Explaining Absenteeism at Workplace Predicted by ML

While employers expect employees to miss a certain number of workdays each year, excessive absences can equate to decreased productivity and can have a major effect on company finances, morale, and other factors. Excessive absenteeism is defined as a person being absence from work for more than 3 hours.

In this project is to help predict absenteeism at Company X during work time. The project is broken up into 3 phases:

• Phase 1: Data Pre-processing and Exploratory Analysis.

Data pre-processing involves data cleaning, data transformation, data reduction, and data selection.

• Phase 2: Predictive Modeling

In the modeling phase, there were two important things that was performed. The first step was data treatment by utilising the scikit-learn module in Python. The data was divided for training and testing, the ratio was 80% : 20% respectively. Then the second step was to apply odds ratios. Odds is the ratio of the probability that an event will occur versus the probability that the event will not occur, or probability / (1-probability).

• Phase 3: Visualizing Predicted Probability Model.

Visualizing of the predicted probability of the model. Click here to view the interactive version.

To view more of my data science projects, head to my Github