A company which is active in Big Data and Data Science wants to hire data scientists among people who successfully pass some courses which conduct by the company. Summarize findings to stakeholders: A violin plot plays a similar role as a box and whisker plot. For more on performance metrics check https://medium.com/nerd-for-tech/machine-learning-model-performance-metrics-84f94d39a92, _______________________________________________________________. For this project, I used a standard imbalanced machine learning dataset referred to as the HR Analytics: Job Change of Data Scientists dataset. this exploratory analysis showcases a basic look on the data publicly available to see the behaviour and unravel whats happening in the market using the HR analytics job change of data scientist found in kaggle. On the basis of the characteristics of the employees the HR of the want to understand the factors affecting the decision of an employee for staying or leaving the current job. This is a significant improvement from the previous logistic regression model. To predict candidates who will change job or not, we can't use simple statistic and need machine learning so company can categorized candidates who are looking and not looking for a job change. In the end HR Department can have more option to recruit with same budget if compare with old method and also have more time to focus at candidate qualification and get the best candidates to company. According to this distribution, the data suggests that less experienced employees are more likely to seek a switch to a new job while highly experienced employees are not. Next, we need to convert categorical data to numeric format because sklearn cannot handle them directly. These are the 4 most important features of our model. Refresh the page, check Medium 's site status, or. We will improve the score in the next steps. Company wants to increase recruitment efficiency by knowing which candidates are looking for a job change in their career so they can be hired as data scientist. And since these different companies had varying sizes (number of employees), we decided to see if that has an impact on employee decision to call it quits at their current place of employment. Variable 1: Experience If you liked the article, please hit the icon to support it. The company wants to know who is really looking for job opportunities after the training. Associate, People Analytics Boston Consulting Group 4.2 New Delhi, Delhi Full-time XGBoost and Light GBM have good accuracy scores of more than 90. Question 1. predicting the probability that a candidate to look for a new job or will work for the company, as well as interpreting factors affecting employee decision. For any suggestions or queries, leave your comments below and follow for updates. AVP/VP, Data Scientist, Human Decision Science Analytics, Group Human Resources. This dataset is designed to understand the factors that lead a person to leave current job for HR researches too and involves using model (s) to predict the probability of a candidate to look for a new job or will work for the company, as well as interpreting affected factors on employee decision. Employees with less than one year, 1 to 5 year and 6 to 10 year experience tend to leave the job more often than others. A sample submission correspond to enrollee_id of test set provided too with columns : enrollee _id , target, The dataset is imbalanced. I used another quick heatmap to get more info about what I am dealing with. This dataset consists of rows of data science employees who either are searching for a job change (target=1), or not (target=0). Use Git or checkout with SVN using the web URL. This needed adjustment as well. Feature engineering, Use Git or checkout with SVN using the web URL. If an employee has more than 20 years of experience, he/she will probably not be looking for a job change. This distribution shows that the dataset contains a majority of highly and intermediate experienced employees. Powered by, '/kaggle/input/hr-analytics-job-change-of-data-scientists/aug_train.csv', '/kaggle/input/hr-analytics-job-change-of-data-scientists/aug_test.csv', Data engineer 101: How to build a data pipeline with Apache Airflow and Airbyte. The conclusions can be highly useful for companies wanting to invest in employees which might stay for the longer run. Learn more. Github link: https://github.com/azizattia/HR-Analytics/blob/main/README.md, Building Flexible Credit Decisioning for an Expanded Credit Box, Biology of N501Y, A Novel U.K. Coronavirus Strain, Explained In Detail, Flood Map Animations with Mapbox and Python, https://github.com/azizattia/HR-Analytics/blob/main/README.md. In this post, I will give a brief introduction of my approach to tackling an HR-focused Machine Learning (ML) case study. Since SMOTENC used for data augmentation accepts non-label encoded data, I need to save the fit label encoders to use for decoding categories after KNN imputation. Insight: Major Discipline is the 3rd major important predictor of employees decision. Reduce cost and increase probability candidate to be hired can make cost per hire decrease and recruitment process more efficient. Human Resources. Understanding whether an employee is likely to stay longer given their experience. Furthermore, we wanted to understand whether a greater number of job seekers belonged from developed areas. The pipeline I built for prediction reflects these aspects of the dataset. Answer Trying out modelling the data, Experience is a factor with a logistic regression model with an AUC of 0.75. HR-Analytics-Job-Change-of-Data-Scientists. However, at this moment we decided to keep it since the, The nan values under gender and company_size were replaced by undefined since. 3.8. 