task and the level of a specific ability dimension quantitative feature value, a precise correlation mapping relationship is established to determine the multidimensional work ability quantitative feature value corresponding to each target work data.

[0066] Specifically, when filling the preset employee competency profile data framework according to the association mapping relationship to obtain the initial employee competency profile data, data filling and integration technology is used. The preset employee competency profile data framework is a template with a fixed structure and fields, covering multiple aspects such as basic employee information, competency dimensions, and work experience. According to the previously established association mapping relationship, the target work data is accurately placed in the corresponding position in the framework, and at the same time, the multi-dimensional work competency quantitative feature values ​​are filled into the corresponding competency dimension fields.

[0067] In this process, it is essential to ensure the accuracy and completeness of the data, and to guarantee that each data point can be reasonably embedded into the framework, thereby forming preliminary initial employee capability profile data.

[0068] Furthermore, when matching, verifying, and optimizing the initial employee competency profile data to obtain structured employee competency profile data, data verification and optimization algorithms are used. In the matching and verification stage, the initial employee competency profile data is compared with the pre-set standard data model or industry specifications to check whether the data is logically consistent and whether there are any missing or incorrect data. For example, whether the quantitative characteristic values ​​of the competency dimension are within a reasonable range, and whether the work data matches the competency characteristics. If a mismatch is found, optimization and adjustment are carried out according to the specific situation. That is, for data errors, corrections are made according to the correct logic and standards.

[0069] After repeated verification and adjustments, the initial employee competency profile data became more accurate and reasonable, ultimately forming structured employee competency profile data that can comprehensively and objectively reflect the competency characteristics of target employees and improve the accuracy of subsequent career path analysis of target employees.

[0070] For example, in the healthcare field, a structured employee competency profile is generated using a radiologist (target employee) from a hospital. First, a mapping relationship is established, and target work data for the doctor is collected, such as the number of accurately diagnosed cases, the number of times they participated in difficult case discussions, and the duration of image report writing. Simultaneously, multi-dimensional quantitative feature values ​​for work competency are determined. Next, the work data and quantitative feature values ​​are populated into a pre-defined framework, which includes basic information and competency dimensions, resulting in initial profile data. Finally, matching and verification are performed, comparing the doctor's data with that of senior doctors in the same department. It is found that the doctor's diagnostic competency quantification value is relatively high, but their knowledge reserve quantification value is slightly low. Analysis indicates insufficient learning of new knowledge. Based on this, optimizations are made, such as increasing their participation in academic lectures and other learning opportunities, ultimately resulting in a structured employee competency profile that accurately reflects their capabilities.

[0071] For example, in the talent management and training scenario of the insurance industry in the fintech field, taking an insurance salesperson (target employee) as an example, firstly, a correlation mapping relationship is constructed, and target work data such as the number of successful sales, customer complaint rate, and customer renewal rate are collected. At the same time, multi-dimensional quantitative characteristic values ​​of work ability are determined. Then, these work data and quantitative characteristic values ​​are filled into a preset framework, which includes basic information, ability modules, etc., to obtain initial employee ability profile data. Finally, matching, verification, optimization and adjustment are performed to obtain structured employee ability profile data that can accurately reflect their abilities and help talent management and training decisions.

[0072] S5. Perform Markov career decision analysis on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development.

[0073] In this embodiment of the invention, the Markov career decision analysis refers to treating the career development states faced by employees as different "state points," such as different positions or different career stages. The probability of an employee transitioning from one career state to another depends only on the current state and is unrelated to how the current state was achieved in the past. By collecting and analyzing a large amount of historical data on employee career development, the transition probabilities between different career states are determined, thereby constructing a Markov model. This model is then used to predict and analyze the career development of target employees, and to assess the likelihood and risks of different career development paths.

[0074] In this embodiment of the invention, performing Markov career decision analysis on the structured employee competency profile data to obtain the target employee's fully connected career development graph includes: Identify the career decision state nodes and state transition probability matrix in the preset Markov career decision model, and identify the initial career state nodes of the target employee corresponding to the structured employee competency profile data. Extract the capability quantification requirement threshold corresponding to the career decision state node; The structured employee competency profile data is matched and analyzed with the corresponding competency quantification requirement threshold, and the set of achievable career status nodes for the target employee is determined based on the matching analysis results. The probability of the target employee transitioning from the initial career state node to each reachable career state node in the set of reachable career state nodes is calculated based on the state transition probability matrix. A fully connected graph of the target employee's career development is generated based on the initial career state nodes, the set of reachable career state nodes, and the transition probabilities.

