Machine Learning with R
summary
Classifies handwritten digits from 8x8 pixel vectors using 5‑nearest neighbors via caret, yields a confusion matrix (common confusions: 1↔8, 9↔7). Fits linear regression on the built‑in prestige dataset with lm, finds education and log(income) significant while women is not. Clusters Chicago assault and burglary lat/long into six groups with kmeans and plots centers on Google Maps using ggplot2/ggmap; clustering visualizes but adds little insight beyond the raw map. All code is on GitHub and KataCoda.
AI-written summary · about
chapters (21)
- 0:00Intro
- 0:47About me
- 2:15Machine Learning?
- 3:26Movies Genres
- 5:27Data-based classification
- 7:35Data Visualization
- 14:05Publishing the model
- 18:00Supervised Machine Learning workflow
- 19:21Classification problem
- 19:58Data preparation
- 20:42K-Nearest Neighbours Algorithm
- 21:41Naive Bayes classifier
- 22:29Logistic regression
- 22:40Evaluation methods for classification
- 27:04Ordinary linear regression
- 27:57Evaluation methods for regression
- 29:15Prestige dataset
- 31:57Categorical data for regression
- 33:26Clustering problem
- 34:13K-means Algorithm
- 35:03Chicago crimes dataset
related talks
51mMachine Learning with Ruby
44mHello Deep Learning
52mMachine Learning Crash Course
40mTools and Tricks from a Pragmatic Data Scientist
37mJavaScript Machine Learning from Magic to Science
54mMachine Learning explained with Examples
40mBuilding a Handwriting Recognition System
45mFrom Ancient Greeks to Modern Geeks - Basic Machine Learning Algorithms