This repo contains 6 regression models of supervised learning under Machine Learning :)
The dataset 'Salary_Data(for Simple LR)' is a linearly separable data and is for the following model(s) :-
- Simple Linear Regression.
The dataset '50-Startups(for Multi LR)' is a linearly separable data and is for the following model(s) :-
- Multiple Linear Regression.
The dataset 'Position_Salaries' is a non-linearly separable data and is for the following model(s) :-
- Polynomial regression.
- Support Vector regression.
- Decision Tree regression.
- Random Forest Regression.
This simple dataset comprises of 30 rows and 2 feature columns. The first feature is the 'Year of Experience' which is the independent variable here and the second feature 'Salary' is the target value(dependent feature).This is a linear data (Plotting and visualization in code).Salaries can be predicted based on the years of experience.
This dataset comprises of 50 rows and 5 feature columns. The first feature is 'R&D Spend' which shows the amount of money spent by each startup on R&D.The second and third features are 'Administration' and 'Marketing Spend' respectively which show the amount of money spent by each startup on the former and latter.Th fourth feature is the 'State' in which the startups are based. All these features are dependent variables. The fifth column 'Profit' is the profit of each Startup and this is the target value. The fact that the target values depend on more than one independent feature makes it a multiple linear regression problem.This is a linear data (Plotting and visualization in code).
This dataset comprises of 10 rows and 3 columns. The first and the second columns are the independent features and 'Salary' is the target feature.Here, positions of employees, their corresponding levels and estimate salaries are given. Suppose a new employee comes for a job and he mentions his position and salary in the previous company. The HR, before hiring, has to tell whether the information given by the new employee is appropriate or is he bluffing. This is a non-linear data (Plotting and visualization in code).