# Machine Learning Algorithm Methods

As a basis for understanding machine learning datasets, several machine learning algorithms will be studied, namely logistic regression, a decision tree for classification problems, and linear regression for regression problems. Logistic regression is one of the most popular basic classification algorithms. In simple terms, Logistic regression is almost the same as linear regression but linear regression is used for Label or Target Variables in the form of numeric or continuous values, while Logistic regression is used for Labels or Targets in the form of categorical/discrete values – related site.

Examples of continuous value are house prices, stock prices, temperature, etc.; and examples of categorical value are predictions of SPAM or NOT SPAM (1 and 0) or predictions of customers SUBSCRIBE or UNSUBSCRIBED (1 and 0). Generally, Logistic Regression is used for binary classification (1/0; Yes/No; True/False) problems, but some data scientists also use it for multiclass classification problems. Logistic regression is a linear classifier, therefore, logistic regression also uses the same formula or function as linear regression. So here are the machine learning algorithm methods;

1. Supervised machine learning algorithms
Supervised machine learning is a machine learning algorithm that can apply existing information to data by assigning certain labels, such as previously classified data (directed). This algorithm can provide a target for the output made by comparing past learning experiences.

2. Unsupervised machine learning algorithms
Unsupervised machine learning is a machine learning algorithm that is used on data that does not have information that can be applied directly (undirected). This algorithm is expected to be able to find hidden structures in unlabeled data.

3. Semi-supervised machine learning algorithms
Semi-supervised machine learning is an algorithm used to perform labeled and unlabeled data learning. Systems that use this method can increase the efficiency of the resulting output.

4. Reinforcement machine learning algorithms
Reinforcement machine learning is an algorithm that can interact with the learning process carried out, this algorithm will give points (rewards) when the given model gets better or reduce points (errors) when the resulting model gets worse. One application that is often found is in search engines.