Introduction to data mining (especially for healthcare domain)
summary
Hands-on demo using the Orange GUI to build a binary heart‑disease classifier on a cleaned dataset reduced to 14 attributes (originally >76). Data come from the UCI‑style heart file of about 3002 patients and the target is binarized to 0/1. Workflow shows import, edit domain, simple preprocessing, a 70/30 train/test split, decision tree (train precision ≈0.94, similar on test) and a neural network (~87%); confusion matrices and overfitting issues are visualized.
AI-written summary · about
chapters (19)
- 0:00Data mining tutorial
- 2:15Tutorial goals
- 3:21Why Orange
- 5:54Heart disease prediction
- 8:05Dataset attributes
- 14:19Orange tool
- 18:36Purpose of data mining
- 20:24Data mining steps
- 21:51Importing datasets
- 27:29Data preprocessing
- 31:09Target attribute for classification
- 38:39Applying data mining
- 40:03Supervised learning overview
- 47:02Decision tree classifier
- 56:18Model performance
- 1:01:33Confusion matrix
- 1:11:33Visualizing decision tree
- 1:14:31Neural networks
- 1:17:07Resources and next steps
related talks
01h 51mMorning Session: Genomics, Medicine, Epidemiology and Data Interpretation
07mBig Data and Cardiac Rehabilitation
30mReducing Bias in Artificial Intelligence...
47mA Practical(ish) Introduction to Data Science
30mFeature Engineering: An Apprentice’s Guide to the “Dark Art” of Machine Learning
38mThe Real World of Data Analytics
44mA Practical-ish Introduction to Data Science
27mMining social media data of people with rare diseases