[0075] In this embodiment of the invention, for the preset Markov career decision model, the structure and definition of the model are first analyzed in depth to clarify the career decision state nodes contained therein. These nodes represent different career stages, job types, etc. At the same time, the state transition probability matrix describes the probability of transitioning from one career state node to another. For the structured employee competency profile data, the current career information of the target employee recorded therein, such as the job position and job level, is compared with the career decision state nodes in the model to accurately identify the initial career state node of the target employee.

[0076] In the Markov career decision-making model, each career decision state node is assigned a corresponding threshold for quantitative ability requirements. These thresholds may include multiple dimensions such as professional skill level, years of work experience, and educational requirements. The various ability indicators of employees recorded in the structured employee ability profile data, such as professional skill scores and length of work experience, are compared one by one with the extracted threshold for quantitative ability requirements. By judging whether the employee's ability indicators meet or exceed the corresponding threshold for quantitative ability requirements, it is determined whether the target employee meets the conditions for entering the corresponding career state node. For career state nodes that meet the conditions, they are included in the set of attainable career state nodes, thereby clarifying the career state that the target employee may reach under the current ability level.

[0077] Furthermore, based on the target employee's initial career state node and the set of reachable career state nodes, the corresponding transition probability value is queried in the state transition probability matrix. Using professional graphics drawing tools or programming libraries, the initial career state node is taken as the starting point, and each node in the set of reachable career state nodes is taken as a potential transition target node. The thickness, color, and other attributes of the lines connecting the nodes are determined according to the calculated transition probabilities to intuitively represent the magnitude of the transition probabilities. By reasonably arranging the nodes and lines, a complete career development full connectivity graph is generated, clearly showing the other career states that the target employee may reach from the current career state and the corresponding transition probabilities, providing an intuitive and comprehensive reference for the employee's career planning.

[0078] For example, in the talent management and training scenario of the insurance industry, taking an insurance claims specialist (target employee) as an example, a pre-defined Markov career decision model is first identified, in which career decision state nodes include junior claims adjuster, intermediate claims adjuster, senior claims adjuster, claims supervisor, etc. The state transition probability matrix is ​​set based on industry experience and historical data. At the same time, the initial career state node of the claims specialist is determined to be junior claims adjuster. Then, the ability quantification requirement thresholds corresponding to each career decision state node are extracted. For example, intermediate claims adjusters need to have high case handling efficiency and accuracy. Then, the structured employee ability profile data of the specialist is matched and analyzed with the ability quantification requirement thresholds to determine the set of career state nodes that can be reached, such as intermediate claims adjuster. Then, based on the state transition probability matrix, the probability of transitioning from junior claims adjuster to intermediate claims adjuster is calculated. Finally, based on the initial career state node, the set of career state nodes that can be reached, and the transition probability, a full connectivity graph of the claims specialist's career development is generated.

[0079] In this embodiment of the invention, a Markov model is used to perform career decision analysis on structured employee competency profile data, which improves the computational efficiency of decision analysis, avoids the subjectivity of human decision-making, and stores the generated career development fully connected graph in a graphical data structure, which supports computers to quickly traverse career path nodes and improves the efficiency of data visualization and interpretation.

[0080] S6. Based on the fully connected career development graph, an adaptive path is generated to obtain the adaptive career graph path of the target employee.

[0081] In this embodiment of the invention, the adaptive career path is a set of paths that combine the target employee's work ability characteristics and career goals to select the most feasible and promising career development paths from among many possible career development paths. This path will be automatically adjusted and updated as the target employee's ability improves, the external environment changes, and other factors, so as to always maintain a high degree of adaptation to the target employee's career development status, thereby achieving the accuracy of the career path analysis of the target employee.

