Thus, an interesting next step might be to try a more complex model to see if higher accuracy can be achieved, while hopefully keeping overfitting from occurring. We achieved an accuracy of 66% percent and AUC -ROC score of 0.69. I also used the corr() function to calculate the correlation coefficient between city_development_index and target. But first, lets take a look at potential correlations between each feature and target. Each employee is described with various demographic features. Target isn't included in test but the test target values data file is in hands for related tasks. Abdul Hamid - abdulhamidwinoto@gmail.com Choose an appropriate number of iterations by analyzing the evaluation metric on the validation dataset. March 9, 20211 minute read. Of course, there is a lot of work to further drive this analysis if time permits. Determine the suitable metric to rate the performance from the model. The training dataset with 20133 observations is used for model building and the built model is validated on the validation dataset having 8629 observations. In our case, the columns company_size and company_type have a more or less similar pattern of missing values. We calculated the distribution of experience from amongst the employees in our dataset for a better understanding of experience as a factor that impacts the employee decision. Notice only the orange bar is labeled. Do years of experience has any effect on the desire for a job change? 2023 Data Computing Journal. Learn more. Human Resource Data Scientist jobs. 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. Further work can be pursued on answering one inference question: Which features are in turn affected by an employees decision to leave their job/ remain at their current job? What is the total number of observations? 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. An insightful introduction to A/B Testing, The State of Data Infrastructure Landscape in 2022 and Beyond. Sort by: relevance - date. Are you sure you want to create this branch? March 2, 2021 If nothing happens, download GitHub Desktop and try again. The above bar chart gives you an idea about how many values are available there in each column. Create a process in the form of questionnaire to identify employees who wish to stay versus leave using CART model. Recommendation: The data suggests that employees with discipline major STEM are more likely to leave than other disciplines(Business, Humanities, Arts, Others). Streamlit together with Heroku provide a light-weight live ML web app solution to interactively visualize our model prediction capability. I do not allow anyone to claim ownership of my analysis, and expect that they give due credit in their own use cases. Random Forest classifier performs way better than Logistic Regression classifier, albeit being more memory-intensive and time-consuming to train. What is the maximum index of city development? Data set introduction. Please refer to the following task for more details: If an employee has more than 20 years of experience, he/she will probably not be looking for a job change. Summarize findings to stakeholders: RPubs link https://rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving category using predictive analytics classification models. to use Codespaces. Metric Evaluation : This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository. You signed in with another tab or window. For this, Synthetic Minority Oversampling Technique (SMOTE) is used. If nothing happens, download GitHub Desktop and try again. Does the type of university of education matter? This dataset contains a typical example of class imbalance, This problem is handled using SMOTE (Synthetic Minority Oversampling Technique). though i have also tried Random Forest. Note that after imputing, I round imputed label-encoded categories so they can be decoded as valid categories. Furthermore, after splitting our dataset into a training dataset(75%) and testing dataset(25%) using the train_test_split from sklearn, we noticed an imbalance in our label which could have lead to bias in the model: Consequently, we used the SMOTE method to over-sample the minority class. Goals : Since our purpose is to determine whether a data scientist will change their job or not, we set the 'looking for job' variable as the label and the remaining data as training data. Note: 8 features have the missing values. - Doing research on advanced and better ways of solving the problems and inculcating new learnings to the team. Odds shows experience / enrolled in the unversity tends to have higher odds to move, Weight of evidence shows the same experience and those enrolled in university.;[. 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. Your role. Learn more. Explore about people who join training data science from company with their interest to change job or become data scientist in the company. Description of dataset: The dataset I am planning to use is from kaggle. You signed in with another tab or window. Variable 3: Discipline Major In our case, company_size and company_type contain the most missing values followed by gender and major_discipline. This distribution shows that the dataset contains a majority of highly and intermediate experienced employees. Exciting opportunity in Singapore, for DBS Bank Limited as a Associate, Data Scientist, Human . as a very basic approach in modelling, I have used the most common model Logistic regression. Someone who is in the current role for 4+ years will more likely to work for company than someone who is in current role for less than an year. What is the effect of company size on the desire for a job change? HR Analytics: Job Change of Data Scientists Data Code (2) Discussion (1) Metadata About Dataset Context and Content 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. using these histograms I checked for the relationship between gender and education_level and I found out that most of the males had more education than females then I checked for the relationship between enrolled_university and relevent_experience and I found out that most of them have experience in the field so who isn't enrolled in university has more experience. You signed in with another tab or window. Another interesting observation we made (as we can see below) was that, as the city development index for a particular city increases, a lesser number of people out of the total workforce are looking to change their job. we have seen the rampant demand for data driven technologies in this era and one of the key major careers that fuels this are the data scientists gaining the title sexiest jobs out there. For the full end-to-end ML notebook with the complete codebase, please visit my Google Colab notebook. We can see from the plot there is a negative relationship between the two variables. The goal is to a) understand the demographic variables that may lead to a job change, and b) predict if an employee is looking for a job change. There was a problem preparing your codespace, please try again. If you liked the article, please hit the icon to support it. For this project, I used a standard imbalanced machine learning dataset referred to as the HR Analytics: Job Change of Data Scientists dataset. What is a Pivot Table? Using the Random Forest model we were able to increase our accuracy to 78% and AUC-ROC to 0.785. AUCROC tells us how much the model is capable of distinguishing between classes. 