/LoanDefault-Prediction

Lending Club Loan data analysis

Primary LanguageJupyter Notebook

LoanDefault-Prediction

This is the code for "Intro to Statistics - Data Lit #2 by Siraj Raval on Youtube at School of AI.

Coding Challenge - Due Date, Feb 9, 12 PM PST

Create a Jupyter notebook with a detailed Exploratory Data Analysis report of this lending club data. Make sure to use the 3 key statistical concepts i mentioned in the video (statistical features, probability distributions, and bayesian stats). Submit your github link in the comments section of the video. I'll give the winner a shoutout in a week!

Model Accuracies:

  • Random Forest with Randomized search CV -- 82.09
  • Logistic Regression with Grid search CV -- 83.18
  • Support Vector Machine with Grid search CV -- 82.50
  • K Nearest Neighbors with Grid search CV -- 77.40
  • Bagging with Base estimator as Random Forest -- 84.10
  • Bagging with Base estimator as Logistic Regression -- 83.10
  • AdaBoost Classifier ----- 83.60
  • MultilLayer Perceptron Classifier ----- 83.40

Note: Check out our project report to find out why we used these algorithms.

Technologies:

  • Programming Language: Python
  • Libraries: Pandas, Scikit-learn, Matplotlib, Seaborn
  • Visualization: plotly

Data Source:

https://www.lendingclub.com/info/download-data.action