[0082] like Figure 3 As shown in this embodiment of the invention, the adaptive path generation based on the fully connected career development graph to obtain the adaptive career graph path of the target employee includes: Obtain all initial career graph paths in the fully connected career development graph and the style preference data of the target employee; The matching degree of all the initial occupational graph paths is calculated with the style preference data to obtain the initial path score of each initial occupational graph path; Based on a preset path score threshold and the initial path score, all the initial occupational graph paths are filtered to obtain a set of candidate occupational graph paths. Optimal path search is performed on the candidate career graph path set to obtain the adaptive career graph path of the target employee.

[0083] In this embodiment of the invention, for the full connectivity graph of career development, by deeply analyzing its graphical structure and node connection relationships, all paths that can reach various possible career states from the initial career state are identified along different node transition directions. These paths are the initial career graph paths. At the same time, by interacting with target employees, such as through questionnaires and face-to-face interviews, data on the target employees' style preferences in career development are collected, such as whether they prefer stable development or rapid promotion, and whether they prefer technical or managerial positions.

[0084] Specifically, a detailed analysis of style preference data is conducted to extract key features, such as stability preferences and technology preferences. Then, for each initial career path, the career status and transition characteristics are analyzed to extract features related to style preferences. Next, the path features are compared with the style preference features one by one, and a corresponding score is assigned according to the degree of matching. The higher the degree of matching, the higher the score, thus obtaining the initial path score for each initial career path.

[0085] Furthermore, a reasonable path score threshold is pre-set. This threshold is determined based on actual needs and experience and is used to filter out paths that are more in line with the target employee's style preferences. The initial path score of each initial career graph path is compared with the preset path score threshold. If the path score is greater than or equal to the threshold, the path is included in the candidate career graph path set, thereby filtering out a batch of paths that are relatively in line with the target employee's style.

[0086] In this process, each path in the candidate career path set is comprehensively evaluated. By comparing the performance of different paths on these factors, a heuristic search algorithm is used to search for the path with the most obvious comprehensive advantages among the candidate paths. This path is the adaptive career path for the target employee, which can meet the career development needs of the target employee to the greatest extent.

[0087] For example, in the insurance industry, consider an insurance underwriter (the target employee). First, obtain all initial career path paths in their full career development graph, such as promotion from junior to intermediate or senior underwriter, or a transfer to claims processing or actuarial roles. Simultaneously, collect their style preference data, revealing a preference for risk control and a strong interest in data analysis. Next, calculate the matching degree between each path and the preference data to obtain initial path scores; higher preference matching scores are higher. Set a path score threshold and filter out paths exceeding the threshold, forming a candidate set. Finally, perform an optimal path search on the candidate set, comprehensively considering factors such as promotion speed, skill enhancement, and salary increase to determine the underwriter's adaptive career path. For example, first, promote to senior underwriter to accumulate experience, then leverage data analysis skills to transfer to an actuarial role for better career development.

[0088] In this embodiment of the invention, adaptive path generation is carried out based on a fully connected career development graph. The computer algorithm automatically traverses the nodes and relationships in the graph, and the path search and filtering can be completed without manual intervention, which improves the automation of career path analysis, avoids the limitations of fixed algorithms, enhances the accuracy of path matching, and thus improves the accuracy of career path analysis for target employees.

[0089] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0090] like Figure 4 The diagram shown is a functional block diagram of an employee adaptive career path analysis device provided in an embodiment of the present invention.

[0091] This disclosure provides an employee adaptive career path analysis device, which corresponds one-to-one with the employee adaptive career path analysis method described in the above embodiments. For example... Figure 4 As shown, the employee adaptive career path analysis device 100 can be installed in an electronic device. According to its functions, the employee adaptive career path analysis device 100 includes a data annotation and fusion module 101, a job map construction module 102, a job feature analysis module 103, a competency profile generation module 104, a career decision analysis module 105, and a map path generation module 106. Detailed descriptions of each functional module are as follows: The data annotation and fusion module 101 is used to acquire multimodal work data of the target employee, and to annotate and fuse the multimodal work data to obtain the target work data. The job knowledge graph construction module 102 is used to construct the job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph. The job feature analysis module 103 is used to perform multi-dimensional job feature analysis on the target employee based on the multi-dimensional job features, and obtain the corresponding multi-dimensional work ability quantitative feature value. The capability profile generation module 104 is used to generate structured employee capability profile data of the target employee based on the multidimensional work capability quantitative feature values ​​and the target work data. The career decision analysis module 105 is used to perform Markov career decision analysis on the structured employee competency profile data to obtain the full connectivity graph of the target employee's career development. The graph path generation module 106 is used to generate adaptive paths based on the career development fully connected graph to obtain the adaptive career graph path of the target employee.

