Lecture

Hierarchical Clustering: Dendrograms and Linkage Functions

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

In course

EE-311: Fundamentals of machine learning

Ce cours présente une vue générale des techniques d'apprentissage automatique, passant en revue les algorithmes, le formalisme théorique et les protocoles expérimentaux. ...

Description

This lecture covers the concepts of hierarchical clustering, including clustering by centroids and density, visualization through dendrograms, and different linkage functions such as complete, single, and average linkage. The instructor explains how hierarchical clustering forms clusters at different levels and the criteria for agglomerating clusters based on distance measures.

Instructors (2)

François Fleuret

Michael Stefan Daniel Liebling

Teaching & PhD Past EPFL PhD Students as codirector Christian Jaques, Olivia Mariani, Adrian Shajkofci, François Thierry M Marelli Courses Fundamentals of machine learning EE-311 This course provides a general overview of machine learning, covering the main algorithms, theoretical formalisms, and experimental protocols.

Official source

Related concepts (10)

Four-bar linkage

In the study of mechanisms, a four-bar linkage, also called a four-bar, is the simplest closed-chain movable linkage. It consists of four bodies, called bars or links, connected in a loop by four joints. Generally, the joints are configured so the links move in parallel planes, and the assembly is called a planar four-bar linkage. Spherical and spatial four-bar linkages also exist and are used in practice. Planar four-bar linkages are constructed from four links connected in a loop by four one-degree-of-freedom joints.

Linkage (mechanical)

A mechanical linkage is an assembly of systems connected to manage forces and movement. The movement of a body, or link, is studied using geometry so the link is considered to be rigid. The connections between links are modeled as providing ideal movement, pure rotation or sliding for example, and are called joints. A linkage modeled as a network of rigid links and ideal joints is called a kinematic chain. Linkages may be constructed from open chains, closed chains, or a combination of open and closed chains.

Cluster analysis

Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar (in some sense) to each other than to those in other groups (clusters). It is a main task of exploratory data analysis, and a common technique for statistical data analysis, used in many fields, including pattern recognition, , information retrieval, bioinformatics, data compression, computer graphics and machine learning.

Watt's linkage

In kinematics, Watt's linkage (also known as the parallel linkage) is a type of mechanical linkage invented by James Watt in which the central moving point of the linkage is constrained to travel on a nearly straight line. It was described in Watt's patent specification of 1784 for the Watt steam engine. Today it is used in automobile suspensions, allowing the axle of a vehicle to travel vertically while preventing sideways motion. Watt's linkage consists of three bars bolted together in a chain.

Five-bar linkage

In kinematics, a five-bar linkage is a mechanism with two degrees of freedom that is constructed from five links that are connected together in a closed chain. All links are connected to each other by five joints in series forming a loop. One of the links is the ground or base. This configuration is also called a pantograph, however, it is not to be confused with the parallelogram-copying linkage pantograph. The linkage can be a one-degree-of-freedom mechanism if two gears are attached to two links and are meshed together, forming a geared five-bar mechanism.

Related lectures (64)

Supervised Learning Overview

Covers CNNs, RNNs, SVMs, and supervised learning methods, emphasizing the importance of tuning regularization and making informed decisions in machine learning.

Clustering Methods and Dimensionality Reduction

Covers clustering methods and dimensionality reduction techniques.

Unsupervised Learning: Clustering

Explores unsupervised learning through clustering techniques, algorithms, applications, and challenges in various fields.

Clustering: Hierarchical and K-means Methods

Introduces hierarchical and k-means clustering methods, discussing construction approaches, linkage functions, Ward's method, the Lloyd algorithm, and k-means++.

Clustering: Unsupervised Learning