Machine Learning
and
Neural Networks
Machine Learning
"A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E"
Tom M. Mitchell
Supervised Learning
- Training based on labeled data
- The "right answer" is known
- Two types: Regression and Classification
Supervised Learning Example
House price prediction
Supervised Learning Example
Type Classification
Unsupervised Learning
- Training based on non-labeled data
- The "right answer" is NOT known
- Finds clusters instead of types.
Unupervised Learning
Cluster Detection
Google News
News clustering. Show all related news together
DNA Data Analysis
Find patterns in very long DNA strings
Social Networks
Find clusters of people (friends, family, coworkers, etc)
Univariable Linear Regression
Gradient Descent
Artificial Neural Networks
Demo Time
The talk is over!
Thank you.