slugStudy of Euclidean and Manhattan Distance Metrics using Simple K-Means Clustering
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2014, Vol 2, Issue 5
Abstract
Clustering is the task of assigning a set of objects into groups called clusters in which objects in the same cluster are more similar to each other than to those in other clusters. Generally clustering is used to find out the similar, dissimilar and outlier items from the databases. The main idea behind the clustering is the distance between the data items. The work carried out in this paper is based on the study of two popular distance metrics viz. Euclidean and Manhattan. A series of experiments has been performed to validate the study. We use two real and one synthetic datasets on simple K-Means clustering. The theoretical analysis and experimental results show that the Euclidean method outperforms Manhattan method in terms of number of iterations performed during centroid calculation.
Authors and Affiliations
Deepak Sinwar, Rahul Kaushik
Keywords
Related Articles
- EP ID EP18115
- DOI -
- Views 264
- Downloads 13
slugDetecting Phishing Websites Based on Visual Cryptography
Most of the time people using internet for online transactions from one place to another place. Th e security in this case should be very high and should not be easily traceable to attacker....
Performance Analysis of Flat Plate Solar Water Collector using Trapezoidal shape and Semicircular Tubes
The flat plate collector forms the heart of any solar energy collection systems designed for operation in the low temperature range, from ambient to 60oc, or the medium temperature range, from ambient to 100oc. A well e...
Android Based Women Tracking System Using GPS and GSM
n recent times, role of women in all industries have increased. Simultaneously, crime against women is also increasing day by day and this is the high time to provide safety for all working women. This paper focuses on...
Dielectric Studies of Ion Induced PVDC (Poly vinylidene chloride)
Dielectric properties of ion induced PVDC (Poly vinylidene chloride) were analyzed. Lithium (50 MeV), carbon (85 MeV) and nickel (120 MeV) ion beams were used to analyze the modifications induced by swift heavy ions as...
Simulation of Hybrid Amplifiers(EDFA and Raman Amplifier) on 40GBPS on 8 and 16 Users WDM System
To fulfill the need of higher bandwidth, enhanced number of users, enhanced speed of communication of today’s users, the concept of WDM and fiber optic was developed. WDM is a wavelength division multiplexing. In which...