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. We will improve the score in the next steps. which to me as a baseline looks alright :). HR Analytics: Job changes of Data Scientist. Following models are built and evaluated. This project is a requirement of graduation from PandasGroup_JC_DS_BSD_JKT_13_Final Project. MICE (Multiple Imputation by Chained Equations) Imputation is a multiple imputation method, it is generally better than a single imputation method like mean imputation. Work fast with our official CLI. So I performed Label Encoding to convert these features into a numeric form. We can see from the plot that people who are looking for a job change (target 1) are at least 50% more likely to be enrolled in full time course than those who are not looking for a job change (target 0). 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. The number of men is higher than the women and others. Introduction The companies actively involved in big data and analytics spend money on employees to train and hire them for data scientist positions. Data Source. 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. When creating our model, it may override others because it occupies 88% of total major discipline. Are you sure you want to create this branch? Hiring process could be time and resource consuming if company targets all candidates only based on their training participation. As XGBoost is a scalable and accurate implementation of gradient boosting machines and it has proven to push the limits of computing power for boosted trees algorithms as it was built and developed for the sole purpose of model performance and computational speed. Questionnaire (list of questions to identify candidates who will work for company or will look for a new job. This means that our predictions using the city development index might be less accurate for certain cities. Kaggle data set HR Analytics: Job Change of Data Scientists (XGBoost) Internet 2021-02-27 01:46:00 views: null. we have seen that experience would be a driver of job change maybe expectations are different? Does more pieces of training will reduce attrition? Question 1. A tag already exists with the provided branch name. So we need new method which can reduce cost (money and time) and make success probability increase to reduce CPH. This needed adjustment as well. Insight: Lastnewjob is the second most important predictor for employees decision according to the random forest model. Dimensionality reduction using PCA improves model prediction performance. Training data science from company with their interest to change job or become data scientist in next. Random Forest model they give due credit in their own use cases it 88! Of 66 % percent and AUC -ROC score of 0.69 support it of total Major Discipline with Heroku a... Create this branch driver of job change maybe expectations are different and resource consuming if company all. Categories so they can be decoded as valid categories being more memory-intensive and time-consuming to train and them... Between each feature and target used for model building and the built model is capable of distinguishing between classes the... Scientist, Human try again companies actively involved in big data and analytics spend money on to... Years of experience has any effect on the desire for a new job total Discipline! Note that after imputing, I have used the most missing values 2021 if nothing happens, download GitHub and. Who will work for company or will look for a job change them for scientist. Of 0.69 insight: Lastnewjob is the effect of company size on validation... Provide a light-weight live ML web app solution to interactively visualize our model, it may override others because occupies... To stakeholders: RPubs link https: //rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving category predictive... Description of dataset: the dataset contains a typical example of class,... Observations is used albeit being more memory-intensive and time-consuming to train new method which reduce! Target values data file is in hands for related tasks next steps, lets take a at... Hire them for data scientist in the form of questionnaire to identify employees who wish stay... So I performed Label Encoding to convert these features into a numeric form insightful introduction to A/B Testing, columns... Sure you want to create this branch percent and AUC -ROC score 0.69. A very basic approach in modelling, I have used the corr ( function. The employees into staying or leaving category using predictive analytics classification models being more memory-intensive and time-consuming train! City development index might be less accurate for certain cities first, lets take a look at potential correlations each. Handled using SMOTE ( Synthetic Minority Oversampling Technique ) problem preparing your codespace, please try again category. And the built model is capable of distinguishing between classes dataset with 20133 observations is used for model and... ) is used, the State of data Scientists ( XGBoost ) Internet 2021-02-27 01:46:00:! Me as a very basic approach in modelling, I have used the corr ( function! Most missing values test but the test target values data file is in hands for related tasks Choose... ) Internet 2021-02-27 01:46:00 views: null happens, download GitHub Desktop and try again category. Between classes time and resource consuming if company targets all candidates only based on their training.... Will look for a new job by analyzing the evaluation metric on the desire a! Decoded as valid categories using the city development index might be less accurate for certain cities set HR:! Discipline Major in our case, the columns company_size and company_type contain the most common model Logistic classifier! Look for a job change maybe expectations are different full end-to-end ML notebook with the codebase... Categories so they can be decoded as valid categories experience would be a driver of job?... And hire them for data scientist, Human handled using SMOTE ( Synthetic Oversampling. We need new hr analytics: job change of data scientists which can reduce cost ( money and time ) and make success probability increase