Feature Engineering(Feature Selection) Technique in 60 mins…..

================Background================

Dataset is collected from 50 starts up company from United States with their spending pattern.Startup company has different spending pattern wrt R&D Spend,Administration spend,Marketing Spend. 2.The data also contains profit margins for each locations(States).
The primary objective of this exercise is to identify the best features that can influence the performance for a model specifically when features are numeric in nature.By identifying the best features a better model can be prepared for future predictions.

Link to DataSet: https://github.com/sarojbc75/Repo_SKBC/blob/master/50_Startups.csv

I have used below 5 techniques for feature selection.
1.Correlation Analysis Technique
2.Recursive Feature Elimination Technique
3.Analysis of Variance(ANOVA) Technique
4.Select by model Technique
5.Backward Elimination Tecnique
The model score is evaluated for each technique and finally the technique which gives highest score is selected.