Introduction to Machine Learning

Hello Technotizers, it’s pretty obvious that at this moment you are sitting in front of your device and gazing through these lines of text on the screen of your machine. Machines have played a very important role for the mankind from time they have been come into being. In these hilarious months of the global pandemic, every person on this planet was stuck at home seeing more machines than people. The day started with scrolling tweets on the mobile and ended by watching Netflix on television. Every minute of the day was dependent on some machine. An activity like playing games also got restricted to playing games online with a computer. And it also ended up winning sometimes!! So have you ever thought who gave your computer the superpower to play games and also win against its master? The answer to this question lies in Machine Learning. What exactly machine learning is and how does it work? There are numerous definitions of machine learning and they all decently define this magic field. Some of them are mentioned below:

“Machine Learning at its most basic is the practice of using algorithms to parse data, learn from it, and then make a determination or prediction about something in the world.” – Nvidia 

 “Machine learning algorithms can figure out how to perform important tasks by generalizing from examples.” – University of Washington

“The field of Machine Learning seeks to answer the question “How can we build computer systems that automatically improve with experience, and what are the fundamental laws that govern all learning processes?” – Carnegie Mellon University

Just reading definitions won’t do the job; let’s peek into an example which will probably help us in clearly understanding Machine learning’s purpose.

Consider we're creating a game of Rock, Paper, Scissors. When we play this with a human, it's very basic; every child can learn it in just a few minutes. Now, let's have a look at the most root part of a game that the human mind is really very good at, that is recognizing what it's actually looking at. Humans can look at the hand gesture and instantly recognize which ones are rock, which ones are paper, and which ones are scissors. But how would we program computer to recognize them? Think about all of the diversity of hand types, skin color, and even people who do scissors with their thumb sticking out, and people who do scissors with their thumb in. If you’ve ever written any kind of code, you’ll instantly realize that this is a really difficult task. It might take us thousands or tens of thousands of lines of code, and that's just to play rock, paper, or scissors. So what if there was a different way to teach a computer to recognize what it sees? What if we could have a computer learn in the same way that a human does? That's the core of machine learning and the path to artificial intelligence. So traditional programming looks like this. We have data, for example, a feed from the webcam, and we have rules that act on this data. These rules are expressed in a programming language and are the bulk of any code that we write. Ultimately, these rules will act on the data and give us an answer. Maybe the computer sees a rock, maybe a paper, and maybe it sees scissors. But what if we turn this around, and instead of we as the programmer figuring out the rules, we instead give it answers with the data and have the computer figure out what the rules are. That's machine learning!! So now, it is possible to tell the computer that how rock looks, how a paper looks and how a scissor looks by having a lot of pictures of the three gestures. And can have a computer figure out the patterns that match them to each other. Then, the computer will have learned to recognize a rock, paper, and scissors. That's the core of building something that uses machine learning. We get a set of data that has patterns inherent in it, and we have a computer learn what those patterns are. Those patterns will be then used to play the game we desired for. So machine learning twists the traditional programming method by giving the answers beforehand. 

Machine learning is like teaching a baby how to talk or do certain things. The baby sees, listens and absorbs what the parents say and learns everything step by step growing better with each step. It observes the surrounding and tries to imitate and modify what it has learnt earlier. Same is with machine learning, it learns from the given data and gets better with each iteration reducing the errors every time. So as this little baby is growing day by day, it’s being used by various companies in different forms to make their products and services much more efficient than before.

Virtual Personal Assistants: Most common examples of these are Google assistant, Apples’s Siri, Amazon’s Alexa, etc. These products are developed to provide personal assistance to customers while they use their devices and try to give smoother user experience.

Product Recommendations: We all have might experienced seeing various advertisements of products and services while using different applications. Recommendation engines are built using Machine learning in a way to filter your recent searches and activities and try to recommend to you similar products by matching over thousands of options. 

Image Recognition: Image recognition is a part of computer vision and is counted as one of the most common uses of machine learning. Security, surveillance, visual geolocation, object recognition, industrial automation, gesture recognition, etc. are wide applications of image recognition.

Speech Recognition: Speech recognition is the translation of spoken words into the text. It is being widely used in virtual assistants, automotives, by medical practitioners to capture voice notes while diagnosis and also for voice-based authentication.

Face Recognition: It is another powerful application of Machine learning which more widely used for security purposes to verify people that can cause a theft. It is also applied in Forensic investigations and is also the backbone of trending social media platforms like Snapchat, Instagram and Facebook.

Classification: Machine learning models are used to classify objects or phenomenon very easily using various algorithms. Examples include segregating emails as spam and not spam, classifying flowers into various species, identifying cancer tumors, etc.

Prediction: As the name suggests, prediction is to predict certain values, events or properties. Weather forecasting, anomaly prediction, sales forecasting, prediction of labels or types, etc. are the various applications.

And this list continues to grow until it fascinates you with the outspreaded applications of Machine learning. Machine learning has the ability to automate almost everything and will create wonders in the near future.

Hope this article helped you in knowing what Machine learning exactly is. See you then!! Keep coding and exploring new techs.!!