Mashinali o‘qitishga kirish Nosirov Xabibullo xikmatullo o‘gli Falsafa doktori (PhD), tret kafedrasi mudiri

Unsupervised learning 

• The only requirement to be called an unsupervised learning strategy is to learn a new feature space that captures the characteristics of the original space by maximizing some objective function or minimising some loss function.
• Therefore, generating a covariance matrix is not unsupervised learning, but taking the eigenvectors of the covariance matrix is because the linear algebra eigendecomposition operation maximizes the variance; this is known as PCA.

Unsupervised learning 

• Some of the most common algorithms used in unsupervised learning include:
• (1) Clustering,
• (2) Anomaly detection,
• (3) Neural Networks, and
• (4) Approaches for learning latent variable models.

Each approach uses several methods as follows:

• Clustering
• Hierarchical clustering,
• K-means
• Mixture models
• Dbscan
• OPTICS algorithm
• Anomaly detection
• Local outlier factor
• Isolation forest
• Neural networks
• Autoencoders
• Deep belief nets
• Hebbian learning
• Generative adversarial networks
• Self-organizing map
• Approaches for learning latent variable models such as
• Expectation–maximization algorithm (EM)
• Method of moments
• Blind signal separation techniques
• Principal component analysis
• Independent component analysis
• Non-negative matrix factorization
• Singular value decomposition

Unsupervised learning 

• The goal of clustering is to
• group data points that are close (or similar) to each other
• identify such groupings (or clusters) in an unsupervised manner

Unsupervised learning 

• A cluster is represented by a single point, known as centroid (or cluster center) of the cluster
• Centroid is computed as the mean of all data points in a cluster
• Cluster boundary is decided by the farthest data point in the cluster

Application of Clustering

• Example 1: groups people of similar sizes together to make “small”, “medium” and “large” T-Shirts.
• Tailor-made for each person: too expensive
• One-size-fits-all: does not fit all.
• Example 2: In marketing, segment customers according to their similarities
• To do targeted marketing.
• Example 3: Given a collection of text documents, we want to organize them according to their content similarities,
• To produce a topic hierarchy

Thank you! Contacts Khabibullo Nosirov, Phd Project Manager, Head Of The Department Tashkent University Of Information Technologies named after Muhammad Al-Khwarizmi Radio And Mobile Communications Faculty 100084, Amir Temur 108, Tashkent, Uzbekistan [e-mail] [phone] (WhatsApp) [phone] (Telegram) www.tuit.uz www.spacecom.uz www.intras.uz