Machine Learning finds it’s applications everywhere in the digital era- Machine Learning is at the core of your email accounts helping you filter spam messages and other unwanted emails, it is at the heart of growing search power of iPhone’s Siri, it is in your Smartphone helping you organize and classify your photos, it is on Walmart recommending you products based on preferences to customize your shopping experience. Gadgets, Equipment’s and all other services in the digital era are getting smarter by helping humans make decisions based on recommended practices and data trends. Wondering, what is machine learning and why is there such a craze?
In digital economy, consumers and producers need to associate with each other much before a transaction can happen. Before the advent of Internet, one could buy books only from the local store which had limited shelf space. The digital era allows book lovers to download any book at any time but the problem with this is the vast number of options available. It is difficult for the consumers to browse through the virtual shelves of an online bookstore which has millions or billions of books for sale. This is applicable to any service or product that has to be procured remotely like booking a hotel room, looking for a job, investments, gifting cakes and flowers, gadgets, looking for a perfect date, tutoring classes, etc. The same scenario applies to songs, movies, blogs, news items, videos or any other webpage. This is the problem of the information era and Machine Learning is a vital part of the solution. Machine Learning algorithms provide perfect matchmaking between consumers and producers by reducing the information overload.
Sept 1st, 2015, Technology Review-“Machine Learning Algorithm Predicts Which New Faces Will Make It as Fashion Models”
Aug 31st, 2015, Forbes-“Qualcomm's Snapdragon 820 uses machine learning to fight malware.”
Oct, 2015, KDnuggets- “Deep Learning Finds What Makes a Good #selfie” by Andrej Karpathy.
All the above headlines relate to how machine learning and deep learning (broader set of machine learning methods to model high-level abstractions in data) are streamlining various business processes and decision making. Machine Learning is an exciting new era for technology that is ruling the world by luring computers to act without specific programming.
What is Machine Learning?
Joseph Sirosh, VP of Machine Learning at Microsoft explains Machine Learning process as - “You take data from your enterprises and make several hypotheses and experiment with them. When you find a hypothesis that you can believe in, you want to put that into production so you can keep monitoring that particular hypothesis with new data.”
A headline on New York Times about Machine Learning – “Brain like Computers, Learning from Experience”
Machine learning is a science that is not new but is gaining fresh momentum now. Machine Learning is a powerful AI technique for crunching petabytes of data to make sense of it in a complicated world by automating analytical model building. Machine Learning helps cognitive systems to learn and engage with the world in a customized way. Machine Learning algorithms learn from the data that they process and analyse without explicitly having to be programmed where to look .The basic concept behind machine learning algorithms is to give the computer an end goal, let the algorithm fail over it again and again till it learns from those mistakes to ultimately achieve the goal i.e. the machine learns from experience.
Machine learning is an iterative concept- as the models are exposed to more data, they learn from fresh feeds of data and tend to independently adapt with new data without constant human intervention. Machine learning algorithms learns from former computations to produce reliable, effective, repeatable decisions and results.
Machine Learning can be broadly classified into 3 different categories-
- Regression –Basically deals with numerical answers.
- Classification –Multiple Choice Answers
- Recommendation –Given a particular user, what are the different products you can show him based on his interests and preferences?
Machine Learning-The Game Changer for Businesses
Learn Data Science in Python to build better Machine Learning Models!
Machine learning, predictive analytics and APIs for that matter are not technologies of the future, but important technologies of the present.” – said Janet Wagner, Machine learning and predictive analytics foster growth
It has been estimated that machine learning is used in more than half of the mobile apps. Machine learning is the unsung hero that powers various business applications whether it is Google’s Self Driving Car, Facebook’s facial recognition, speech recognition of Apple’s Siri or effective web search. Every week there are news headlines about companies finding novel uses of machine learning algorithms that adapt as they are exposed to new data.
Machine learning has great potential to transform business but for it to work effectively the business needs to have access to all possible data sets because more data always wins with machine learning. The best example is to understand how machine learning can help a Telecom company to predict which customers might toss after their contract expires. Machine Learning can help telecom companies predict this by analysing their billing patterns and call usage along with sentiment analysis in the call notes recorded by the customer service executives.
An ex-Google employee said “Everything in the company is really driven by machine learning.”
Machine learning finds extensive use among business executives that want to make the best use of their company’s data by understanding what it is, what it can do for business and what is that they need to watch out for in the data when using it. Machine learning was earlier used to predict what’s happening with a business but now the machine learning algorithms can suggest various strategies for businesses to move forward. Machine learning algorithms that were just analysts before have now become he strategists.
Machine learning contributes to the ROI of big data applications in two ways-
- Increasing the productivity of data scientists
- Finding hidden patterns which even great enterprise data scientists might have overlooked.
Machine Learning Applications
- One of the most interesting application of machine learning is - identifying cat faces in videos. Google connected 1000 PC’s and a machine learning algorithm was subjected to analyse 10 million YouTube videos. Using unsupervised learning, where, the algorithm was not notified upfront to look for cat faces -the machine learning algorithm identified cat faces by identifying which parts of the video would be relevant to the audience after analysing the data patterns.
Image Credit : media.giphy.com
- Cornell University has built a machine learning algorithm that identifies whales in the ocean by analysing the audio recordings to protect them so that ships don’t hit them.
- Amazon is building a machine learning model to identify an employee’s access requirements in the office based on his/her designation. This will help Amazon restrict permission and access for new employees saving them lot of time, money and documentation work.
- Moments app launched by Facebook uses facial recognition technology to classify photos on your smartphone based on when and where they were taken. Their machine learning model then identifies who are the friends in them so that users can sync those photos quickly with only those specific friends.
- Amazon’s machine learning algorithm has been an excellent weapon to fight fake product reviews on the e-commerce website. The algorithms learn which reviews are most useful to customers –identifying the ones that are real and fake. Amazon has litigated charges against several websites that specialize in creating fake Amazon product reviews.
- Google has built a machine learning model to train machines on reading natural language documents and how to summarize them. Just similar to IBM Watson, Google’s machine learning model is built to answer several complex questions without having much knowledge of the language structure.
2016 will be the year of AI and machine learning- as the tech billionaire Mark Zuckerberg takes on 2016 New Year’s resolution to build a personal digital assistant that can run his home and help him with his daily tasks. The king of social media wants to build something like J.A.R.V.I.S (Just a Rather Very Intelligent System) based on the idea from the Iron Man movies, which will understand Zuckerberg’s voice to control everything in his house- temperature, AC, lights, music, etc.
CLICK HERE to get the 2016 data scientist salary report delivered to your inbox!
Machine learning has lots of applications and it all depends on what interests a reader, which is subjective. We have just listed few of them above and there are several others like identifying human genomes, predicting housing prices, predicting taste in movies and music, fraud detection in financial institutions and more. 2016 is definitely going to be the year of Machine Learning and AI.
If know any interesting machine learning applications that are set to remake our world, please share with us in comments below.
Learn Data Science in Python and R to build interesting Machine Learning Applications