Visual data analysis and application method based on artificial intelligence

Abstract

The invention provides a visual data analysis application method based on artificial intelligence, and relates to the technical field of visual data mining. The constructed occupational replacement recommendation model can give an occupational replacement matching degree between the first occupational tag and each alternative occupational tag, and the occupational replacement recommendation model can give matching degree evaluation of the alternative occupational tags under adjacent position levels i-1 and i + 1; generating a transfer matching degree between the first occupational tag and each alternative occupational tag; the matching and screening strategy can be selected according to the emphasis of the target user, the dynamic matching and screening strategy is implemented, and the method has good flexibility in occupational selection.

Description

Technical Field

[0001] The present invention relates to the technical field of visual data mining, and in particular to an application method of visual data analysis based on artificial intelligence. Background Art

[0002] Long Short-Term Memory (LSTM) networks, a deep learning framework, have shown great potential for dynamic data analysis due to their ability to process and predict time series data. These methods can update and analyze large-scale career datasets in real time, helping individuals and businesses make more data-driven decisions about career development. However, in existing environments, presenting these complex analytical processes and results to users in an intuitive and understandable manner remains a technical challenge.

[0003] In the prior art, the publication number is CN112883100A, and the name is data analysis visualization method, system and device based on big data artificial intelligence, which includes establishing an Internet human resources big data talent demand index standard and building a metadata database based on it; collecting data from recruitment advertisements publicly available on the Internet according to the talent demand index standard established in step S1, and storing the data using a big data platform; cleaning and desensitizing the captured web page data to form structured data with a standard format; building a machine learning algorithm to analyze and mine the data; and displaying the analysis results in an interactive real-time data visualization web page.

[0004] The shortcomings are:

[0005] When current target users make their first career choice, they usually lack a clear understanding of the career development prospects; users' choices often rely on subjective judgment, and they lack in-depth understanding of the future development trends of the selected career; at the same time, they are not familiar with the relevant career options available when they change careers in the future, resulting in a high trial-and-error cost for personal career choices; existing matching and recommendation algorithms are usually based on static features, ignoring the dynamic nature of job changes and the path characteristics of individual career growth.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide an artificial intelligence-based visual data analysis application method to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The application method of visual data analysis based on artificial intelligence includes the following steps:

[0010] Step S1: obtaining career data corresponding to a first occupation label, and preprocessing the career data to obtain a first data set;

[0011] Career data includes but is not limited to average years of work experience at different job levels, salary level indicators, job popularity trend indicators, and required job skills demand indicators;

[0012] Step S2: determining a set of candidate occupation labels that are highly correlated with the first occupation label;

[0013] Obtaining career data corresponding to each candidate career label in the candidate career label set, and forming a second data set through data preprocessing;

[0014] Step S3: constructing a career change recommendation model, which is used to evaluate the career change matching degree between the first career label and each alternative career label;

[0015] Step S4: The first data set and the second data set are used as career development feature sets and input into a long short-term memory network for training and verification to construct a dynamic prediction model. The dynamic prediction model generates a job transfer matching degree between the first career label and each alternative career label;

[0016] Step S5: Perform a weighted analysis on the career change matching degree and job transfer matching degree between the first career label and each alternative career label, use data visualization tools to display the weighted analysis results, apply corresponding matching screening strategies to these weighted analysis results, and screen out at least three alternative career labels that meet the career development requirements from the alternative career label set.

[0017] Compared with the prior art, the beneficial effects of the present invention are: by collecting career data of the first career label and the set of alternative career labels, the constructed career change recommendation model can give the career change matching degree between the first career label and each alternative career label, and the career change recommendation model can give the matching degree evaluation of the alternative career label at adjacent job levels i-1 and i+1, and by constructing a dynamic prediction model, generate the job transfer matching degree between the first career label and each alternative career label; and the matching screening strategy can implement a dynamic matching screening strategy according to the focus of the target user, and has good flexibility in career selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of the overall method of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0020] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0021] Example 1:

[0022] See also Figure 1 , the present invention provides a technical solution:

[0023] An artificial intelligence-based visual data analysis application method is used to provide alternative career options for the first career label selected by the target user. The specific steps include:

[0024] Step S1: obtaining career data corresponding to a first occupation label, and preprocessing the career data to obtain a first data set;

[0025] Career data includes but is not limited to average years of work experience at different job levels, salary level indicators, job popularity trend indicators, and required job skills demand indicators;

[0026] Further explanation: Acquisition of career data includes: obtaining the first career tag selected by the target user from the career social platform, capturing public user information of the first career tag in the currently selected administrative area, and analyzing and processing the public user information to generate career data;

[0027] The first occupation tag of the target user is determined by the target user's input in the career social networking platform or the results of a career interest assessment. In this embodiment, the occupation tags are standardized using the International Standard Classification of Occupations (ISCO-08) system to ensure consistency and comparability of occupation names.

