Logistic Regression

• It is a predictive modeling technique.
• It estimates the relationship between dependent and independent variables.
• It is used when the output is in binary format(discrete or in categorical format) whereas linear regression is used when the output is continuous in range.
• Logistic regression curve converts any value to discrete values.
• There is a threshold that segregates the output as positive or negative(0 or 1).
• It comes under the category of Supervised Learning under classification.

Linear vs Logistic Regression

Logistic Regression Equation

Use-Cases:

1. Weather Prediction(will it rain or not)
2. Classification Problem(bird or not a bird)
3. Determines Illness(patient is ill or not)

Steps

1. Teaching the model with the dataset
2. Dropping the non-essential components
3. Determining the output and evaluating the model

Sigmoid Curve

A threshold has to be set above which the value is 1 and below it, the value is 0.