Machine

Learning
A machine behaves like
HUMAN
 What is machine learning?
 Applications of machine learning
 Types of learning algorithms
◦ Supervised
◦ Unsupervised
 Conclusion
 Final Remarks

Overview
 “Field of study that gives computers the
ability to learn without being explicitly
programmed”
◦ Arthur Samuel (1959)
 “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 (1998)

What is 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”

 As per above,
◦ T: classify emails as spam or not spam
◦ E: watch the user label emails as spam or not spam
◦ P: percentage of correctly-identified emails

Example: Spam Email Detection

 Face recognition

Applications of Machine Learning

 Speech recognition

Applications of Machine Learning

 Other applications:
◦ Spam detection
◦ Internet search
◦ Predictions
◦ Weather Forcasting
◦ Many more!

Applications of Machine Learning

 Supervised learning algorithms learn a
function that can be used to predict the
output associated with new inputs.
 The goal is to approximate the mapping
function so well that when you have new
input data (x) that you can predict the
output variables (Y) for that data.

 i.e Y=f(x)

Supervised Learning
 Given labeled training examples
 Find correct prediction for an unlabeled
example
 E.g. Handwriting recognition

Supervised Learning
 Learning and analysing from input data
itself.

Unsupervised Learning
 Learn from delayed feedback
 E.g. Machine learns how to play chess

Reinforcement Learning
 Machine learning: computers learn from
experience to improve performance of
certain tasks
 Many useful applications that we use
today
 Types of machine learning
◦ Supervised
◦ Unsupervised
◦ Reinforcement
 Conclusion

Remarks-Summary
 Machine learning approaches applied in
systematic quality improvement .
 Increased reviewer agreement appeared
to be associated with improved predictive
performance.
 Helps in detecting spam or wrong with a
perfection.

COnCLUSION
Thank you
Any questions?