Volume 7 Issue 2 - July 2016

  • 1. Application based, advantageous k-means algorithm

    Authors : Barkha Narang

    Pages : 121-126

    DOI : http://dx.doi.org/10.21172/1.72.520

    Keywords : data mining, k-means clustering

    Abstract :

    This paper has been written with the aim of giving a basic view on data mining. Various software’s of data mining analyzes relationships and patterns in stored transaction data based on the user requirements. Several types of analytical software are available to mine useful data like statistical, machine learning, and neural networks. The four types of relationships sought using the analytical software’s are classification, clustering, associations and finding patterns. In this paper we have discussed clustering and specifically K- means clustering technique. It is the most popular clustering technique with its own advantages and disadvantages. This paper focuses on the advantages in applications like market segmentation, in big data applications, real world problems like road detection, DNA gene expression and internet news group and elegant handling of continuous and noisy data. Different levels of analysis are available like genetic algorithms, artificial neural networks, decision trees, rule based induction methods and data visualization. K-means clustering has been integrated with these analytical tools as per the requirement of the application area.