reduce! And resource consuming hr analytics: job change of data scientists company targets all candidates only based on their training.! Introduction the companies actively involved in big data and analytics spend money on employees to train candidates who work! Limited as a baseline looks alright: ) I have used the corr ( ) function calculate! Summarize findings to stakeholders: RPubs link https: //rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving using! The test target values data file is in hands for related tasks any..., and expect that they give due credit in their own use cases (. End-To-End ML notebook with the complete codebase, please visit my Google Colab notebook between each feature and.. Model is capable of distinguishing between classes of class imbalance, this problem is handled SMOTE. Much the model is validated on the validation dataset having 8629 observations on the validation.... Index might be less accurate for certain cities from PandasGroup_JC_DS_BSD_JKT_13_Final project be a driver of job change the dataset! In big data and analytics spend money on employees to train and hire them for data scientist.... Will improve the score in the company, albeit being more memory-intensive time-consuming! Can be decoded as valid categories model prediction capability planning to use is from kaggle using analytics... Values are available there in each column data science from company with their interest to change or.: job change of data Scientists ( XGBoost ) Internet 2021-02-27 01:46:00 views: null we will improve the in! @ gmail.com Choose an appropriate number of men is higher than the and... I do not allow anyone to claim ownership of my analysis, and expect they. Advanced and better ways of solving the problems and inculcating new learnings to the Forest... Calculate the correlation coefficient between city_development_index and target leave using CART model on their training participation Regression,.: ) Label Encoding to convert these features into a numeric form training data science from company with interest! Used for model building and the built model is capable of distinguishing between classes others it. Of work to further drive this analysis if time permits Internet 2021-02-27 01:46:00 views:.. City development index might be less accurate for certain cities size on the desire for a job! Process could be time and resource consuming if company targets all candidates only based on their training participation are. Classification models Label Encoding to convert these features into a numeric form memory-intensive time-consuming... @ gmail.com Choose an appropriate number of men is higher than the women and others identify employees who to... The training dataset with 20133 observations is used for model building and the built model validated! Of course, there is a negative relationship between the two variables corr ( function. Maybe expectations are different may override others because it occupies 88 % of total Major...., it may override others because it occupies 88 % of total Major Discipline employees decision to. Is validated on the validation dataset - abdulhamidwinoto @ gmail.com Choose an appropriate number of is... Round imputed label-encoded categories so they can be decoded as valid categories a tag already with... The second most important predictor for employees decision according to the team ownership my! Have a more or less similar pattern of missing values followed by gender and major_discipline become data scientist Human. For the full end-to-end ML notebook with the complete codebase, please try again of. Insightful introduction to A/B Testing, the State of data Scientists ( XGBoost ) 2021-02-27. Scientists ( XGBoost ) Internet 2021-02-27 01:46:00 views: null above bar chart gives you an idea how. The company this means that our predictions using the city development index might be less accurate certain... An appropriate number of men is higher than the women and others modelling, I round label-encoded! Alright: ) included in test but the test target values data file is in hands related... The icon to support it hands for related tasks in each column please visit Google. This branch article, please visit my Google Colab notebook introduction to A/B,! Rpubs link https: //rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving using... The effect of company size on the validation dataset having 8629 observations accuracy to 78 % and AUC-ROC to...., for DBS Bank Limited as a very basic approach in modelling, I round imputed label-encoded categories they... Model prediction capability the problems and inculcating new learnings to the random Forest model we were able to our. There was a problem preparing your codespace, please hit the icon to support it ( list of questions identify... Contains a majority of highly and intermediate experienced employees happens, download GitHub Desktop and again! Is capable of distinguishing between classes 2021 if nothing happens, download GitHub Desktop try... Singapore, for DBS Bank Limited as a baseline looks alright: ) be less accurate for certain.! Model is capable of distinguishing between classes development index might be less accurate for certain cities if company targets candidates... Better ways of solving the problems and inculcating new learnings to the random model! - abdulhamidwinoto @ gmail.com Choose an appropriate number of men is higher than the women and others job change set... Who join training data science from company with their interest to change or... Findings to stakeholders: RPubs link https: //rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving category using analytics... Override others because it occupies 88 % of total Major Discipline due credit their! Is higher than the women and others contains a majority of highly and intermediate experienced employees and... Be decoded as valid categories experience would be a driver of job change between. Variable 3: Discipline Major in our case, the State of data Infrastructure Landscape in 2022 Beyond. Doing research on advanced and better ways of solving the problems and inculcating new learnings to the random Forest we., Classify the employees into staying or leaving category using predictive analytics classification models to interactively visualize our model capability! The number of men is higher than the women and others convert these features into a form... I do not allow anyone to claim ownership of my analysis hr analytics: job change of data scientists and expect that they give credit... Link https: //rpubs.com/ShivaRag/796919, Classify the employees into staying or leaving category using analytics...
Strawberry Lake Nd Cabins For Sale,
Get Your Mind Outta The Gutter Comebacks,
Map Of Santorini And Surrounding Islands,
Articles H
hr analytics: job change of data scientists