UN-BOXING THE BLACK BOX – MACHINE LEARNING WITH PYTHON

If you like this, Share!

“ALPHABET LAUNCHES CHRONICLE TO TURN MACHINE LEARNING ON HACKERS

The latest company to evolve out of Alphabet’s X moonshot program hopes it can mark the spot where hackers are breaking in. The project is entering a crowded market.”- JANUARY 24, 2018 2:13 PM PST

“APPLE’S AR APPS WILL FINALLY BE ABLE TO STICK THINGS ON WALLS

When iOS 11.3 launches, it’ll pack tech developers can use to make their augmented reality apps do things they’ve never been able to do before.” –JANUARY 24, 2018 6:17 AM PST

Reference: Machine Learning updates

Two new developments, which broke up in a gap of just 8 hours. This manifests that its high time to at least understand WHAT is it all about and WHY it is important to know more about it, that is Machine Learning.

This is going to be a series of tutorials through which we will be giving you all the important aspects of machine learning, various techniques and so on. We will also try to present examples using python towards the end. Hopefully by the end of this tutorial you will be confident enough to talk and work on machine learning using python.

Interesting right??? Ok here we go….

INTRODUCTION

Let us start from the text book definition for machine learning. Tom Mitchell described;

“The field of machine learning is concerned with the question of how to construct computer programs that automatically improve with experience. 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”.

AI and ML are two very much misused buzz words in the field of analytics. It is very important to know the line of separation between AI and Machine Learning. AI is a broader term which make us call machines “SMART”, whereas Machine Learning can be termed as a sub part or building blocks where we provide data to the machine and let them learn and make sense out of it and thereby make computers more intelligent.

Let us take an example which is now part and parcel of everybody’s life, Email. The simple problem is to identify the so called spam messages. In order to do this, we should know what pushes a mail into spam category. The nature of spam mails is different for different people. So what is to be done? The machine itself will fish out the required algorithm automatically to find the spam mails, by keeping the known spam pile as base, that is Machine Learning.

To bring a clarity to the concept of machine learning let us take 1 or 2 simple and more common use case. We want to develop an interface to identify whether the person in a photo is male or female, image recognition. So what we will do is we will give the system different pictures of male and female and the machine will automatically extract an algorithm which can accurately predict whether a newly fed photo have male or female.

APPLICATION

  1. Speech recognition: Automatic Speech Recognition (ASR) systems is used to test and optimize continuously.
  2. Effective Web Search: Using Naïve Bayes  we pull out the categories of problems from the huge pool of users entered queries. This will intensify the quality of search results. The query logs are used for the training stage.
  3. Recommendation systems: Prediction of rating or preference a customer would give over other products. Machine learning is used to predict the product or service recommendation based on the huge set of user responses.
  4. Computer Vision: Add-on of AI and cognitive neuron science where the contents of images are automatically figured out by labeling the images. This is done from the training set and by using ML algorithms we can figure out which pixels are relevant and which are not.
  5. Information retrieval: Desired set of information is retrieved from millions of data available. Machine learning algorithms are used for these data mining.
  6. Fraud detection: Machine Learning algorithms helps you in finding the fraudulent in transactions based on a large similar scenario.

MACHINE LEARNING TYPES

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

In the coming tutorials we will be discussing about all these topics with the examples using python.

“Google’s self-driving cars and robots get a lot of press. But the company’s real future is in machine learning, the technology that enables computers to get smarter and more personal.”

– Eric Schmidt (Google Chairman)

So let us start our journey to equip ourselves with the hot skill set of the time, MACHINE LEARNING.

Machine Learning

Leave a Reply

1 Comment on "UN-BOXING THE BLACK BOX – MACHINE LEARNING WITH PYTHON"

  Subscribe  
newest oldest most voted
Notify of

Wow an interesting article. As I read I realize the enormous potential of MI. Looking forward to more updates from you guys!