Machine Learning(MI)

Today is the era of machines and robotics in the world.. Human gives the instruction to the machines they follow it. We know that humans learns from the past experience.. like that why would machines can do the same? Training the machine by accessing the past data and makes machines to  learn from it to get  an accurate result or predictions from it... This is how Machine Learning works.These predictions can be done using programming languages like python,R-language..

How MI is helpful for us ???

Machine learning can be helpful for future predictions in some of the sector such as banking, medical diagnosis ,weather predictions, classification determine area cost in future, company profit etc

There are many confusions between artificial intelligence , Machine Learning and deep learning. lets discuss one by one. Artificial intelligence is the parent of Machine Learning .Here machine will carry out the task as human does. If we ask machine some questions they will think and responds to it. Now machine learning .. as i said before.. we will give the machine the past data and we will make them to learn about the data and it will build a model using some algorithm  for future predictions and the last is deep learning .It is subset of machine learning to train deep neutral network to achieve better accuracy



Machine Learning types:
Supervised learning: Here we are giving labelled data to the machine..For eg : The training data set contains the labelled picture of cats and non-cats.This is more likely to teacher supervise the learning process by telling the machine how exactly cat looks by showing different pictures of cat ...if any of the non-cat image is shown teacher will make the machine that it is not cat.

Unsupervised learning:It is a type of machine learning solutions used to draw inferences from datasets consisting of input data without labelled responses. The most common unsupervise learning method  is clustering analysis  which is used for exploratory data analysis to find hidden pattern or grouping in data.
 

How MI works????

Data which we are giving will be labelled or unlabelled , they need to trained using algorithm to produce the model to get accuracy. When new input data is giving to the machine learning algorithm it makes predictions based on the model.If it attains the accuracy ,the algorithm that used is successful and can be used for future predictions and better decisions .If accuracy is not  acceptable then we need to apply different algorithm to the data set. 

 

 

What are application of MI??

Medical sector:  MI helps in prediction, analysis and even in  classification in  medical field for example: detecting cancer, heart monitoring , retinal scan with the help of new deep learning algorithm  and even it can predict the disease is normal or dangerous.

Banking Sector: It is used to find if any credit card fraud happens, and even used in customer service and mobile banking

Recognition of sound :  It used to differentiate similar sounding word and can be filtered out.

Image Recognition : This is one of the important application of MI in daily basis. 

Predictions : It helps to predict the weather condition and natural calamities.

 

What is the future of MI??

This is one of the fastest growing field in next future. Many industries are using this for various purposes like prediction , analysis and classification. It has computer based  and processing power which are key factor of MI. AI, MI and Deep learning are having bright future as they influenced in market for business growth