Table of Contents

Introduction

1. Linear classifiers
   1. Geometric interpretation of the data points
   2. Classification formalized
   3. Logistic regression
2. Support vector machines
   1. Maximum margin hyperplane
   2. Lagrangian methods for optimization
      1. Primal problem
      2. Dual problem
      3. Application to maximum margin hyperplane
   3. Kernels and kernel trick
      1. Linear kernel
      2. Polynomial kernel
      3. Gaussian kernel (Radial basis function)
      4. Other kernels
   4. Soft margin classifiers
   5. Algorithm implementation
3. Model building
   1. Data preprocessing
      1. Categorical variables
      2. Scaling and standardizing
      3. Grid search
      4. Missing values
   2. Feature selection
      1. Forward selection
      2. Backward selection
      3. F-score
      4. Genetic algorithm
   3. Model selection
      1. Hold-out cross validation
      2. k-fold cross validation
      3. Leave-one-out cross validation
   4. Model evaluation
4. Support vector machines in Credit risk
5. Application
   1. Peer to Peer Lending
   2. Data description
   3. Benchmark model: Logistic regression
   4. SVM with Linear kernel
      1. Optimal cost parameter search
   5. SVM with Gaussian kernel
   6. Model results
   7. Quantifying the edge

Conclusion

Bibliography

Appendix