Title: Fuzzy Neighborhood Grid-Based DBSCAN Using Representative Points

Year of Publication: Jul - 2016
Page Numbers: 63-73
Authors: Oguz Altun, Abdallah Mekky
Conference Name: The Third International Conference on Data Mining, Internet Computing, and Big Data (BigData2016)
- Turkey

Abstract:


Clustering process is considered as one of the most important part in data mining, and it passes through many levels of developments. One of the most famous algorithm is Density-Based Spatial Clustering of Application with Noise (DBSCAN) [1,2,4]. It is a density-based clustering algorithm that uses a crisp neighborhood function to calculate the neighbor sets, and basically depends on distance function. In fuzzy clustering [9], which is considered as a soft clustering algorithm, it uses a fuzzy neighborhood function that allow a node in the dataset to have a membership degree in each point in the dataset. In this paper we propose a new algorithm that depends on both bases the speed of DBSCAN and the accuracy of fuzzy clustering. FNGMDBSCAN-UR is a Fuzzy Neighborhood Gridbased Multi-density DBSCAN Using Representative points. That uses grid-based to separate the dataset into small nets and fuzzy neighborhood function to create neighborhood sets.it is noticeable that FNGMDBSCAN-UR is much accurate than crisp DBSCAN with nested shapes and multi-dense datasets as we will see in the result section in this paper.