In very simple terms, Machine Learning is the process of using existing data to create a mathematical function.

---

Example

For example, we have an ice cream shop's data that includes the average temperature of the past 30 days, along with the number of ice creams sold on each day. We can analyze this data (average temperature and number of ice creams sold) and can create a mathematical function that fits with the existing data.

Further, this function can take the average temperature as input and return the prediction of the number of ice creams that can be sold on that day.

---

Terminologies

From the existing data, the data that we use as the condition, in our case, the average temperature, is called a Feature. The main data that we are targeting to predict, in our case, the number of ice creams sold, is called the Label.

In mathematical terms, the features are referred using $x$ . Hence, the features can be represented as [ $x_1, x_2, x_3, ..., x_n$ ]; whereas the label is referred to as $y$ .

The process of defining the function using the existing data is called Training. This is the step where the model (a general term used for the mathematical function we create) learns the relationship between features and labels.

Further, the process of using the defined function to get the predicted value is called Inferencing.

In mathematical terms, the function is referred to as $f(x)$ . Moreover, the predicted value by the function is referred to as $\hat{y} = f(x)$

---

There are multiple types of Machine Learning; but at the core, Machine Learning can be classified into two types:

1. Supervised Machine Learning
2. Unsupervised Machine Learning

---

Supervised Machine Learning

Supervised Machine Learning is a type of Machine Learning algorithm in which the training data includes both the feature values and the known label values. This type of algorithm is used to train models by determining a relationship between the features and labels in past observations. This helps the model predict unknown labels for new inputs where the feature values are known.

---

Unsupervised Machine Learning

Unsupervised Machine Learning is a type of algorithm in which the training data only includes the feature values, but no known labels. It determines the relationships and patterns between the features of the observations in the training data.

---

Citation

This blog post is inspired by a Microsoft Learn course's module. While the foundational concepts are based on the course material, I have expanded on them with additional explanations, examples, and insights to better simplify and contextualize the information for readers.