Machine Learning is at the heart of Decision Making and Data Science. Under its umbrella of different Supervised and Unsupervised algorithms, lies Logistic Regression which is important in dealing with categorical data for statistical analysis. In this video on Logistic Regression on Customer Data, we will explore core concepts of Logistic regression from an application perspective and operate on Customer data to generate insights practically.
The logistic model or the logit model as they call it is used to model the probability of a certain class or event existing that can be pass/fail, win/lose, alive/dead or red/blue, Young/Old, healthy/sick. It can be extended to model several classes of different events which can be determining whether an image contains apple, orange, peach, etc. Each object is detected in the image will be assigned a probability that is between 0 and 1, with a sum of one.
Logistic Regression is one of the most popular classification algorithms based on probability threshold. Logistic regression is a predictive analysis which is used in predicting discrete values for a given set of features. The algorithm makes use of a sigmoid function to give values between 0 and 1. In this video, the we will walk you through a case study on Customer data of a financial institution. The objective of the case study is to reduce expenses by targeting only those customers who are likely to subscribe using predictive modeling.