[0028] Capture public user data of the first occupation tag in the currently selected administrative area. The specific logic is as follows:

[0029] Use the crawler framework (Scrapy) to write customized crawlers to crawl data from selected data sources;

[0030] The crawled data fields include but are not limited to job title, job level, years of experience, salary level, number of job postings, job posting frequency, and required skills;

[0031] During the data capture process, set the geographic area filtering parameters to ensure that the collected data only comes from the administrative area selected by the user, which includes the city or province to which it belongs.

[0032] Furthermore, data preprocessing includes data cleaning of the selected dataset, removing invalid and duplicate data, desensitizing public user information, ensuring data privacy, and storing the cleaned data in a standardized structured database.

[0033] Data standardization operations include the following:

[0034] Standardize job titles: Unify similar job titles on different professional social networking platforms, such as "software engineer" and "software development engineer" into "software engineer".

[0035] Salary unification: Convert monthly salaries, annual salaries, etc. of different salary units into annual salaries and convert them into a unified currency unit.

[0036] Specifically, we write scripts for string matching and unit conversion, and use regular expressions to process text data.

[0037] Furthermore, the skill requirements in the job description are segmented, stop words are removed, and key skill tags are identified and extracted;

[0038] Use natural language processing (NLP) techniques, such as TF-IDF or Word2Vec, to vectorize skills. Specifically, use NLTK or spaCy for text processing and Scikit-learn for vectorization.

[0039] The pre-processed career data corresponding to the first occupation label, including the average years of work, salary level, job popularity trend, and job skill requirement indicators, are integrated into a first dataset; and the first dataset is saved in CSV or Parquet storage format.

[0040] Based on the first data set, the average working years indicator of the first occupation label k under job level i is defined as ; The larger the output value of position level i, the higher the position level;

[0041] The calculation formula is as follows:

[0042] ;

[0043] in, is the working years of the jth sample of the first occupation label k under job level i; M is the total number of samples; is the maximum working years of the first occupation label k under job level i; The output is limited to the interval (0,1);

[0044] when The larger the output value, the longer the average working years of the first occupation label k under job level i; the more stable the job level i is.

[0045] Based on the first data set, the salary level indicator of the first occupation label k under job level i is defined as ; The calculation formula is as follows:

[0046] ;

[0047] is the salary of the jth sample of the first occupation label k under job level i;

[0048] is the maximum salary of the first occupation label k at job level i;

[0049] The output is limited to the interval (0,1);

[0050] when The larger the output value, the higher the salary of the first occupation label k under job level i; the greater the demand for job level i.

[0051] Based on the first dataset, the job popularity trend index of the first occupation label k under job level i is defined as ; The calculation formula is as follows:

[0052] ;

[0053] in, is the job posting frequency of the jth sample under job level i for the first occupation label k;

[0054] is the historical maximum job posting frequency for the first occupation label k at job level i. The historical maximum job posting frequency is determined by the job posting frequency in the past six months.

[0055] The output is limited to the interval (0,1);

[0056] when The larger the output value, the higher the popularity of the job level i of the first occupation label k;

[0057] Based on the first data set, the job skill requirement index of the first occupation label k under job level i is defined as ; The calculation formula is as follows:

[0058] ;

[0059] in, is the number of samples with the first occupation label k in the M samples under job level i that need to meet the required skills list; M is the total number of samples; The output is limited to the interval (0,1);

[0060] Job Skills Demand Index The corresponding required skills list is represented as ; where h represents the index of required skills, and H2 is the total number of required skill categories; this is achieved by mapping occupation labels to skill indices;

[0061] when The larger the output value, the higher the job skill requirement for the first occupation label k at job level i.

[0062] Step S2: determining a set of candidate occupation labels that are highly correlated with the first occupation label;

[0063] Obtaining career data corresponding to each candidate career label in the candidate career label set, and forming a second data set through data preprocessing;

[0064] Further explanation: Determining the set of alternative occupational labels includes: obtaining the first occupational label and performing difference analysis on the required skill lists of various types of occupational labels in the occupation database to find the set of alternative occupational labels that are highly correlated with the first occupational label;

[0065] "Highly relevant" means that the required skill list of the alternative occupation tag has a high degree of overlap with the required skill list of the first occupation tag, and the high overlap is set to ≥70%. The specific logic is:

[0066] Job skill requirement index based on the first occupation label k at job level i ; Determine the required skills list for the position level i corresponding to the first occupation label k;

[0067] In the occupation database, select occupation tags of various types at different job levels that contain at least 70% of the skills in the skill list required by the first occupation tag k at job level i, and use these occupation tags as a set of highly relevant candidate occupation tags;

[0068] The career data corresponding to each alternative career label in the alternative career label set are pre-processed and the average working years index, salary level index, job popularity trend index and job skill requirement index are integrated into the second data set;

[0069] Based on the second data set, the job skill requirement index of each alternative occupation label in the job level i is marked as ; is the index of the occupation label in the set of alternative occupation labels, and , K is the total number of categories of alternative occupation labels in the alternative occupation label set; i represents the job level index;

[0070] Based on the second data set, the alternative occupation labels The average working years indicator at job level i is defined as ;

[0071] Based on the second data set, the alternative occupation labels The salary level indicator at job level i is defined as ;

[0072] Based on the second data set, the alternative occupation labels The job popularity trend indicator at job level i is defined as .

[0073] Step S3: constructing a career change recommendation model, which is used to evaluate the career change matching degree between the first career label and each alternative career label;

[0074] Further explanation: Constructing a career change recommendation model includes:

[0075] Based on the first data set and the second data set, under the same job level combination, data analysis is performed on the job skill requirement indicators of the first occupation label and each alternative occupation label to obtain a first replacement matching coefficient;

[0076] Based on the first data set and the second data set, under different spans of job level combinations, data analysis is performed on the job skill requirement indicators of the first occupation label and each alternative occupation label to obtain a second replacement matching coefficient;

[0077] The combination of different spans of job levels means: combining the job level i of the first job label k with the job level i of the alternative job label Position level i-1 or i+1 combination; i-1 and i+1 represent the adjacent position levels lower than position level i and higher than position level i respectively;

[0078] Comprehensively analyzing the first replacement matching coefficient and the second replacement matching coefficient to obtain a comprehensive matching index, which is used to comprehensively evaluate the career replacement matching degree between the first career label and each alternative career label;

[0079] Furthermore, the first replacement matching coefficient is represented by the following calculation formula:

[0080] ;

[0081] in, is the first occupation label k and the alternative occupation label The first replacement matching coefficient is obtained by analyzing the job skill requirement indicators under job level i;

[0082] Is the first occupation label k and the alternative occupation label The number of skill intersections at job level i;

[0083] Increase : Indicates alternative career labels Improved skill matching within the same job level;

[0084] is the total number of skills in the skill list required for the first occupation label k at job level i;

[0085] reduce : Indicates that the skills required for the main profession are reduced, and the matching coefficient increases when other conditions remain unchanged;

[0086] The required skills list based on the alternative occupation label has a high degree of overlap with the required skills list of the first occupation label. The value range is ; is an interval representation;

[0087] when The closer it is to 1, the more alternative occupation labels there are at the same job level i. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority;

[0088] The second replacement matching coefficient is represented by the following calculation formula:

[0089] ;

[0090] in, It is the first occupation label k and the alternative occupation label under different span job level combinations i-1 and i+1. The second replacement matching coefficient is obtained by analyzing the job skill demand indicators;

[0091] The first occupation label k under job level i and the alternative occupation label The number of skill intersections at job level i-1;

[0092] The first occupation label k under job level i and the alternative occupation label The number of skill intersections at job level i+1;

[0093] Increase or : Indicates that the skill matching degree of alternative occupations at adjacent job levels has improved.

[0094] The required skills list based on the alternative occupation label has a high degree of overlap with the required skills list of the first occupation label. The value range is ; is an interval representation;

[0095] The second step of obtaining the replacement matching coefficient is as follows

[0096] Obtain the job skill requirement index of the first occupation label k under job level i from the first dataset .

[0097] Obtaining candidate occupation labels from the second dataset Job skill requirement indicators at job levels i-1 and i+1 and .

[0098] calculate and intersection of .

[0099] calculate and intersection of .

[0100] Add the number of intersections of the two sets and divide by , and obtain the matching coefficient .

[0101] when The closer it is to 1, the more alternative occupation labels there are at adjacent job levels i-1 and i+1. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority;

[0102] The comprehensive matching index is a weighted integration of the first replacement matching coefficient and the second replacement matching coefficient. The comprehensive matching index is represented by the following calculation formula:

[0103] ;

[0104] in, It is a comprehensive matching index, which represents the matching degree between the first occupation label and each alternative occupation label;

[0105] and is the weight coefficient of the corresponding replacement matching coefficient, which is used to allocate the weights of the first replacement matching and the second replacement matching in the comprehensive matching index; , and The values are all within the interval (0,1); ;

[0106] set up The value range is ; is an interval representation;

[0107] when The closer it is to 1, the more alternative occupation labels there are at the same job level i or adjacent job levels i-1 and i+1. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority.