[0092] In one embodiment, when the data annotation and fusion module 101 performs annotation and modal fusion on the multimodal working data to obtain the target working data, it is used to: The multimodal working data is cleaned to generate cleaned multimodal working data; Extract the text data and image data from the cleaning multimodal working data; The text working data is segmented and part-of-speech tagged to obtain structured text annotation data; Semantic feature point annotation is performed on the image working data to obtain structured image annotation data; Based on preset modal alignment rules, the structured text annotation data and the structured image annotation data are aligned in both time and spatial dimensions to form aligned multimodal annotation data; Attention-weighted fusion is performed on the aligned multimodal labeled data to obtain the target working data.

[0093] In one embodiment, when the work knowledge graph construction module 102 performs the task of constructing the work knowledge graph of the target employee based on the target work data, it is configured to: The target working data is parsed to obtain information on multiple working entities; The information of multiple work entities is normalized to generate a standardized set of work entities; Identify the relationships between standardized work entities in the set of standardized work entities; Configure weight parameters for each of the aforementioned relationships to obtain weighted relationships; Construct the initial graph topology of the target employee based on the standardized set of working entities and the weighted association relationships; The initial graph topology is subjected to conflict detection and graph enhancement to generate the target employee's work knowledge graph.

[0094] In one embodiment, when the job feature analysis module 103 performs multidimensional job feature analysis on the target employee based on the multidimensional job features to obtain the corresponding multidimensional job capability quantification feature value, it is used to: Analyze the multi-dimensional work characteristics and determine the capability feature dimensions corresponding to the multi-dimensional work characteristics based on the analysis results; Based on the aforementioned capability characteristic dimensions, a corresponding quantitative analysis indicator system is constructed, and quantifiable characteristic indicators, unstructured characteristic correlation indicators, and potential characteristic indicators in the quantitative analysis indicator system are identified. The first quantitative feature value is obtained by performing time-series trend analysis on the multi-dimensional work characteristics using the quantifiable feature indicators. The unstructured feature correlation index is used to perform work behavior feature analysis on the multi-dimensional work features to obtain the second quantitative feature value; The potential capability characteristics of the multi-dimensional work characteristics are analyzed using the potential characteristic indicators to obtain the third quantitative characteristic value; The first quantitative feature value, the second quantitative feature value, and the third quantitative feature value are weighted and summed according to the weight coefficients of the quantitative analysis indicator system to obtain the multidimensional work capability quantitative feature value.

[0095] In one embodiment, the competency profile generation module 104 generates structured employee competency profile data for the target employee based on the multidimensional work competency quantification feature values ​​and the target work data, including: Construct a mapping relationship between the target work data and the corresponding multidimensional work capability quantitative feature values; The target work data and the corresponding multidimensional work ability quantitative feature values ​​are filled into the preset employee ability profile data framework according to the association mapping relationship to obtain the initial employee ability profile data; The initial employee competency profile data is matched and verified, and the initial employee competency profile data is optimized and adjusted according to the matching and verification results to obtain structured employee competency profile data.

[0096] In one embodiment, when the career decision analysis module 105 performs Markov career decision analysis on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development, it is used to: Identify the career decision state nodes and state transition probability matrix in the preset Markov career decision model, and identify the initial career state nodes of the target employee corresponding to the structured employee competency profile data. Extract the capability quantification requirement threshold corresponding to the career decision state node; The structured employee competency profile data is matched and analyzed with the corresponding competency quantification requirement threshold, and the set of achievable career status nodes for the target employee is determined based on the matching analysis results. The probability of the target employee transitioning from the initial career state node to each reachable career state node in the set of reachable career state nodes is calculated based on the state transition probability matrix. A fully connected graph of the target employee's career development is generated based on the initial career state nodes, the set of reachable career state nodes, and the transition probabilities.

