Binary Classification using Logistic Regression and Neural Networks

29 Oct 2020

Goal of this project

The goal of this project is to solve classification problems using logistic regression and neural networks. There are 2 parts to this analysis:

Predict whether Ad was clicked on or not using logistic regression model and compute different metrics to evalute the model
Predict synthetically generated data using neural network and compare it with logistic regression performance

Analysis 1 - Logistic regression decision boundary

precision_recall roc K-means clustering

Analysis 2 - Neural Network decision boundary

Guassian Data Logistic regression Neural Network

Data Source :

Advertising dataset from Kaggle

Links to artifacts