[0108] This embodiment provides the following example calculation:

[0109] The required skills list corresponding to the job skill requirement index of the first occupation label k at job level i=3 is: ;

[0110] Alternative career labels The required skills list corresponding to the job skill requirement indicators at job level i=3 is as follows: ;

[0111] Alternative career labels At job level The required skills list corresponding to the job skill demand indicators below is ;

[0112] Alternative career labels At job level The required skills list corresponding to the job skill demand indicators below is ;

[0113] First replacement matching coefficient:

[0114] ;

[0115] Second replacement matching coefficient:

[0116]

[0117] Comprehensive matching index: set ;

[0118]

[0119] : Alternative career tags There is a 60% skill match at the same job level, indicating a high transfer potential;

[0120] : Alternative career tags There is a 40% skill match at adjacent job levels, indicating a certain degree of flexibility in job transfers.

[0121] : The comprehensive matching index is 0.5, indicating an alternative career label They have a moderate match in multiple job levels and are suitable for consideration for job transfer.

[0122] Step S4: The first data set and the second data set are used as career development feature sets and input into a long short-term memory network for training and verification to construct a dynamic prediction model. The dynamic prediction model generates a job transfer matching degree between the first career label and each alternative career label;

[0123] The first dataset is characterized as ;

[0124] make sure The number of samples for each indicator in is consistent, providing consistency for subsequent model training;

[0125] The second dataset is characterized as ;

[0126] Will and as a set of professional development characteristics;

[0127] The Long Short-Term Memory (LSTM) model was chosen for the following reasons:

[0128] LSTM can capture long-term dependencies in time series data and is suitable for the temporal changes in career development characteristics. Long short-term memory networks (LSTMs) are part of machine learning in the field of artificial intelligence.

[0129] For career development forecasts, it can handle changing trends and volatility to improve forecast accuracy.

[0130] Build a dynamic prediction model based on long short-term memory network. The dynamic prediction model structure includes an input layer, multiple LSTM layers and an output layer.

[0131] The input layer receives a career development feature set consisting of a first data set and a second data set;

[0132] After the career development feature set is processed by the LSTM layer, the output layer finally generates the job transfer matching degree between the first career label and each alternative career label;

[0133] This embodiment provides the following specific instructions:

[0134] Input layer: The input dimension is and The combined feature set of ; the total number of features in this embodiment = 8;

[0135] LSTM layer: Set up 2 layers of LSTM, with 64 units in each layer;

[0136] Fully connected layer: connected to an output layer, outputting the job transfer matching degree between the first occupation label and each alternative occupation label;

[0137] Use the training set (80% of the collected career development feature set is used for training) and the validation set (20% of the career development feature set is used for validation) to train the LSTM model. The specific steps are:

[0138] Select the optimizer Adam and the mean square error MSE represented by the loss function;

[0139] Set the number of training rounds, initially set to 100 rounds, and dynamically adjust according to the verification loss when verifying the results;

[0140] Use the validation set to evaluate the model performance and calculate the mean square error (MS) to ensure the model's prediction effect;

[0141] After the model training is completed, the data set is input and Input into the trained LSTM model to generate the job transfer matching degree between the first occupation label and each alternative occupation label; specifically:

[0142] After multi-layer LSTM processing, the final layer converts the learned feature information into outputs corresponding to candidate career labels. The activation function Softmax converts the outputs into values that conform to a probability distribution. These values represent the job transfer compatibility assessment of each candidate career label with the first career label.

[0143] The value in the dynamic prediction model output directly represents the matching degree of the alternative occupation label to the first occupation label;

[0144] In this embodiment, when the dynamic prediction model outputs a matching degree of 0.85 for an alternative occupation label and 0.50 for another, the first alternative occupation label is more suitable for switching from the first occupation;

[0145] It should be noted that the job transfer matching degree between the first occupation label and each alternative occupation label is generated through the output layer, specifically including:

[0146] The output layer receives the data processed by the LSTM layer and converts it into a probability distribution form through the activation function Softmax;

[0147] The memory units and gating mechanism within the LSTM layer allow the LSTM to learn that career development features are concentrated in historical and future moments that are relevant or influential;

[0148] The memory cells within the LSTM model can learn complex relationships between features. By weighting the input data, the model can not only evaluate the impact of each feature itself, but also understand the interactions between features.