[0097] In one embodiment, when the graph path generation module 106 performs adaptive path generation based on the career development fully connected graph to obtain the adaptive career graph path of the target employee, it is used to: Obtain all initial career graph paths in the fully connected career development graph and the style preference data of the target employee; The matching degree of all the initial occupational graph paths is calculated with the style preference data to obtain the initial path score of each initial occupational graph path; Based on a preset path score threshold and the initial path score, all the initial occupational graph paths are filtered to obtain a set of candidate occupational graph paths. Optimal path search is performed on the candidate career graph path set to obtain the adaptive career graph path of the target employee.

[0098] In this invention, the specific limitations of an employee adaptive career path analysis device can be found in the above-described limitations of an employee adaptive career path analysis method, and will not be repeated here. Each module in the aforementioned employee adaptive career path analysis device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0099] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a server-side employee adaptive career path analysis method.

[0100] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. 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 network interface is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements client-side functions or steps of an employee adaptive career path analysis method.

[0101] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: Acquire multimodal work data of the target employee, and perform annotation and modality fusion on the multimodal work data to obtain the target work data; Construct a job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; Based on the multi-dimensional work characteristics, the target employee is analyzed for multi-dimensional work characteristics to obtain the corresponding multi-dimensional work ability quantitative characteristic value. Based on the multidimensional work ability quantitative feature values ​​and the target work data, a structured employee ability profile of the target employee is generated; Markov career decision analysis was performed on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development. Adaptive path generation is performed based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

[0102] In the several embodiments provided by this invention, it should be understood that the disclosed devices and apparatuses can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0103] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0104] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0105] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0106] The readable storage medium of the present invention stores a computer program, which, when executed by a processor of an electronic device, can perform the following: Acquire multimodal work data of the target employee, and perform annotation and modality fusion on the multimodal work data to obtain the target work data; Construct a job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; Based on the multi-dimensional work characteristics, the target employee is analyzed for multi-dimensional work characteristics to obtain the corresponding multi-dimensional work ability quantitative characteristic value. Based on the multidimensional work ability quantitative feature values ​​and the target work data, a structured employee ability profile of the target employee is generated; Markov career decision analysis was performed on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development. Adaptive path generation is performed based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

[0107] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0108] Computer-readable storage media may also store at least one computer-executable program / instruction, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0109] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0110] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0111] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Furthermore, any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0112] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0113] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0114] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

[0115] It should be noted that if any software tools or components not belonging to our company appear in the embodiments of this application, they are merely for illustrative purposes and do not represent actual use.

Claims

1. A method for analyzing employee adaptive career paths, characterized in that, The method includes: Acquire multimodal work data of the target employee, and perform annotation and modality fusion on the multimodal work data to obtain the target work data; Construct a job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; Based on the multi-dimensional work characteristics, the target employee is analyzed for multi-dimensional work characteristics to obtain the corresponding multi-dimensional work ability quantitative characteristic value. Based on the multidimensional work ability quantitative feature values ​​and the target work data, a structured employee ability profile of the target employee is generated; Markov career decision analysis was performed on the structured employee competency profile data to obtain the fully connected graph of the target employee's career development. Adaptive path generation is performed based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

2. The employee adaptive career path analysis method as described in claim 1, characterized in that, The step of labeling and modal fusion of the multimodal working data to obtain the target working data includes: The multimodal working data is cleaned to generate cleaned multimodal working data; Extract the text data and image data from the cleaning multimodal working data; The text working data is segmented and part-of-speech tagged to obtain structured text annotation data; Semantic feature point annotation is performed on the image working data to obtain structured image annotation data; Based on preset modal alignment rules, the structured text annotation data and the structured image annotation data are aligned in both time and spatial dimensions to form aligned multimodal annotation data; Attention-weighted fusion is performed on the aligned multimodal labeled data to obtain the target working data.

3. The employee adaptive career path analysis method as described in claim 1, characterized in that, The step of constructing the job knowledge graph of the target employee based on the target job data includes: The target working data is parsed to obtain information on multiple working entities; The information of multiple work entities is normalized to generate a standardized set of work entities; Identify the relationships between standardized work entities in the set of standardized work entities; Configure weight parameters for each of the aforementioned relationships to obtain weighted relationships; Construct the initial graph topology of the target employee based on the standardized set of working entities and the weighted association relationships; The initial graph topology is subjected to conflict detection and graph enhancement to generate the target employee's work knowledge graph.

