Machine Learning FAQs

Machine Learning Basic Doubts:

What is Machine  Learning?

Machine learning is a core sub-area of AI, it enables computers to get into  a mode of self-learning without being explicitly programmed. When exposed to new data, these computer programs are enabled to learn, grow, change and develop by themselves.

To put simply, the iterative aspect of ML is the ability to adapt new data independently. This is possible as programs learn from previous computations and use "pattern recognition" to produce reliable results.

Why Machine Learning?

Consider some of the instances where ML is applied:the self-driving Google car, cyber fraud detection, online recommendation engines- like Fb's friend suggestion, Youtube's video recommendations, Netflix's movies/shows recommendations and "more item to consider" and get yourself a little something" on Amazon- are all examples of applied ML.

Machine Learning has 2 phases:

• Learning Phase: In learning phase, training data is fed for pre-processing where it goes through normalization, dimension reduction,image processing, etc. Next is learning where supervision / minimization etc. takes place. The third part is error analysis where precision, recall, over-fitting and cross validation of data is tested.

• Prediction Phase: Based on the learning phase, machine would have built a prediction model, now the new data is pushed to the ML model which would predict the future based on test data results. Model keeps on evolving based on new inputs and actual results.