Machine Learning: Fundamental Concepts 💻📚

Hello hello, in this blog series, we dive into Machine Learning! This is the first part. First, let’s start with the following question.

What is The Machine Learning?

The scientific field that aims to develop various algorithms and techniques to enable computers to learn in a manner similar to humans is called machine learning.

There are three main types of machine learning

Reinforcement Learning

Reinforcement Learning is used for training in robotics, gaming, and autonomous technologies. The system learns from its failures, and after experiencing something new, it begins new learning.

Unsupervised Learning

Unsupervised learning is used for clustering and dimensionality reduction. This type of machine learning occurs when there are no labels.

Supervised Learning

Supervised learning is the most commonly used method in this field. It involves having a labeled dataset where there is a clear relationship between the independent variables and the dependent variable. It is utilized for building predictive models and performing classification tasks.

Problem Types

Regression Problems

In a Regression Problem, the dependent variable, also known as the target variable, is numeric.

Classification Problems

In a Classification Problems, the dependent variable, also known as the target variable, is categorical.

Methods for Evaluating Model Performance

Linear Regression Models

We use the following formulas to measure the success of regression models. There is no superiority of one over the other. Essentially, they calculate the difference between the predicted value and the actual value to assess performance.

1. MSE (Mean Squared Error)

1. RMSE (Root Mean Squared Error)

1. MAE (Mean Absolute Error):

Logistics Regression Models

1- Accuracy

In the below Logistic Regression section, we’ll delve into other metrics in greater detail at a later stage.

Model Validation

The effort to more accurately evaluate and validate the success of models.

Hold Out Method

The dataset is divided into two parts: the training set and the test set. The modeling or training process is performed on the training set. Subsequently, the model is tested or evaluated using the test set.

K-Fold Cross Validate

The dataset is divided into training and test sets. The training set is further divided into subsets, and within these divided subsets, modeling and testing take place. Then, the model is evaluated using the test set, which it has never encountered before.

Bias

It indicates how far the model is from the real values.

Variance

It specifies how much the model varies across different datasets.

Big thanks to Vahit Keskin and Miuul

Contact me on Linkedin :) yaseminderyadilli