Introduction to Machine Learning.

Machine learning allows the computers learn automatically without human intervention or assistance and adjust actions accordingly. Let’s think about a newly born baby. That baby learn the things around him day by day. Then perform the actions based on past experiences.

Main thing we can identified in Machine learning concept is a Classifier. Classifier is a model which we can input Features (Measurable properties of data set) and labels/targets (value to be predicted by our model) for training. Then this classifier can predict the result(label) for the given testing features.

classifier

Think about a classifier which can identify an apple and an orange. First we have to train the classifier with features and labels. height, width , color , texture are some features for that classifier. Orange, apple are the two labels for that classifier.

Machine learning is a subset of Artificial Intelligence and that is a super-set of Deep Learning.

There are 3 types of Machine learning.

1. Supervised Learning.-learned in the past to new data using labeled examples to predict future events. We have to train the model with a set of features and labels. As an example, when teacher gives the set of questions and answers in the students study that and in the exam when the question is given students answer it according to the previous experiences.
2. Unsupervised Learning .-Used when the information used to train is neither classified nor labeled. Unsupervised learning studies how systems can infer a function to describe a hidden structure from unlabeled data. No labels are given to the learning algorithm, leaving it on its own to find structure in its input.
3. Reinforcement Learning .- Use both labeled and unlabeled data for training. useful when only incomplete labels are available. For examples, driving a vehicle or playing a game against an opponent.

Types of Machine learning based on the type of output(results or labels)

1. Classification- This can be used when the results are qualitative. For predicting discrete labels(class,category). This modeling is the task of approximating a mapping function(f) from input variables(X) to discrete output variables(y)or classes.Then the mapping function predict the labels.

Classification Algorithms (Supervised):

• Support vector machines
• Nearest Neighbors
• Decision Trees and Random Forests
• Gaussian naive Bayes

Classification Algorithms (unsupervised):

• K-Means Clustering.
• Means Shift Clustering
• Gaussian Mixture Models
• Spectral Clustering

2. Regression- This can be used when the results are quantitative.For predicting continuous labels(numbers between negative infinite to positive infinite).This model approximating a mapping function(f)from input variables(X) to a continuous output variable(y).

Regression Algorithms

• Linear regression
• Support Vector Machines
• Random forest regression

3. Clustering- Set of inputs is to divided into groups. Groups are not known beforehand, making this typically an unsupervised task.

First article on Machine Learning.In the next article I will write on how to work with regression problems with examples. Touch with my future articles too.