Classification of threats using Clustering Algorithms

Clustering belongs to unsupervised learning algorithms. The unsupervised algorithm is a method in which the references are drawn from the datasets consisting of nonlabeled input data. Clustering is a process of separating the data into groups where members of one group are more similar to each other than members of other groups. In the following picture, we can see clusters of data into a plot.

Clustering is important as it brings out the general property of a group of points. The property of a group could be subjective to the users. The algorithm also needs to assume some points in order to form the clusters among the data points.

There can be various methods of clustering in machine learning. Some of these methods are explained below.

1. Density-based method:- This method assumes the dense regions of the plot have some similarities and to be different from the other regions. It can also be used to merge more than one cluster.

2. Hierarchical Methods:- The cluster formed from these methods are of a tree-shaped hierarchy. Old clusters can be used to form new clusters.

3. Partition methods:- This method divide the points into multiple clusters. Each partition can be used to form another cluster.

4. Grid method:- In this method the data points are divided into a finite number of cells called grids.

Implement this algorithm on the Global Terrorism Database(GTD) for the required result.