Hi Crawbears,
Today, we’ll look into the question: Will a new bank customer default on his or her loan? We’ll optimize, train, make predictions with, and evaluate four classification models – K Nearest Neighbor (KNN), Decision Tree, Support Vector Machine (SVM), and logistic regression – for loan status of new customers. We’ll work with a bank data set of 346 customers with key variables such as loan status, principal, terms, effective date, due date, age, education, and gender. A bank’s department head, for example, could apply a predictive model to better structure loans and tailor terms to various target customers. Let’s break it down:
Part 1: Cleaning and wrangling, including converting data types, using .to_datetime, and replacing values
Part 2: Exploratory analysis, including plotting stratified histograms, working with groupby, creating new relevant variables, and making observations to determine key features
Part 3: One hot encoding to convert categorical variables with multiple categories to binary variables using pd.get_dummies and adding new features using pd.concat
Part 4: Feature selection of predictors (X) and labeled target (y)
Part 5: Normalizing feature set using scikit learn’s preprocessing.StandardScaler.fit.transform
Part 6: KNN, including determining and plotting optimal k value, training model and making predictions on test set, generating a confusion matrix heatmap and report, evaluating jaccard and F1 scores
Part 7: Decision Tree, including determining and plotting optimal max depth, training model and making predictions on test set, visualizing decision tree using pydotplus and graphviz, generating a confusion matrix heatmap and report, evaluating jaccard and F1 scores
Part 8: SVM, including determining and plotting optimal kernel function, training model and making predictions on test set, generating a confusion matrix heatmap and report, evaluating jaccard and F1 scores
Part 9: Logistic Regression, including determining and plotting optimal regularization and numerical solver, training model and making predictions on test set, calculating probability, generating a confusion matrix heatmap and report, evaluating jaccard, F1, and log loss scores
Part 10: Evaluating model performance head-to-head by creating a dataframe of accuracy scores for KNN, Decision Tree, SVM, and Logistic Regression models to make comparisons
We’ll cover cleaning, wrangling, and visualizing techniques, apply important scikit learn libraries to develop, optimize, train, and make predictions, and walk through evaluating and comparing models. Let’s dig in.
If there’s something you’d like to explore, model out, or visualize, reach out to info@crawstat.com!
Our family will be away in the mountains and wilderness for the next week (I got four walkie talkies!) so we’ll dive back into modeling when I return.
See you next week!
Rish