AI is giving the human abilities to machines so that they can see, hear, speak, learn like humans.
Studying machine learning
source - https://ml.berkeley.edu/blog/tutorials/
Machine Learning :
Humans have the ability to learn anything. ML is the process of giving this ability to machines.
Intro to ML
- machine learning is all about drawing lines through training data.
- different machine learning algorithms draw different lines and curves
- ML algorithms learn patterns by drawing lines through training data, and then generalizes the patterns it sees to new data.
Example of orange and apple
we have apples and oranges. their size and colour are plotted on graph
a machine learning algorithms draws a line through the data to separate them
now we can predict if a fruit of a given size and colour is an apple or orange
This is the power of machine learning. We take some training data, run a machine learning algorithm which draws a decision boundary on the data, and then extrapolate what we’ve learned to completely new pieces of data. (This is supervised learning using classification technique)
Other applications of ML are :
- classifying tumors as malignant or benign,
- marking emails as spam or not spam, or
- analysing fingerprints for security systems.
Supervised Learning - (need training data)
Classification technique - It is about drawing lines to separate data.
The line is called 'decision boundary'. Example :Regression technique - It is about drawing lines to describe data. The line is called ‘predicter’.