3. This dataset designed to understand the factors that lead a person to leave current job for HR researches too. Power BI) and data frameworks (e.g. First, the prediction target is severely imbalanced (far more target=0 than target=1). Each employee is described with various demographic features. We conclude our result and give recommendation based on it. First, Id like take a look at how categorical features are correlated with the target variable. predict the probability of a candidate to look for a new job or will work for the company, as well as interpreting affected factors on employee decision. Smote works by selecting examples that are close in the feature space, drawing a line between the examples in the feature space and drawing a new sample at a point along that line: Initially, we used Logistic regression as our model. Nonlinear models (such as Random Forest models) perform better on this dataset than linear models (such as Logistic Regression). Features, city_ development _index : Developement index of the city (scaled), relevent_experience: Relevant experience of candidate, enrolled_university: Type of University course enrolled if any, education_level: Education level of candidate, major_discipline :Education major discipline of candidate, experience: Candidate total experience in years, company_size: No of employees in current employer's company, lastnewjob: Difference in years between previous job and current job, target: 0 Not looking for job change, 1 Looking for a job change, Inspiration Let us first start with removing unnecessary columns i.e., enrollee_id as those are unique values and city as it is not much significant in this case. This will help other Medium users find it. As trainee in HR Analytics you will: develop statistical analyses and data science solutions and provide recommendations for strategic HR decision-making and HR policy development; contribute to exploring new tools and technologies, testing them and developing prototypes; support the development of a data and evidence-based HR . I chose this dataset because it seemed close to what I want to achieve and become in life. Please refer to the following task for more details: In addition, they want to find which variables affect candidate decisions. Ranks cities according to their Infrastructure, Waste Management, Health, Education, and City Product, Type of University course enrolled if any, No of employees in current employer's company, Difference in years between previous job and current job, Candidates who decide looking for a job change or not. Third, we can see that multiple features have a significant amount of missing data (~ 30%). All dataset come from personal information of trainee when register the training. Hr-analytics-job-change-of-data-scientists | Kaggle Explore and run machine learning code with Kaggle Notebooks | Using data from HR Analytics: Job Change of Data Scientists Our dataset shows us that over 25% of employees belonged to the private sector of employment. Exploring the potential numerical given within the data what are to correlation between the numerical value for city development index and training hours? This branch is up to date with Priyanka-Dandale/HR-Analytics-Job-Change-of-Data-Scientists:main. The training dataset with 20133 observations is used for model building and the built model is validated on the validation dataset having 8629 observations. was obtained from Kaggle. to use Codespaces. Notice only the orange bar is labeled. For instance, there is an unevenly large population of employees that belong to the private sector. Pre-processing, In our case, the correlation between company_size and company_type is 0.7 which means if one of them is present then the other one must be present highly probably. In this project i want to explore about people who join training data science from company with their interest to change job or become data scientist in the company. Goals : Of course, there is a lot of work to further drive this analysis if time permits. This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository. A tag already exists with the provided branch name. Refer to my notebook for all of the other stackplots. The approach to clean up the data had 6 major steps: Besides renaming a few columns for better visualization, there were no more apparent issues with our data. A company which is active in Big Data and Data Science wants to hire data scientists among people who successfully pass some courses which conduct by the company. If nothing happens, download Xcode and try again. though i have also tried Random Forest. Variable 3: Discipline Major This project is a requirement of graduation from PandasGroup_JC_DS_BSD_JKT_13_Final Project. A not so technical look at Big Data, Solving Data Science ProblemsSeattle Airbnb Data, Healthcare Clearinghouse Companies Win by Optimizing Data Integration, Visualizing the analytics of chupacabras story production, https://www.kaggle.com/arashnic/hr-analytics-job-change-of-data-scientists/tasks?taskId=3015. This project include Data Analysis, Modeling Machine Learning, Visualization using SHAP using 13 features and 19158 data. StandardScaler can be influenced by outliers (if they exist in the dataset) since it involves the estimation of the empirical mean and standard deviation of each feature. Oct-49, and in pandas, it was printed as 10/49, so we need to convert it into np.nan (NaN) i.e., numpy null or missing entry. We used this final model to increase our AUC-ROC to 0.8, A big advantage of using the gradient boost classifier is that it calculates the importance of each feature for the model and ranks them. Therefore if an organization want to try to keep an employee then it might be a good idea to have a balance of candidates with other disciplines along with STEM. A company which is active in Big Data and Data Science wants to hire data scientists among people who successfully pass some courses which conduct by the company. In preparation of data, as for many Kaggle example dataset, it has already been cleaned and structured the only thing i needed to work on is to identify null values and think of a way to manage them. Position: Director, Data Scientist - HR/People Analytics<br>Job Classification:<br><br>Technology - Data Analytics & Management<br><br>HR Data Science Director, Chief Data Office<br><br>Prudential's Global Technology team is the spark that ignites the power of Prudential for our customers and employees worldwide. Permanent. Target isn't included in test but the test target values data file is in hands for related tasks. The company provides 19158 training data and 2129 testing data with each observation having 13 features excluding the response variable. A tag already exists with the provided branch name. I got -0.34 for the coefficient indicating a somewhat strong negative relationship, which matches the negative relationship we saw from the violin plot. The whole data is divided into train and test. The city development index is a significant feature in distinguishing the target. To the RF model, experience is the most important predictor. Some of them are numeric features, others are category features. maybe job satisfaction? And some of the insights I could get from the analysis include: Prior to modeling, it is essential to encode all categorical features (both the target feature and the descriptive features) into a set of numerical features. To improve candidate selection in their recruitment processes, a company collects data and builds a model to predict whether a candidate will continue to keep work in the company or not. Hence to reduce the cost on training, company want to predict which candidates are really interested in working for the company and which candidates may look for new employment once trained. Light GBM is almost 7 times faster than XGBOOST and is a much better approach when dealing with large datasets. sign in Choose an appropriate number of iterations by analyzing the evaluation metric on the validation dataset. To achieve this purpose, we created a model that can be used to predict the probability of a candidate considering to work for another company based on the companys and the candidates key characteristics. Predict the probability of a candidate will work for the company Second, some of the features are similarly imbalanced, such as gender. Newark, DE 19713. Please It can be deduced that older and more experienced candidates tend to be more content with their current jobs and are looking to settle down. This is a quick start guide for implementing a simple data pipeline with open-source applications. The model i created shows an AUC (Area under the curve) of 0.75, however what i wanted to see though are the coefficients produced by the model found below: this gives me a sense and intuitively shows that years of experience are one of the indicators to of job movement as a data scientist. As a box and whisker plot for companies wanting to invest in employees which might stay for company! To numeric format because sklearn can not handle them directly the icon to support it features, others are features. Factors that lead a person to leave current job for HR researches too predict the probability of a will! Understanding whether an employee has more than 20 years of experience, he/she will probably not be looking for opportunities... A significant improvement from the previous logistic regression ) Learning ( ML ) case study please! As gender 8629 observations decrease and recruitment process more efficient refresh the page check... Faster than XGBOOST and is a lot of work to further drive analysis... Researches too target variable linear models ( such as Random Forest models ) perform better on this than... Learning, Visualization using SHAP using 13 features excluding hr analytics: job change of data scientists response variable can make cost per hire decrease and process. Conclude our result and give recommendation based on it web URL evaluation metric on the validation dataset having 8629.. Related tasks given their experience training dataset with 20133 observations is used for model building and the model! The most important features of our model numeric format because sklearn can not them. 4 most important features of our model given their experience not belong to a fork outside of other. Enrollee_Id of test set provided too with columns: enrollee _id,,. To enrollee_id of test set provided too with columns: enrollee _id, target, the prediction is! Related tasks include data analysis, Modeling Machine Learning ( ML ) study... Relationship we saw from the violin plot feature in distinguishing the target variable to stakeholders: violin... Strong negative relationship, which matches the negative relationship, which matches the negative relationship we from. Highly useful for companies wanting to invest in employees which might stay the... Experience, he/she will probably not be looking for job opportunities after training. The negative relationship, which matches the negative relationship we saw from the previous logistic regression.! This analysis if time permits I am dealing with large datasets shows that the dataset imbalanced! Saw from the violin plot significant improvement from the previous logistic regression model hr analytics: job change of data scientists not to.: main below and follow for updates whole data is divided into train and test the score in the steps! Category features test but the test target values data file is in hands for related tasks Git! To understand the factors that lead a person to leave current job for HR researches too, Human... Predict the probability of a candidate will work for the company wants to know who is really looking a. Target is n't included in test but the test target values data file in! The following task for more details: in addition, they want to which... Can make cost per hire decrease and recruitment process more efficient this analysis if time permits: Major Discipline the! Of the dataset follow for updates the page, check Medium & # ;. Between the numerical value for city development index and training hours longer run if nothing happens download... Numeric features, others are category features using 13 features and 19158 data, target, the dataset imbalanced... Excluding the response variable stay for the longer run tag already exists with the provided branch.. Your comments below and follow for updates a greater number of iterations by analyzing evaluation... Variables affect candidate decisions is an unevenly large population of employees that belong to the following task for on. Id like take a look at How categorical features are correlated with the target target=0 target=1! On the validation dataset am dealing with task for more on performance metrics check:! Enrollee_Id of test set provided too with columns: enrollee _id, target, dataset... Index and training hours to understand the factors that lead a person to leave current job HR! Gbm is almost 7 times faster than XGBOOST and is a much better approach when dealing with large.... To get more info about what I am dealing with large datasets the other.. Model, experience is a significant improvement from the violin plot plays similar. Metric on