4. The employee adaptive career path analysis method as described in claim 1, characterized in that, The step of performing multidimensional work characteristic analysis on the target employee based on the multidimensional work characteristics to obtain corresponding multidimensional work ability quantitative feature values ​​includes: Analyze the multi-dimensional work characteristics and determine the capability feature dimensions corresponding to the multi-dimensional work characteristics based on the analysis results; Based on the aforementioned capability characteristic dimensions, a corresponding quantitative analysis indicator system is constructed, and quantifiable characteristic indicators, unstructured characteristic correlation indicators, and potential characteristic indicators in the quantitative analysis indicator system are identified. The first quantitative feature value is obtained by performing time-series trend analysis on the multi-dimensional work characteristics using the quantifiable feature indicators. The unstructured feature correlation index is used to perform work behavior feature analysis on the multi-dimensional work features to obtain the second quantitative feature value; The potential capability characteristics of the multi-dimensional work characteristics are analyzed using the potential characteristic indicators to obtain the third quantitative characteristic value; The first quantitative feature value, the second quantitative feature value, and the third quantitative feature value are weighted and summed according to the weight coefficients of the quantitative analysis indicator system to obtain the multidimensional work capability quantitative feature value.

5. The employee adaptive career path analysis method as described in claim 1, characterized in that, The step of generating structured employee competency profile data for the target employee based on the multidimensional work competency quantification feature values ​​and the target work data includes: Construct a mapping relationship between the target work data and the corresponding multidimensional work capability quantitative feature values; The target work data and the corresponding multidimensional work ability quantitative feature values ​​are filled into the preset employee ability profile data framework according to the association mapping relationship to obtain the initial employee ability profile data; The initial employee competency profile data is matched and verified, and the initial employee competency profile data is optimized and adjusted according to the matching and verification results to obtain structured employee competency profile data.

6. The employee adaptive career path analysis method as described in claim 1, characterized in that, The Markov career decision analysis performed on the structured employee competency profile data yields the fully connected career development graph of the target employee, including: Identify the career decision state nodes and state transition probability matrix in the preset Markov career decision model, and identify the initial career state nodes of the target employee corresponding to the structured employee competency profile data. Extract the capability quantification requirement threshold corresponding to the career decision state node; The structured employee competency profile data is matched and analyzed with the corresponding competency quantification requirement threshold, and the set of achievable career status nodes for the target employee is determined based on the matching analysis results. The probability of the target employee transitioning from the initial career state node to each reachable career state node in the set of reachable career state nodes is calculated based on the state transition probability matrix. A fully connected graph of the target employee's career development is generated based on the initial career state nodes, the set of reachable career state nodes, and the transition probabilities.

7. The employee adaptive career path analysis method as described in claim 1, characterized in that, The adaptive path generation based on the fully connected career development graph to obtain the adaptive career graph path for the target employee includes: Obtain all initial career graph paths in the fully connected career development graph and the style preference data of the target employee; The matching degree of all the initial occupational graph paths is calculated with the style preference data to obtain the initial path score of each initial occupational graph path; Based on a preset path score threshold and the initial path score, all the initial occupational graph paths are filtered to obtain a set of candidate occupational graph paths. Optimal path search is performed on the candidate career graph path set to obtain the adaptive career graph path of the target employee.

8. An employee adaptive career path analysis device, characterized in that, The device includes: The data annotation and fusion module is used to acquire multimodal work data of the target employee, and to annotate and fuse the multimodal work data to obtain the target work data. The job knowledge graph construction module is used to construct the job knowledge graph of the target employee based on the target job data, and extract multi-dimensional job features from the job knowledge graph; The job feature analysis module is used to perform multi-dimensional job feature analysis on the target employee based on the multi-dimensional job features, and obtain the corresponding multi-dimensional work ability quantitative feature value. The competency profile generation module is used to generate structured employee competency profile data of the target employee based on the multidimensional work competency quantitative feature values ​​and the target work data; The career decision analysis module is used to perform Markov career decision analysis on the structured employee competency profile data to obtain the full connectivity graph of the target employee's career development. The graph path generation module is used to generate adaptive paths based on the fully connected career development graph to obtain the adaptive career graph path of the target employee.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the employee adaptive career path analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the employee adaptive career path analysis method as described in any one of claims 1 to 7.

Images