The purpose was to build a model to derive insights on what issues are faced by employees of the courier company and forecast the work loss for next year.

Machine Learning algorithms such as Decision Tree, Random Forest & Linear Regression are used. I have also used Ludwig algorithm launched by Uber which is a one-shot place for building any model.

New Concepts Used:

Ludwig Algorithm:

• The algorithm is launched by UBER- Artificial Intelligence team. They are using this algorithm since past many years but just before few months, they made this algorithm public in GitHub. • I tried to apply on our data set and was successfully able to do that after several failed efforts. • The Error Metrics I got was very high RMSE was around 2.34 in numbers which are not at all acceptable.

• Description: o Ludwig is a TensorFlow based toolbox that allows to train and test deep learning models without the need to write code.

• Properties:
o Requires Minimal Coding. o We can perform Sentiment Analysis, Natural Language Processing, Classification, Regression and what not! o We can also perform Deep Learning models. o Currently, visualizations are not available for Jupyter Environment. o This algorithm is going to make a disruptive change in the coming time.

Data Explorer Library: This library helps to create basic analysis report with interactive charts. This library generates a report which contains Missing Value calculation, Data Distribution, Data Analysing report.