CO 1 - SESSION 2
• Introduction
• Well posed learning problems
• Terminology
• Applications
Introduction
• ML is a branch of AI concerned with design and
development of algorithms that deals with
“construction and study of systems that can learn from
data.”
• ML can empower computers learn and behave more
intelligently. It is enabled by iterating over past data
and be ready to predict any future data.
Ex. Humans – learn from past experiences
Robot—Learn from past data
Introduction
Few examples:
• We have a small house of $70,000 and big house for
$160,000.Now how to predict the cost of a medium house.
• Detecting spam emails (e.g. with words like “cheap’, “free” etc.).
What is the probability that these mails containing the word
“cheap” is spam. We will categorize based on the word. If 80%
of mails indicate it is spam, we can conclude the mails with
word “ cheap” is spam.(Naive Bayes)
2 3/10 4/10 No
3 7/10 6/10 ?
Introduction
Other examples
“ Is this cancer?”,
“What is the market value of this house?”,
“Which of these people are good friends with each
other?”,
“Will this rocket engine explode on take off?”,
“Will this person like this movie?”,
2) Tom Mitchell gave a “well-posed” definition that has proven more useful
to engineering types: “A computer program is said to learn from experience
E with respect to some task T and some performance measure P, if its
performance on T, as measured by P, improves with experience E.”
E.g. So if you want your program to predict, for example, traffic patterns at a
busy intersection (task T), you can run it through a machine learning
algorithm with data about past traffic patterns (experience E) and, if it has
successfully “learned”, it will then do better at predicting future traffic
patterns (performance measure P).
Terminology
In machine learning we use the following terms quite frequently.
Features: These are distinct characteristics that can be used to describe
each item in a quantitative manner.
COLOUR: RED
TYPE : FRUIT
SHAPE: ROUND
Terminology
Sample: It is an item to process (e.g. classify)
Ex: Document Classification
Terminology
Picture: