Machine learning is a type of artificial intelligence (AI) that provides computers with the ability to learn without being explicitly programmed. Machine learning focuses on the development of computer programs that can teach themselves to grow and change when exposed to new data.
Because of new computing technologies, machine learning today is not like machine learning of the past. It was born from pattern recognition and the theory that computers can learn without being programmed to perform specific tasks; researchers interested in artificial intelligence wanted to see if computers could learn from data. They learn from previous computations to produce reliable, repeatable decisions and results. It’s a science that’s not new – but one that’s gaining fresh momentum.
Essentially, it is a method of teaching computers to make and improve predictions or behaviors based on some data. What is this "data"? Well, that depends entirely on the problem. It could be readings from a robot's sensors as it learns to walk, or the correct output of a program for certain input.
Facebook's News Feed uses machine learning to personalize each member's feed. If a member frequently stops scrolling in order to read or "like" a particular friend's posts, the News Feed will start to show more of that friend's activity earlier in the feed. Behind the scenes, the software is simply using statistical analysis and predictive analytics to identify patterns in the user's data and use to patterns to populate the News Feed. Should the member no longer stop to read, like or comment on the friend's posts, that new data will be included in the data set and the News Feed will adjust accordingly.
Why is machine learning important?
Resurging interest in machine learning is due to the same factors that have made data mining more popular than ever. Things like growing volumes and varieties of available data, computational processing that is cheaper and more powerful, and affordable data storage.
All of these things mean it's possible to quickly and automatically produce models that can analyze bigger, more complex data and deliver faster, more accurate results – even on a very large scale. And by building precise models, an organization has a better chance of identifying profitable opportunities – or avoiding unknown risks.
Using Machine Learning, its possible to handle previously unseen scenarios. Once a Machine Learning model with good generalization capabilities is learned, it can handle previously unseen scenarios and take decisions accordingly. Note that in a traditional program, you need to tell what decisions need to be taken if a particular scenario occurs. Now imagine a billion scenarios are present, you clearly cannot write a code which can handle all these new scenarios. Hence the need for machine learning.
What's required to create good machine learning systems?
- Data preparation capabilities.
- Algorithms – basic and advanced.
- Automation and iterative processes.
- Ensemble modeling.
Some well-known applications of Machine Learning are as follows:
- Customer feedback for businesses on Twitter
- Online recommenders in e-commerce sites such as Amazon or Netflix?
- The self-driving Google Car
- Fraud detection systems
- Facebook news feed
What is the future of Machine Learning?
The variety of applications that Machine Learning supports includes search engines, image recognition, speech analysis, filtering tools, and robotics. The author of the article, Where Machine Learning Is Headed, predicts that in the coming year, the global community will witness a tremendous growth of smart apps, digital assistants, and main-stream use of Artificial Intelligence.
Machine Learning will proliferate the mobile market and enter the territories of drones and self-driving cars. The demand for making algorithms more easily available will push vendors to offer many new Machine Learning tools. Though such canned products will be available in the market, the skills required to fine tune existing algorithms, tweak the data, and develop an advanced model will remain in demand.
Machine Learning is a new tool for better forecasting. In businesses, forecasting demand is increasingly becoming an insurmountable challenge, frequently leading to erroneous results and The trends in the demand data fluctuate so much, and the inherent causes behind those fluctuations are so complex that understanding demand variability is beyond the scope of most business leaders and managers. Moreover, manual factors intensify the human bias in demand planning activities. Now Machine Learning seems to offer a solution for demand forecasting. With the inherent capability to learn from current data, Machine Learning can help to overcome challenges facing businesses in their demand variations.
What are the drawbacks of Machine Learning?
Thomas Frey, a futurist at The Da Vinci Institute, has predicted these three inevitable consequences of a machine driven age. These disadvantages could have long-term effects across the business spectrum:
- The increasingly automated lifestyle presented by Machine Learning systems is gradually eroding the human strength to fight odds and overcome obstacles. If life becomes that simple and trouble free, the entire human race can become quite vulnerable to threats and sudden turn of fortunes!
- Machine-driven solutions will tend to have a “canned” look and feel, thus devaluing originality and reducing the chances of “innovative” solutions.
- Too much machine dependence can dissolve human interdependencies – the core of human civilization. Is that something desirable?
With excerpts from: