A Brief Intro- Artificial Intelligence, Machine learning, Deep Learning and Data Science.
Past a year, when Google DeepMind’s AlphaGo program beated a South Korean Master in the oldest game of Go, media was coming up with terms like Artificial Intelligence, Machine Learning and Deep Learning to describe how a program won. At the same time people Harvard Business Review calls a job of data scientist as “The Sexiest Job of the 21st century”. What are these? Are these all same?
People use these term interchangeably but these are not the same things.
In this blog post, I would like introduce you to these terms and how they are different yet related to each other.
AI is future.
AI is science fiction.
AI can make computers take over humans.
It is the science of making intelligent machines, more specifically intelligent computer programs. It is how intelligently a computer understands human interaction and takes action.
AI is not new!
Research in AI had begun from 1950’s. The earliest achievement in this field was the ‘Turing Test’ presented by British Mathematician Alan Turing. The Turing Test involves a Human Judge asking questions to two entities of which one is human and the other is computer. If the judge fails to distinguish the human and computer then the computer passes the test.
AI involves writing all the steps that we as human can think upon as a set of instructions and then with experience the computer chooses which move increases chances of achieving some goal. There has been humongous amount of research going on in this field done by big daddies of technology to startups aiming to automate and solve human problems.
Remarkable product built in this field is Self Driving cars by Google and Tesla.
AI has made humans think about starting a living on Mars, life on a whole different planet. This is revolutionary and possible due to humans working on AI.
Spoiler alert! This is not at all related to Mechanics.
Machine learning is a part of AI where we make machines learn. Yes, I know Machines ain’t got brains like we do. Then how is possible make it learn?
A human learns with experience similarly a computer learns by data. Imagine a black box to which you feed some examples then ask a question on basis of those examples and it predicts answer from the examples you fed before. This black box which predicts your output is the ML algorithm.
Basically Machine learning is a part of AI in which the machine predicts possible outcomes on the basis of data it already has. Unlike in AI that uses heuristics, you don’t need to specify the instructions explicitly in ML.
We use Machine learning products in our daily lives. Didn’t know this? We use all email these days. Who sorts which email can be a spam? ML does.
We all use e-commerce websites like Amazon or Flipkart. They use recommendation systems to give suggestions on the basis of product you viewed or purchased which is another implementation of ML.
The reseach in this field is going so fast that people have built Stock Prediction algorithms and are working to improve cyber security using ML.
When more than one layer of neural networks is used then it is deep learning. Thus we can say that deep learning is used to implement Artificial Neural Networks. Deep learning again belongs to Machine learning where specific type of algorithms are used.
We try to implement biological system to a computer.
Again, what are the uses?
The new camera phones which predict gender and age of the person being captured is based on Convolutional Neural Networks which is part of Deep Learning.
Creating images or automating subtiles while a video is being played is also deep learning.
Sentiment Analysis of text is another example. Apple’s Siri and Hike’s Natasha ie. the chatbots are also made using deep neural networks.
With the advent of Deep Learning, there are now more emerging field in AI to be discovered. More revolution to arrive.
People confuse a lot whether Machine learning belong to AI or Data Science. Well, it belongs to both.
Before answering this, I would tell you what Data Science is.
Data Science as the name suggests is the science of Data. Wikipedia suggests that Data science is a “concept to unify statistics, data analysis and their related methods” in order to “understand and analyze actual phenomena” with data.
Data Science is a vast field and it consists of mathematics, statistics, computer science with subfields of machine learning, classification, data mining, databases and visualisation.
Data science is all about data, cleaning the data, extracting features from it, applying machine learning and statistics and show some visualizations of data to obtain relations between features etc.
As machine learning has most of its part as statistics, thus, we say that ML belongs to Data Science while the system is still learning it also belongs to AI. ML is the connector between Data Science and AI.
Data Science is extensively used in businesses for their growth as business intelligence and also to get relevant results about data.
For example, having data of city we can tell about Male vs. Female ratio, the average income of a family, birth and death rates, and lots more which can help the government to take action to the desired place.
About CauseCode: We are a technology company specializing in Healthtech related Web and Mobile application development. We collaborate with passionate companies looking to change health and wellness tech for good. If you are a startup, enterprise or generally interested in digital health, we would love to hear from you! Let's connect at email@example.com