Posts

Fuzzy Clustering

C-means algorithm

Fuzzy sets are used in fuzzy logic and can be considered as a generalisation of set theory. An element can be a member of a particular set or not in set theory

Read More

Centroid-Based Clustering

K-Means and PAM

Centroid-based or representative-based clustering is a method of finding the best k clusters of items in the D data set.

Read More

Clustering Algorithms

What is CLustering

Unsupervised machine learning methods aim to find patterns or groups (clusters) in data sets so that the most similar items in the data set will be gathered in the same cluster

Read More

Classification Performance Measures

How to measure classification performance

Multiple methods exist to measure the performance of a classification algorithm and classify a data set into two classes, positive and classified.

Read More

K-Nearest Neighbours

Very simple inefficient algorithm for classification

KNN classifier is a nonparametric lazy classifier. In nonparametric classification the algorithm does not assume any specific distribution for the data sets.

Read More

Support Vector Machine

Elegant mathematical model for classification

Support Vector Machine is a binary parametric classifier that classifies items by creating a hyperplane between classes.

Read More

Rule-Based Classification

A Variation of Decision Tree

A rule-based classifier uses a set of rules to classify items in a data set. The rules are formalised in the form of IF-THEN clause.

Read More

Decision Trees Classification

A Simple and effective classification algorithm

Decision Tree is a classification model which recursively partitions the data space into two parts. The split can be considered as a hyperplane parallel to one axis of the data space.

Read More