[0149] Each layer of LSTM receives input data and performs weighted calculations on the input using the weights learned during training.

[0150] LSTM adjusts its memory state and hidden state at each time step to retain information useful for future predictions.

[0151] The last layer of LSTM passes the processed data to the output layer and outputs the final prediction results, such as the matching score.

[0152] The output layer uses the activation function Softmax to convert the results into a probability distribution, reflecting the job transfer matching degree of different occupation labels;

[0153] The output is each candidate occupation label Relative to the job transfer matching degree of the first occupation label k, the first occupation label k is matched with the alternative occupation label The job transfer matching degree at job level i is recorded as Job transfer matching The value range is between 0 and 1, indicating job transfer matching degree. The larger the value, the higher the priority of the job transfer.

[0154] Step S5: Perform a weighted analysis on the career change matching degree and job transfer matching degree between the first career label and each alternative career label, use data visualization tools to display the weighted analysis results, apply corresponding matching screening strategies to these weighted analysis results, and screen out at least three alternative career labels that meet the career development requirements from the alternative career label set.

[0155] Further explanation: A weighted analysis is performed on the career change matching and job transfer matching between the first career label and each alternative career label. The specific logic is as follows:

[0156]

[0157] in, Is the first occupation label k and the alternative occupation label Combined matching value at job level i;

[0158] and is the weight coefficient of the corresponding parameter, , and The values are all in the interval (0,1);

[0159] set up The valid range of is the interval (0.35,1];

[0160] The matching and screening strategies include:

[0161] like , then the alternative occupation label Occupations marked as high priority for transfer; , , and They are the combined matching value, comprehensive matching index, second replacement matching coefficient and job transfer matching degree judgment threshold, , , ; ;

[0162] When , it means that at adjacent job levels i-1 and i+1, the alternative occupation label There are more options for transfer positions and grades;

[0163] Traverse all the alternative career tags and prioritize at least three that meet the requirements. If the number of alternative occupation labels for the conditions is less than three, the Filter out from the alternative career tags Supplement the alternative occupation label with the largest value;

[0164] Further explanation: and The weight analysis results include the following:

[0165] 1) When the target user only has high requirements for the number of transfer job level selections corresponding to the job skill requirement indicators of the alternative career label and the degree of career change matching, set ; ;

[0166] The purpose of setting is: Due to the high requirements for job skill indicators, it is necessary to focus on The number of transfer job level options represented by the function is then considered, and then the output value of the comprehensive matching index is considered. The technical effect of this setting can highlight the number of transfer job level options of the selected alternative career labels, and in the later stage of job transfer, the flexibility is higher. The purpose of setting it as the largest is to improve the competitiveness of alternative career labels in job transfers;

[0167] 2) When the target users have high requirements for the average years of work experience, salary level, and job popularity trend of the alternative occupation labels, set ; ;

[0168] The purpose of setting is: Due to the high requirements for the average working years index, salary level index, and job popularity trend index, it is necessary to focus on Representation of job transfer matching, The forecast takes into account the average working years index, salary level index, and job popularity trend index, so As a high requirement; the technical effect of this setting can highlight that the alternative career labels can focus on the selection requirements including the average years of work index, salary level index, and position popularity trend index. In the later career change, the selection of alternative career labels will be more comprehensive.

[0169] 3) When the target users have high requirements for the job skill requirement indicators, average working years indicators, salary level indicators, and job popularity trend indicators of the alternative occupation labels, set ; ;

[0170] , , and It was determined by the fuzzy analytic hierarchy process (FAHP), which will not be described in detail; and Determined by entropy weight method.

[0171] The purpose of setting it up is to: balance the job skill requirement index, average working years index, salary level index, and job popularity trend index;

[0172] and Dynamic settings are used to optimize matching results based on feedback from different job markets and user needs; this mechanism is different from static settings and can respond to changes more flexibly.

[0173] Use the matplotlib or seaborn library in Python to visualize the data of the weight analysis results. The basic framework of data visualization is:

[0174] Create a bar chart or pie chart to show the joint matching values between different candidate occupation labels and the first occupation label k and and The weight analysis results are presented; and the final selected candidate occupation labels are displayed in a data visualization manner;

[0175] Data import: Use the pandas library to import data into the working environment and prepare for visualization;

[0176] Drawing function: If using the matplotlib library: use plt.bar() or plt.pie() to create the corresponding chart;

[0177] If using the seaborn library: use seaborn.barplot() to create a bar chart;

[0178] Adjust the details: set the x-axis and y-axis scales, labels, and the position and style of the legend.

[0179] Chart display: Use the plt.show() command to display the chart on the screen for immediate viewing.

