The exploratory analysis helps users gain insights into single or multiple variables. However, it does not determine what combinations may generate a prediction model, so as to predict the temperature. On the other hand, machine learning can generate a prediction model from a training dataset, so that the user can apply the model to predict the possible labels from the given attributes. In this recipe, we will introduce how to predict the temperature and find the correlation between attributes.
Before predicting we need to see how variables are related and what is the confidence level. We need a corrplot
package to see this. Perform the following steps to find the correlation matrix and confidence interval:
> install.packages("corrplot")> require(corrplot)> mydata$Month = airquality$Month # Removing factors, using original dataFollowing command will...