Machine learning is a part of Artificial Intelligence in which software applications learn from input data or experience without being explicitly programmed.
Machine learning focuses on developing computer programs that can access data and use it to learn for themselves.
What Is Machine Learning?
The primary aim of Machine Learning is to allow computers to learn autonomously without human intervention or assistance and adjust actions accordingly.
Machine Learning is dependent on input data, instructions, and data sets similar to how a human brain learns new things, and gains knowledge, and experience.
If we have a model defined up to some parameters, we say that the model is learning if it optimizes the parameters using the training data or past experience.
Mitchell (1997) defines Machine Learning algorithm as :
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
Mitchell (1997)
The machine learning process begins with observations of input data, instructions, and past experiences.
It looks for patterns in data so it can later make predictions for the new data.
Why Machine Learning?
We are living in the age of big data. We all are Producers of data, everything we do we generate data, every time we buy a product, we watch a movie, visit any web page, or post on social media we are generating data.
With this data we want our needs to be understood, for example a supermarket chain wants to be able to predict which customer is likely to buy which product, to maximize sales and profit.
Similarly, each customer wants to find the set of products best matching his/her needs. If we look at things, we find that these things are not random.
Algorithms are used to solve a problem, but unfortunately for predicting customer behavior we don’t have any computer algorithm and similarly for such other things, we have lots of data.
From this data, we can detect certain patterns or regularities and such patterns can be used to make predictions. This is the niche of Machine Learning.
Machine Learning also helps us find solutions to many problems in computer vision, speech recognition, and robotics.
For example: recognizing faces, we humans recognize the faces of friends and family members by simply looking at their faces, how? We do not know how. We are not able to explain it.
We know that faces are not just random collections of pixels (digital pictures). We know that a face has a well-defined symmetric structure.
Each person’s face has eyes, a nose, a mouth, located in certain places on the face. Each person’s face has a particular combination of these.
A learning program analyses sample face images of a person, captures the patterns specific to that person, and then recognizes them by matching those patterns in a given image.
Types Of Machine Learning
Supervised
Supervised Learning used labled data to train the model and predict the output value of the test data.
Un Supervised
Un-Supervised Learning used data which is not labled or classified.
Reinforcement
Reinforcement Learning used trail and reward mechanism to predict the values.