Introduction to Machine Learning Simplified Version

May 17, 2020

As someone just starting out in Machine Learning and Computer Vision. I created this blog in other to share my ideas while learning Machine Learning to beginners out there, I wanted to be able to talk about Machine Leaning in a Simplified way. When I just started out, trying to wrap my head around Artificial Intelligence and Machine Learning, i tried a lot of different sources finding the best way to understand it.

I was so interested in Artificial Intelligence and was excited to understand how robots worked because in short, it was fascinating to think about. I soon found out that to make some very meaningful progress I had to start with Machine-Learning and i did just that. Enough with the Introduction, I’m going to give a basic and simplified introduction to Machine-Learning (often abbreviated as ML) for complete beginners who want to start their journey in this Field of Study.

Machine Learning is a sub-field of Artificial Intelligence. It is the concept that defines the different methods in which a Machine (preferably referred to as model) can learn from real world data and have the ability to make accurate predictions(decisions). The concept of Machine Learning is a field with many directions and techniques. These in total form the different popular methods that Machine Learning is applied in real world where it can bring significant improvements to human lives by performing Automation, Analysis or Prediction tasks.

It becomes a fun thing to think about how we can harness this power and in what ways it can be of benifit to our personal or financial lives. At a very minimal level, those with technical skills therefore think of the ways that models (a program,software capable of making predictions) can be built for use in improving our analysis or prediction power. Below I would give a brief explantion on some of the areas that Machine Learning Technonlogy is being used in real-world

SOME MACHINE LEARNING TECHNOLOGIES THAT ARE POPULAR TODAY

-->
1. Recommendations Systems: A very popular application of ML models or technology is in the business world as recommender systems. In simple words, a recommendations system is a system deployed for use by businesses to channel the right products to different customers, website vistors, or Users. It uses a simple ML algorithm that can store User data and in future , recommend products the software thinks would suit that Users taste perfectly. The data stored can come in different forms depending on the kind of platform the recommender system should be used. An Example of data a recommender system can store for future use includes (items added to cart, pages visited, time spent on a page,etc). A very Important area to consider when building a recommendations system is where it would be used, because this would generally define the kind of data to be collected and fed to the model.



2. Chatbots: Another ML technology growing at a very fast pace is the Chatbots or Dialog APIs. Chatbots are exactly what their name means, they can have conversations with a human and in a humane way, or well, we are getting there. A chatbot model that has been trained with some ML algorithm has the ability to process sentences and give some response, in most cases, this responses are manually trained by a human i.e A human trainer writes the correct response for different questions,as such, It is to be noted that the most common application of Chatbots are in business websites. They assist Users with a lot of information on how to get things done or to navigate different sections of the site to find what the customer searches for. If you have had any interaction with chatbots, i think you can notice that it isn't a human answering the questions directly. With the current technology, Chatbots are not yet equipped with the best humane touch around, but ofcourse there is reason to believe that this might not be so in a few years. I like the idea of Chatbots, they relieve so much workforce and can just be active anytime of the day fulfilling customers needs and making business mor eproductive.



3. Prediction Software: While not being a main thing You should see around, mostly because a lot of softwares are just an integrating prediction with other types of technology, but for the sake of simplicity, I decided to write about this here afterall it is a ML tech being used. A prediction software or modle is at the very basics of how Machine Learning can be use in real-world. It is probably the easiest to build depending on what area it is to be used. But normally as we have the easiest ML Algorithm capable of making predictions, then yhh, prediction softwares is as easy as ML goes. Prediction softwares are models that have the ability to make predictions and truthfully they can be extremely useful because in practice they cover almost all areas of human lives. One thing that would always make the average human being thrive no amtter the environ ment he finds himself is the ability to make predictions. Through past decades as I'm sure, humans have been able to make predictions, but the really productive thing is a Machine having that ability, this significantly cuts the time in half leaving us to focus our energy and time in collecting data and perhaps devicing better algorithms that can give us more accuracy and faster computational power. And that's exactly what we have been doing these past few years. There are so many researces and advances that it becomes hard to keep up with it all. I think the best we can do is to focus our time in our places of interest, that way everyone is doing a part of evrything that matters.



4. Artificial Intelligence Products: There are a lot of AI products around that do so many fascinating things, they all have their core in ML models. These AI products or APIs are actually built using ML algorithms, trained on diffrent ML Frameworks before they are packaged and deployed on beautiful platforms. To mention a few of them: Video Intelliegnce Tools, Image Classification, Photo Tagging Softwares,and so much more. This AI Products greatly increse human Entertainments, Satisfaction and Effectiveness in different areas.

There are so many other applications of Machine Learning that they can not all be mentioned in this article. What many Experts advise is for us to focus on the sub-branches that we find most interesting, focus our time on it and keep learning and making some advancements by challenging ourselves in any way we can.

We would be looking at the basics of Machine Learning in a simplified way. I should update this article and list some ML Algorithms that help us achieve the differnet advanced technologies we see on the internet today.