Ensemble methods project

Ensemble methods are learning algorithms that construct a. set of classifiers and then classify new data points by taking a (weighted) vote of their predictions. In this project we will learn how to implement few ensemble techniques which are used in the industry.

What have we learnt so far ?

• What is ensembling?
• Types of ensembling
• Naive aggregation or voting
• Bootstrap Aggregating or Bagging
• Stacking

What are we going to do in the project ?

• We will observe the bagging classifier model behaves and learns on training and testing dataset with the change in no .of estimators.
• We will implement the stacking classifier and measure the accuracy of how our model performed.

What will you learn by doing this project ?

• You will learn as to why ensemble methods are chosen and widely used now a days as they
  • Average out the biases
  • Reduce the variance
  • Unlikely to overfit
• You will be able to build your own stacking classifier with combination of more than one model and hence increase the accuracy of overall model performance.

About the dataset

The data set we are using in previous projects which is named as housing prices.