This is a Machine Learining approach using Random Forest Regression to predict the average temperature (Farenheit) of different cities-
- Chennai
- Delhi
- Mumbai
- Kolkata on any given date(mm-dd-yyyy)
The temperature dataset of these cities were obtained from https://www.kaggle.com/riturajsaha/temperature-of-different-cities-of-india
R2 Score value of different models are used to compare them, r-square is a statistical measure that represents the goodness of fit of a regression model. The ideal value for r-square is 1, its range is from -1 to 1.
Below is comparison of r-Square value of different cities after applying Random Forest Regression.
City : | Chennai | Delhi | Mumbai | Kolkata |
---|---|---|---|---|
Random Forest Regression | 0.1120 | 0.4816 | 0.0223 | 0.3661 |
Researchers evaluate their models based on r-square values or in other words effect sizes. According to Cohen (1992) r-square value .12 or below indicate low, between .13 to .25 values indicate medium, .26 or above and above values indicate high effect size.