Where the Machine Learning is Used

Today I would like to write few lines about applications of machine learning algorithms in real life. And the honest answer to the question in title is – “Everywhere”. Literally. I will give you few examples, but this only the most common cases and mostly very generalized. So let’s begin.

1. Image recognintion

The perfect example is Facebook with its face recognition of your friends. When you upload a photo of you and your friends to your album Facebook gives you a suggestion – “oooh, look I found that you are with your friend James and John on this photo, would you like to tag them?”. And your first reaction is “What a hell, how do you know that, soulless machine?”. I guess everyone experienced that. And answering your what-the-hell question, Facebook uses machine learning algorithm to recognize faces on your photo, founds your face and few others, then compares it with faces of your friends on FB and when he founds a match he starts to annoy you. Seems simple? Yes and no. For those who work with this – simple, for a person who has nothing to do with this technology and IT in general something unbelievable and fantastic. Later I will write an article about image recognition, trying to explain how machine does it.


You’re buying something on Amazon, watching your favorite movie/series on Netflix/Pornhub  or listening to the audio-book on Audible and below you see: “these items may be interesting for you” or “users who bought/watched/listened this also did this”. There are few algorithms that do that – Apriori, Eclat, Boltzmann Machine, Autoencoders and others. Every one of those has its own logic, but in general they analyze actions of the user, create patterns (features) and look for similarities in bigger database. And yes, big brother knows everything about you.

3. Voice recognition

I guess every person that has smartphone already tried that little button with mic icon in WhatsApp or any other messaging application. Or someone tried to google using voice. Or you were playing with Siri/Alexa/Cortana/Google Assistant asking them dumb questions :D. All these became possible thanks to machine learning algorithms.

4. google search

Have you noticed that when you type Google is already trying to guess what are you going to search? And usually it guesses very well, doesn’t it? While you might be scared that Google knows everything about you… You should. Because it does :D. And it does because it has the history of your searches and familiar searches of other people, so it can predict what you might be looking for. Also it improves the search every time by analyzing your behavior after it presented the results: did you click on first link, did you change the search words and so on. Google gathers this info and next time when you or someone else makes a query using the same keywords it presents better results.  And again – thanks to machine learning.

5. Email spam filters

I think this actually one of the first fields where the machine learning algorithms started to be used. First spam-filters were pretty logical, but with development of technology and available methods it became more intelligent and smart, so now you don’t even notice these spammy letters in your mailbox.

6. cyber security and fraud detection

According to research, every “new” malware has 90-98% of repetitive code, what makes it much easier to detect using machine learning. Security programs that are powered by these algorithms understand the pattern and also can take care of 2-10% really new variations.

Also machine learning is used more and more for fraud detection in different industries. PayPal for example analyses millions of transactions to fight money laundering and gets better and better, because machines learns.

7. healthcare

I still didn’t make a deep research on it, but as far as I know reading the titles, machine learning is a beast (or will become in recent years) in medicine. It detects malicious patterns much better than human. Next sentences are proudly stolen from other article. One study used computer assisted diagnosis (CAD) when to review the early mammography scans of women who later developed breast cancer, and the computer spotted 52% of the cancers as much as a year before the women were officially diagnosed. Additionally, machine learning can be used to understand risk factors for disease in large populations. The company Medecision developed an algorithm that was able to identify eight variables to predict avoidable hospitalizations in diabetes patients.

8. Facebook ads

Here will be really quick example which is enough to understand the power of machine learning. Facebook can target 30-40 years old (or any age range you set) housewives (or any occupation you think of) that live in your city (or any city you choose) and like horror movies (or any interest you decide). And yes, Facebook will show your ad only to that audience. This is really impressive.

9. Predictions

Not the predictions what will happen with your ass in the future based on the lines of your palm. Computer won’t tell you that he sees “a fire… big-big fire in your future and you should be aware of this” (OMG I should make a podcast on this, this sounds hilarious in my head :D). So back to reality. Weather prediction – machine learning, finance trading – the same, trends in economics – yep. Any type of prediction that can be made based on previous data – all using machine learning. And please, don’t think it will tell you that exactly this is going to happen. No. It will be something: based on the data I have in the next few units of time COULD happen this, because it happened in the past.

10. natural language processing

Have you seen those cute chat-bots that literally can understand you? Yep – machine learning in action. Also natural language processing is used to scan comments and understand if they are supportive or not. So imagine you have an app in Google Play or Apple Store with a lot of comments and it is just impossible for a human to go through all of them. You can download all the comments, run through machine learning algorithm and understand what users feel and think about your app.

Guys, machine learning and artificial intelligence is huge, it’s already has tons of applications and in the near future everything will be around it. Ahh, and yes. Although it’s powerful, the level of current AI is far far away from Skynet and other scary things all of you have seen in Hollywood movies. So, don’t panic.

Leave a Reply

Your email address will not be published. Required fields are marked *