Introduction to Machine Learning.

Machine Learning gives machines the ability to learn and perform certain tasks without being explicitly programmed.

“A computer program is said to learn from experience E regarding some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E.”

 The processes involved in Machine Learning

To get started with machine learning, you need to know the machine learning algorithm. Machine Algorithms are a set of instructions and mathematical techniques you pass to the computer to perform specific and required tasks. 

An algorithm is a logic behind the machine learning model. Examples of machine learning models are the Linear regression model, Logistic regression model, Decision tree model, and more. A model is the major component of machine learning. 

A machine learning process involves building a model that can solve any problem statement.

Needs for Machine Learning

•  Increase in data generation
•  Improvement required in decision making
•  Identifying trends and patterns in Data
• Machine learning allows the extraction of the data and predicts the data and patterns with the help of Machine learning Algorithms. These algorithms help in the distribution of data.
• Solve complex problems
• Machine Learning helps to solve complex tasks for example self-driving cars.

Types of Machine Learning?

A machine can learn based on three approaches which are:

1. Supervised Learning
2. Unsupervised Learning
3. Reinforcement Learning

1. Supervised Learning: Supervised learning is a method where the machine is fed with well-labeled data. In this type of machine learning you feed the machine with a guide (labeled data) and then it predicts its output using that guide. The labeled data-set is what you pass into the machine. It might be a collection of images. These images will train the module to understand the data. The labeled data set is nothing but the training data set. That’s why it’s called Supervised machine learning i. e guide or help.
2. Unsupervised Learning: These models aids to train the machine with unlabeled data with no guide or supervision. An example is Generated Adversarial Network (GANS). In this type of machine learning, the model does not know about the data that is being fed, unlike Supervised learning. The model tries to understand the patterns to predict the output. 
3. Reinforcement Learning: This is used for advanced Machine Learning. This is a part of machine learning where an agent is put in an environment and he learns to behave in his environment by performing certain actions and observing the rewards which it gets from those actions. This type of Machine Learning is mainly used in self-driving cars and so on.

Types of Machine Learning Problems

1. Regression
2. Classification
3. Clustering

1. Regression: Regression problems give the output in continuous quality. The regression problem can be solved using Supervised Learning Algorithm. An example of a supervised learning algorithm is Linear regression. An example of its application is in a weather forecast or prediction
2. Classification: Problems give the output in the form of categorical value I.e. it only has true values such as; true or false, yes or no, high or low, and so on. We can solve its problems using Machine Learning classification algorithms. Some examples of Classification algorithms are Logistic regression and Support Vector Machines. An example of its application is in filtering spam messages or emails.
3. Clustering: These Problems are used for creating clusters. It involves assigning the input into two or more clusters based on feature similarity. Its problems can be solved by using Unsupervised learning algorithms.

Why Python is the best for Machine Learning?

1. Python is Easy to use and learn.
2. It has so many libraries for Machine Learning and frameworks.
3. Open-source and large community.
4. Easier implementation.