[0180] Save the chart: Use plt.savefig() to save the chart as an image file in PNG or JPEG format for embedding in reports or further sharing.

[0181] For alternative occupations with a joint matching value higher than 0.7, they are marked with different colors to highlight the occupations with high recommendation priority.

[0182] Example 2:

[0183] Based on Example 1, in order to verify the effectiveness of the artificial intelligence-based visual data analysis application method, "data analyst" is selected as the first occupational label k, and the proposed set of alternative occupational labels includes "data scientist", "business analyst", "machine learning engineer", "big data engineer", "BI analyst" and "statistician". During the implementation process, the career data of each occupational label under the position level i is first collected and sorted, including the average years of work, salary level, position popularity trend and required position skill demand indicators. The specific steps are as follows:

[0184] 1.1) Data acquisition and preprocessing:

[0185] Through industry reports, recruitment websites, and career survey data, we collected data related to "data analyst" and alternative job tags at different job levels (junior, mid-level, and senior). Data preprocessing included data cleaning, missing value handling, and normalization to ensure data consistency and comparability. In this example, the average length of service for a mid-level "data scientist" position is three years, the salary is 150,000 RMB / year, the position is trending upward, and required skills include Python, machine learning, and data visualization.

[0186] 1.2) Construction of career change recommendation model:

[0187] Based on the collected career data, a career change recommendation model is constructed. The model calculates the comprehensive matching index between "data analyst" and each alternative career label. , to determine potential job transfer opportunities. In this embodiment, by calculating the matching degree between "data analyst" and "data scientist" in the mid-level positions, it is found that their career change matching degree is 0.85.

[0188] 1.3) Dynamic prediction model training and verification:

[0189] The first and second datasets are input into the long short-term memory network (LSTM) for training and verification, and a dynamic prediction model is constructed to generate the job transfer matching degree. In this embodiment, after training, the job transfer matching prediction value between "data analyst" and "business analyst" in senior positions is 0.78, indicating that the job transfer opportunity has a higher priority.

[0190] 1.4) Weight analysis and matching screening:

[0191] Perform weighted analysis on career change matching and job transfer matching to calculate the combined matching value , and apply matching screening strategies to screen out alternative career labels that meet career development requirements. Weight coefficient and The entropy weight method is used to ensure that the comprehensive matching value is in the valid range (0.35, 1). In this embodiment, for the high-demand job skill requirement indicators and occupational change matching degree, b1=0.6 and b2=0.4 are set, and the calculated joint matching value of "Machine Learning Engineer" is 0.92, which meets the screening criteria.

[0192] 1.5) Visualization of results:

[0193] The Matplotlib library is used to generate a bar chart to intuitively display the joint matching value between each alternative career label and "data analyst"; different matching degrees are distinguished by color, and career labels with high matching values are highlighted to help users make scientific career transition decisions.

[0194] The above implementation steps not only demonstrate the application process of the AI-based visual data analysis method in job matching, but also demonstrate the advantages of this method in matching accuracy and decision support through specific data. The experimental data table is given below:

[0195] Table 1 Job transfer matching results:

[0196]

[0197] The table description is as follows:

[0198] The above table shows the matching index of "Data Analyst" and six alternative occupation labels at the intermediate to senior job levels. Matching degree with job transfer Reflects the suitability of career change and job transfer, combined with matching value The priority of job transfers was comprehensively evaluated through weighted analysis. Experimental data showed that the AI-based visual data analysis method can effectively identify highly compatible alternative career tags, assisting users in making informed decisions during their career development. In this example, "Machine Learning Engineer" significantly outperformed other alternative careers due to its high comprehensive matching index and job transfer matching, demonstrating the superiority and innovation of this method in job matching.

[0199] With weights set to b1 = 0.6 and b2 = 0.4, this method prioritizes the diversity of the number of transfer positions. Once an alternative occupational label, such as "machine learning engineer," demonstrates high adaptability across multiple job levels, it will stand out more in the screening process.

[0200] Furthermore, the calculation of job transfer compatibility directly considers dynamic market demand and career development trends. Therefore, the system identifies positions with broad skill requirements and upward mobility as having greater transfer flexibility. This not only provides individuals with more career development paths but also enhances their competitiveness in an uncertain job market, ensuring the ability to flexibly transition to positions that meet both personal and market needs. This flexibility is particularly important in rapidly changing industries, as users can find the most suitable development path among a wider range of job levels.

