Partition and Hierarchical based Clustering Techniques for Analysis of Neonatal Data

Mago, Nikhit and Shirwaikar, Rudresh D and Acharya, Dinesh U and Hegde, Govardhan and Lewis, Leslie Edward Simon and Shivakumar, M (2017) Partition and Hierarchical based Clustering Techniques for Analysis of Neonatal Data. In: International Conference on Cognition and Recognition, 29/12/2016, Bangalore.

[img] PDF
springer-gh.pdf - Published Version
Restricted to Registered users only

Download (378kB) | Request a copy


With the increase of data in the medical domain over the years, it is extremely crucial that we analyze useful information and recognize patterns that can be used by the clinicians for better diagnosis of diseases. Clustering is a Machine Learning technique that can be used to categorize data into compact and dissimilar clusters to gain some meaningful insight. This paper uses partition and hierarchical based clustering techniques to cluster neonatal data into different clusters and identify the role of each cluster. Clustering discovers hidden knowledge which helps neonatologists in identifying neonates who are at risk and also helps in neonatal diagnosis. In addition, this paper also evaluates the number of clusters to be formed for the techniques using Silhouette Coefficient

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Partition based Clustering Hierarchical based Clustering. Silhouette Coefficient. Neonate. K-means.
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Depositing User: MIT Library
Date Deposited: 09 Dec 2017 10:17
Last Modified: 29 Dec 2017 08:04

Actions (login required)

View Item View Item