Unsupervised ML

What is Unsupervised ML?

Unsupervised learning is about discovering general patterns in data. The most popular example is clustering or segmenting customers and users. This type of segmentation is generalizable and can be applied broadly, such as to documents, companies, and genes. Unsupervised learning consists of clustering models, that learn how to group similar data points together, or association algorithms, that group different data points based on pre-defined rules.

For example with this model, an online clothing shop could group certain people based on the products they shop online and advertise different things to them, based on the group they are located in.

---

Types of Unsupervised Learning:

• Clustering

• Dimension Reduction

• Anomaly Detection

• Association

---

Clustering

Clustering is an unsupervised machine learning method of identifying and grouping similar data points in larger datasets without concern for the specific outcome. Clustering is usually used to classify data into structures that are more easily understood and manipulated.

Clustering algorithms:

• K-Means

• Hierarchical
 Clustering

• Gaussian Mixture Models (GMM)

• DBSCAN

---

Dimensiona Reduction

Dimension reduction techniques aim to reduce the number of features in a dataset while preserving its essential characteristics. This is often done to mitigate the curse of dimensionality, improve computational efficiency, and prevent overfitting in machine learning models.

Dimensionality Reduction Algorithms:

• PCA

• SVD

---

Anomaly detection

Anomaly detection, also known as outlier detection, focuses on identifying rare or unusual instances in a dataset that deviate significantly from the norm. Anomalies may represent errors, intrusions, or other unexpected events.

Anomaly detection algorithms:

• Isolation Forest

---

Association

Association rule learning is a rule-based machine learning method for discovering interesting relations between variables in large databases. It is intended to identify strong rules discovered in databases using some measures of interestingness. It is an easy model to learn and understand.

Association algorithms:

• Apriori algorithm