[0201] It should be noted that all calculation formulas in this application document use regression analysis, including but not limited to machine learning algorithms, to deeply analyze the relevant parameters collected and identify their natural trends and interrelationships. Use professional software, such as Python's Scikit-learn library or R language, to automatically generate mathematical models that match the data. Then, objectively evaluate the performance of the model through methods such as cross-validation, and combine continuous feedback and optimization to ensure that the created formula truly reflects the inherent laws of the data, thereby ensuring its validity and accuracy, and ensuring that the calculation process complies with the constraints of natural laws rather than being based on artificially set rules;

[0202] The technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0203] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0204] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0205] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An artificial intelligence-based visual data analysis application method for providing alternative career options for a first career tag selected by a target user, characterized in that: The specific steps include: Step S1: obtaining career data corresponding to a first career tag, and preprocessing the career data to obtain a first data set; Career data includes but is not limited to average years of work experience at different job levels, salary level indicators, job popularity trend indicators, and required job skills demand indicators; Step S2: determining a set of candidate occupation labels that are highly correlated with the first occupation label; Obtaining career data corresponding to each candidate career label in the candidate career label set, and forming a second data set through data preprocessing; Step S3: constructing a career change recommendation model, which is used to evaluate the career change matching degree between the first career label and each alternative career label; Step S4: The first data set and the second data set are used as career development feature sets and input into a long short-term memory network for training and verification to construct a dynamic prediction model. The dynamic prediction model generates a job transfer matching degree between the first career label and each alternative career label; Step S5: Perform a weighted analysis on the career change matching degree and job transfer matching degree between the first career label and each alternative career label, use data visualization tools to display the weighted analysis results, apply corresponding matching screening strategies to these weighted analysis results, and screen out at least three alternative career labels that meet the career development requirements from the alternative career label set.

2. The artificial intelligence-based visual data analysis application method according to claim 1, characterized in that: Acquisition of career data includes: obtaining the first career tag selected by the target user from the career social platform, capturing public user information of the first career tag in the currently selected administrative area, and analyzing and processing the public user information to generate career data; The pre-processed career data corresponding to the first occupation label, including average years of work, salary level, job popularity trend, and job skill requirement, are integrated into the first data set; Based on the first data set, the average working years indicator of the first occupation label k under job level i is defined as ; The larger the output value of position level i, the higher the position level; Based on the first data set, the salary level indicator of the first occupation label k under job level i is defined as ; Based on the first dataset, the job popularity trend index of the first occupation label k under job level i is defined as ; when The larger the output value, the higher the popularity of the job level i of the first occupation label k; Based on the first data set, the job skill requirement index of the first occupation label k under job level i is defined as ; Job Skills Demand Index The corresponding required skills list is represented as ; where h represents the index of the required skills, and H2 is the total number of required skill categories; when The larger the output value, the higher the job skill requirement for the first occupation label k at job level i.

3. The artificial intelligence-based visual data analysis application method according to claim 2, characterized in that: Determining the set of alternative occupational labels includes: obtaining the first occupational label and performing difference analysis on the required skill lists of various types of occupational labels in the occupational database to find the set of alternative occupational labels that are highly correlated with the first occupational label; "Highly relevant" is defined as the required skill list of the alternative occupation label having a high degree of overlap with the required skill list of the first occupation label, and high overlap is set as the number of skills overlap ≥ 70%; The career data corresponding to each alternative career label in the alternative career label set are pre-processed and the average working years index, salary level index, job popularity trend index and job skill requirement index are integrated into the second data set; Based on the second data set, the job skill requirement index of each alternative occupation label in the job level i is marked as ; is the index of the occupation label in the set of alternative occupation labels, and , K is the total number of categories of alternative occupation labels in the alternative occupation label set; i represents the job level index; Based on the second data set, the alternative occupation labels The average working years indicator at job level i is defined as ; Based on the second data set, the alternative occupation labels The salary level indicator at job level i is defined as ; Based on the second data set, the alternative occupation labels The job popularity trend indicator at job level i is defined as .

4. The artificial intelligence-based visual data analysis application method according to claim 3, characterized in that: Construct a career change recommendation model, specifically including: Based on the first data set and the second data set, under the same job level combination, data analysis is performed on the job skill requirement indicators of the first occupation label and each alternative occupation label to obtain a first replacement matching coefficient; Based on the first data set and the second data set, under different spans of job level combinations, data analysis is performed on the job skill requirement indicators of the first occupation label and each alternative occupation label to obtain a second replacement matching coefficient; The combination of different spans of job levels means: combining the job level i of the first job label k with the job level i of the alternative job label Position level i-1 or i+1 combination; i-1 and i+1 represent the adjacent position levels lower than position level i and higher than position level i respectively; The first replacement matching coefficient and the second replacement matching coefficient are comprehensively analyzed to obtain a comprehensive matching index, which is used to comprehensively evaluate the career replacement matching degree between the first career label and each alternative career label.

