Logistic regression is a well-known algorithm used in the field of statistics and machine learning. It is best used to predict responses which are categorical in nature (yes or no).
Similar to linear regression, the goal of this algorithm is to find the values for the coefficient that weight each input variable. Unlike linear regression whose outcome is continuous, the outcome of logistic regression is binary. The prediction of the outcome is transformed using a non-linear function called logistic function. Its an S shaped curve which translates values into the range of 0 and 1. The equation is
P = 1/ 1+e –(a+bX)
Where P is the probability of 1, e is the base of the natural logarithm and a & b are the parameters.
The way model is learned, the predictions made by logistic regression can be used as the probability of class 0 or class 1. As with other algorithms, we need to remove attributes which are unrelated to the output variable as well as attributes which are correlated to each other. These are best used to solve binary classification problems.
Now, will I win this week’s awesome machine learning book or not ? ¯_(ツ)_/¯
By: Jaya Kuppuswamy