offers data science lesson videos made simple!

Sign up or log in to Magoosh Data Science.

7 Machine learning examples

7 Machine learning examples -magoosh

Machine Learning is an interdisciplinary area in nature that brings in techniques from the areas of computer science, statistics, and artificial intelligence, among others. The main components of machine learning research are algorithms which enable this automatic enhancement of experience and which can be functional in a variety of diverse fields.

In this article, you will learn about some examples of Machine Learning. Keep reading!

Smart Email Classification

Classification of your emails into primary, social, and promotion inboxes is all done with the help of an AI approach by Google and also the labeling of your emails as important. Google has outlined its machine learning approach in a research paper titled, “The learning behind Gmail Priority Inbox.” It also highlights large variations in the number of users preferring for volumes of important mail.

Google requires its users to manually intervene to be able to tune their threshold. Whenever a user labels messages in a certain manner, Google makes an increment to their threshold. This way Gmail succeeds in learning every time an Email is marked as important. In a research conducted on Google employees itself to test the fruitfulness of Priority Inbox, it was concluded that people using priority inbox spent around 6% less time reading their emails and wasted 13% less time on reading unimportant emails.

ML for detecting frauds

Various different types of algorithms are currently used by PayPal to manage risks with mostly all pertaining to machine learning: long network, deep learning network, and neural network. One lesson learned by the company over the years is that the best approach to combat fraud is to use the algorithms all at once! Linear algorithms form the norm and are widely in use at PayPal. Drawing a clear straight partition between good and bad customers is something the company used to do. But the approach fails to provide accurate results provided the world being non-linear.

An early realization was made that instead of using one line for marking the differentiation, a better approach would be to use multiple lines or even better, bend the existing straight lines. Hence, most of their algorithms are based on hierarchy. Along with large quantities of data required to be fed into the algorithms and today’s advanced computing infrastructure, PayPal has very effectively boosted neutral net efficiency and the assessment of risks at a very low scale.

Uber Ridesharing App

How are your ride prices determined by Uber? How does the wait time gets minimized once you hail your car? How are you so optimally matched with other passengers so as to minimize detours when booking for your trips? One answer to conquer all these questions – Machine learning.

ATC Jeff Schneider, engineering lead for Uber, discussed in an NPR interview how Machine Learning is used by the company to precisely predict rider demands and thus, ensuring that “surge pricing” (periods during which prices are incremented to increase driver supply and which also decreases rider demands) be no longer needed. Also, the head of Machine Learning in Uber, Danny Lange agreed to the use of machine learning for computing better pickup locations, ETA’s for riders and also for fraud detection.

Healthcare uses ML

The proper combination of data analysis and Artificial Intelligence is greatly enabling the implementation of smarter as well as better healthcare solutions. Regular health telemetry, which was just a thought a few years ago, is now a reality with the rising of smartwatches and other wearable gadgets. Machine Learning is taking everything to a level ahead, in this case, very easily allowing doctors and relatives to constantly monitor the health of their elderly family members. Greater the personal data fed into the algorithm, better the user profile is understood and thus, earlier is the detection of anomalies done by health-care professionals.

Retail uses ML

Backstage of your favorite online retailers also marks the presence of various machine learning algorithms. Tech-giants such as Amazon utilizes this technology to provide more personalized services to its customers. Machine Learning on the basis of your previous purchases and activity, displays you personalized recommendations, largely helping retailers thereby. This results in price variations of retail items over a certain period of time. ML has a big hand in tracking patterns in these price fluctuations and as a result supporting e-commerce companies set prices on the basis of the demand.

ML on Cybersecurity

The automation of cybersecurity tasks, and even their quick deployment in real time to allow the detection of these activities before any damage is done, can be easily acquired with a machine learning approach.

For an example, let’s take a well-trained machine learning model that is able to detect unusual traffic on the network and simultaneously shut these connections as well when they occur. A well-trained model is also capable of detecting malware that can evade human generated signatures, and probably quarantine the samples of malware too, well before they can even execute.

In addition to the above advantages, a trained machine learning model consisting of the standard operating procedure of a given endpoint may even be able to spot when the endpoint gets engaged in odd behavior, which happens most probably when a malicious insider in attempting to steal or destruct sensitive information from the server.

ML on Social Media

Ever wondered how Facebook personalizes your feed and ensures that you see posts of your interests at the top? It’s all thanks to Artificial Intelligence and more precisely, Machine Learning. Due to Facebook’s business interests, the pop-ups and display of ads exactly relevant to your interests–surprising right?! Precisely targeted ads exponentially increase your chances of clicking them and you end up buying something from the advertisers. Facebook and Google during the first half of 2016, efficiently acquired a total of 85% of the online ad market, majorly because of better and smartly targeted advertisements.

The applications of Machine Learning are very varied and important. Machine Learning also contributed heavily to the technical advancement of the world, as you can realize through the above-mentioned examples 🙂

Comments are closed.


Magoosh blog comment policy: To create the best experience for our readers, we will only approve comments that are relevant to the article, general enough to be helpful to other students, concise, and well-written! 😄 Due to the high volume of comments across all of our blogs, we cannot promise that all comments will receive responses from our instructors.

We highly encourage students to help each other out and respond to other students' comments if you can!

If you are a Premium Magoosh student and would like more personalized service from our instructors, you can use the Help tab on the Magoosh dashboard. Thanks!