Learning Machine Learning Concepts and a today story

Hi

I hope you all are fine. So, I decided in my previous article that this semester I won't waste any time and will focus on the things that actually need to be done. I also mentioned that I have two major courses: one is Machine Learning, and the other is Software Engineering. There are some things I already know well, but for now, it’s Saturday, a holiday from university. I am actually talking about the previous day since I am writing this article at 3:08 AM on Sunday. However, the learning I did was on Saturday.

So, what did we learn? First, I analyzed the machine learning course content provided by the university. It’s good, but I need some significant changes just for myself because I want to learn machine learning concepts in-depth and contribute to the AI industry, whether as a consultant or an engineer. I believe every engineer must know machine learning and all these concepts of Artificial Intelligence because it’s already part of our university syllabus and is a trending technology. We need to be aware of it. So, I’ve decided to dive into learning everything about machine learning. So What we have learned today.

Machine learning is a technique that enables a computer to learn from data, allowing it to make decisions or predictions based on patterns within that data. The type of data we use is important for choosing the right learning approach, as it can be either labeled or unlabeled.

What is Labeled Data?

Labeled data is data where each example has both an input and a known output. In other words, each data point is paired with an "answer" or "label" that identifies its category, class, or expected outcome.

• Example: Imagine we have a picture of a cat. If we annotate this picture by writing "Cat" on it, then this becomes labeled data because we’ve provided the information that it is a cat.
• Usage: Labeled data is used in supervised learning, where the model learns from this labeled data to predict labels for new data.

What is Unlabeled Data?

Unlabeled data is data that doesn’t have associated labels. Here, we only have the input data without any identifying information or target values.

• Example: Consider the same picture of a cat, but this time there’s no label or identifier saying "Cat." This is unlabeled data because the system doesn’t know what the image represents.
• Usage: Unlabeled data is used in unsupervised learning, where the model identifies patterns or groupings in the data without any guidance from labels.

Types of Machine Learning

There are two main types of learning approaches in machine learning based on the type of data:

1: Supervised Learning

Data: Uses labeled data.

How It Works: The model is trained on input-output pairs and learns to map inputs to the correct outputs. It "supervises" itself by comparing its predictions to the known labels, adjusting as needed to improve accuracy.

Example: In a spam filter, labeled emails as "Spam" or "Not Spam" help the model learn to identify which messages should be filtered out.

Common Algorithms: Linear Regression, Decision Trees, Support Vector Machines, and Neural Networks.

2: Unsupervised Learning

Data: Uses unlabeled data.

How It Works: The model looks for patterns, similarities, or structures within the data itself. Since there are no correct answers to guide it, the model has to find relationships in the data on its own.

Example: In customer segmentation, an e-commerce platform might group customers based on shopping behaviors, even though it doesn’t know specific customer labels.

Common Algorithms: K-Means Clustering, Hierarchical Clustering, Principal Component Analysis (PCA), and Autoencoders.

Well, I also learned about the concepts of AI vs. ML vs. Deep Learning. If you want to know more about this, just go to Irfan Malik's YouTube channel—he explains very well how to understand the differences between AI, ML, and DL. It’s very late now, so I need to sleep. Sorry for not designing any banners; I might add one in the future. So its a start will do wonders in future. Your comments are very valuable for me. I also worked on my Web Development projects just need to put some more changes and then I will add them on Linkedin and Github as well so Good By for now. Looking forward to meet you again.