Tutorial - Intro to Data Science
summary
Hands‑on build of a loan‑default classifier using a 614×13 CSV: code runs in Google Colab with pandas & scikit‑learn and matplotlib. Workflow: data import, null handling, map categorical to integers, feature selection, train/test split, decision tree then random forest. Initial tree shows 98% train accuracy but ~70% test (overfitting); adding dependents, gender and co‑applicant income raises test accuracy to ~76%. Predict_proba returns default probabilities for each applicant.
AI-written summary · about
chapters (12)
- 0:00Introduction to session
- 3:02What is data science
- 22:37Data science skills and AI
- 45:54Business case loan default
- 1:09:13Setting up Google Colab
- 1:45:23Exploring data with pandas
- 2:37:56Decision tree modeling
- 3:03:41Training and testing models
- 3:30:49Random Forest algorithm
- 3:47:02Handling null values
- 3:50:39Using prediction probabilities
- 4:01:36Q&A and wrap-up
related talks
27mData Science and ML with an emphasis on Banking...
57mHow To Evaluate and Improve Machine Learning Models
39mMachine Learning in Financial Credit Risk Assessment
44mA Practical-ish Introduction to Data Science
46mIntroduction to Machine Learning
04h 02mHands-on Session on Data Science for Beginners
30mFeature Engineering: An Apprentice’s Guide to the “Dark Art” of Machine Learning
27mUsing Machine Learning to solve a classification problem with scikit-learn