Clustering with Neural Network and Index

TL;DR

This paper introduces CNNI, a neural network-based clustering model that uses an internal evaluation index as a loss function, demonstrating its effectiveness especially on non-convex data shapes compared to traditional methods.

Contribution

The paper presents CNNI, a novel neural network clustering approach that incorporates an internal index as a loss function, capable of handling complex data geometries.

Findings

CNNI effectively clusters data with complex shapes.

CNNI with MMJ-SC outperforms traditional clustering models.

First parametric inductive clustering model for non-convex data.

Abstract

A new model called Clustering with Neural Network and Index (CNNI) is introduced. CNNI uses a Neural Network to cluster data points. Training of the Neural Network mimics supervised learning, with an internal clustering evaluation index acting as the loss function. An experiment is conducted to test the feasibility of the new model, and compared with results of other clustering models like K-means and Gaussian Mixture Model (GMM). The result shows CNNI can work properly for clustering data; CNNI equipped with MMJ-SC, achieves the first parametric (inductive) clustering model that can deal with non-convex shaped (non-flat geometry) data.