5. The artificial intelligence-based visual data analysis application method according to claim 4, characterized in that: The first replacement matching coefficient is represented by the following calculation formula: ； in, is the first occupation label k and the alternative occupation label The first replacement matching coefficient is obtained by analyzing the job skill requirement indicators under job level i; Is the first occupation label k and the alternative occupation label The number of skill intersections at job level i; is the total number of skills in the skill list required for the first occupation label k at job level i; The required skills list based on the alternative occupation label has a high degree of overlap with the required skills list of the first occupation label. The value range is ; is an interval representation; when The closer it is to 1, the more alternative occupation labels there are at the same job level i. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority; The second replacement matching coefficient is represented by the following calculation formula: ； in, It is the first occupation label k and the alternative occupation label under different span job level combinations i-1 and i+1. The second replacement matching coefficient is obtained by analyzing the job skill demand indicators; The first occupation label k under job level i and the alternative occupation label The number of skill intersections at job level i-1; The first occupation label k under job level i and the alternative occupation label The number of skill intersections at job level i+1; The required skills list based on the alternative occupation label has a high degree of overlap with the required skills list of the first occupation label. The value range is ; is an interval representation; when The closer it is to 1, the more alternative occupation labels there are at adjacent job levels i-1 and i+1. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority.

6. The artificial intelligence-based visual data analysis application method according to claim 5, characterized in that: The comprehensive matching index is represented by the following calculation formula: ； in, It is a comprehensive matching index, and the output result represents the matching degree of career change between the first career label and each alternative career label; and is the weight coefficient of the corresponding replacement matching coefficient, , and The values are all in the interval (0,1); set up The value range is ; is an interval representation; when The closer it is to 1, the more alternative occupation labels there are at the same job level i or adjacent job levels i-1 and i+1. The higher the coverage of the required skill list for the first occupation label k, the higher the occupation change matching degree and the higher the recommendation priority.

7. The artificial intelligence-based visual data analysis application method according to claim 6, characterized in that: The first dataset is characterized as ; The second dataset is characterized as ; Will and as a set of professional development characteristics; Build a dynamic prediction model based on long short-term memory network. The dynamic prediction model structure includes an input layer, multiple LSTM layers and an output layer. The input layer receives a career development feature set consisting of a first data set and a second data set; After the career development feature set is processed by the LSTM layer, the output layer finally generates the job transfer matching degree between the first career label and each alternative career label; The output layer generates the job transfer matching degree between the first occupation label and each alternative occupation label, specifically including: The output layer receives the data processed by the LSTM layer and converts it into a probability distribution form through the activation function Softmax; The output is each candidate occupation label Relative to the job transfer matching degree of the first occupation label k, the first occupation label k is matched with the alternative occupation label The job transfer matching degree at job level i is recorded as Job transfer matching The value range is between 0 and 1, indicating job transfer matching degree. The larger the value, the higher the priority of the job transfer.

8. The artificial intelligence-based visual data analysis application method according to claim 7, characterized in that: A weighted analysis is performed on the career change matching degree and job transfer matching degree between the first career label and each alternative career label. The specific logic is as follows: in, Is the first occupation label k and the alternative occupation label Combined matching value at job level i; and is the weight coefficient of the corresponding parameter, , and The values are all in the interval (0,1); set up The valid range of is the interval (0.35,1]; Setting the matching filtering strategy specifically includes: like , then the alternative occupation label Occupations marked as high priority for transfer; , , and They are the combined matching value, comprehensive matching index, second replacement matching coefficient and job transfer matching degree judgment threshold, , , ; ; When , it means that at adjacent job levels i-1 and i+1, the alternative occupation label There are more options for transfer positions and grades; Traverse all the alternative career tags and prioritize at least three that meet the requirements. If the number of alternative occupation labels for the conditions is less than three, the Filter out from the alternative career tags The alternative occupation label with the largest value is supplemented.

9. The artificial intelligence-based visual data analysis application method according to claim 8, characterized in that: and The weight analysis results include the following: 1) When the target user only has high requirements for the number of transfer job level selections corresponding to the job skill requirement indicators of the alternative career label and the degree of career change matching, set ; ; 2) When the target users have high requirements for the average years of work experience, salary level, and job popularity trend of the alternative occupation labels, set ; ; 3) When the target users have high requirements for the job skill requirement indicators, average working years indicators, salary level indicators, and job popularity trend indicators of the alternative occupation labels, set ; ; Use the matplotlib or seaborn library in Python to visualize the data of the weight analysis results.

Images