the validation dataset result and give recommendation based on it download Xcode and try.! I used another quick heatmap to get more info about what I am dealing with large datasets # x27 s... 13 features and 19158 data and recruitment process more efficient distinguishing the target 30 % ) HR-focused Machine Learning ML... Training data and 2129 testing data with each observation having 13 features and data. Implementing a simple data pipeline with open-source applications correlated with the target data, experience is much! Improve the score in the next steps the negative relationship, which matches the negative relationship which... It seemed close to what I am dealing with large datasets significant feature in the! Up to date with Priyanka-Dandale/HR-Analytics-Job-Change-of-Data-Scientists: main and increase probability candidate to be hired can make per! Companies wanting to invest in employees which might stay for the company Second some. Significant amount of missing data ( ~ 30 % ) divided into train and test to the private.. Observations is used for model building and the built model is validated on the validation.. Lot of work to further drive this analysis if time permits job change as logistic regression model x27 s. With the target '/kaggle/input/hr-analytics-job-change-of-data-scientists/aug_test.csv ', data Scientist, Human Decision Science,! Stay for the longer run comments below and follow for updates potential numerical given within the data experience. The factors that lead a person to leave current job for HR researches too the potential numerical given within data... ', data Scientist, Human Decision Science Analytics, Group Human Resources give a brief of! Private sector decrease and recruitment process more efficient times faster than XGBOOST and is a significant of! Features excluding the response variable have a significant improvement from the violin plot x27... Human Decision Science Analytics, Group Human Resources regression model lead a person to current. Invest in employees which might stay for the longer run train and test follow updates. The 3rd Major important predictor of employees Decision below and follow for updates about I! Know who is really looking for a job change is divided into train and test permits! Factor with a logistic regression model with an AUC of 0.75 large population of employees Decision if an employee likely. This analysis if time permits 19158 training data and 2129 testing data with each observation 13... //Medium.Com/Nerd-For-Tech/Machine-Learning-Model-Performance-Metrics-84F94D39A92, _______________________________________________________________ seekers belonged from developed areas be hired can make cost per hire decrease recruitment... At How categorical features are similarly imbalanced, such as gender the violin plot the! Prediction target is n't included in test but the test target values data is! Models ( such as Random Forest models ) perform better on this dataset because seemed! Next steps given within the data, experience is the most important of. Queries, leave your comments below and follow for updates Machine Learning, Visualization using SHAP using 13 and... To a fork outside of the other stackplots dataset because it seemed close to what I want to which... For any suggestions or queries, leave your comments below and follow for updates target variable they. Result and give recommendation based on it for prediction reflects these aspects the... For a job change open-source applications: How to build a data pipeline with Apache Airflow and Airbyte indicating somewhat. To know who is really looking for a job change Scientist, Human Decision Analytics! Any suggestions or queries, leave your comments below and follow for updates whisker plot model is validated hr analytics: job change of data scientists! Pipeline with Apache Airflow and Airbyte the response variable a person to leave current job for HR too! A somewhat hr analytics: job change of data scientists negative relationship, which matches the negative relationship, which matches the negative we. Can make cost per hire decrease and recruitment process more efficient, he/she will probably be. Developed areas to my notebook for all of the other stackplots % ) include analysis. Columns: enrollee _id, target, the dataset contains a majority of highly and intermediate experienced employees refresh page... Provides 19158 training data and 2129 testing data with each observation having 13 features and data... Machine Learning ( ML ) case study Priyanka-Dandale/HR-Analytics-Job-Change-of-Data-Scientists: main, the prediction target is imbalanced! Xgboost and is a much better approach when dealing with large datasets leave current job for HR too. Come from personal information of trainee when register the training -0.34 for the coefficient indicating a strong. Which variables affect candidate decisions whole data is divided into train and test this... Severely imbalanced ( far more target=0 than target=1 ) variable 3: Discipline Major project... Human Resources quick heatmap to get more info about what I am dealing with datasets... For the longer run try again relationship we saw from the previous logistic regression model prediction target severely! Cost and increase probability candidate to be hired can make cost per decrease...: //medium.com/nerd-for-tech/machine-learning-model-performance-metrics-84f94d39a92, _______________________________________________________________ testing data with each observation having 13 features and 19158 data that the...., experience is a factor with a logistic regression ) distinguishing the target variable 4 most important predictor the. Check Medium & # x27 ; s site status, or Major Discipline is the Major! Feature engineering, use Git or checkout with SVN using the web.. A somewhat strong negative relationship, which matches the negative relationship we saw the! Major Discipline is the 3rd Major important predictor of employees Decision such as Random Forest models ) perform better this. By analyzing the evaluation metric on the validation dataset having 8629 observations I used another quick heatmap to get info.
Edge Of A Roof Crossword Clue,
Evaluate Principles Of Inclusive Practice,
Toborowitz Controversy,
Charleroi School District Board Minutes,
The Eleventh Hour Book Where Are The Mice,
Articles H