Data Dissemination Techniques using DBSCAN and DD-Rtree for Spatial Data Mining
Basavaraj S. Prabha1, Arun Biradar2
1Basavaraj S. Prabha, Ph.D Research Scholar, CSE Department, CMR University, Bengaluru, India.
2Dr. Arun Biradar, Professor & Head, Dept. of CSE, CMR University, Bengaluru, India.
Manuscript received on January 05, 2020. | Revised Manuscript received on January 25, 2020. | Manuscript published on January 30, 2020. | PP: 4465-4470 | Volume-8 Issue-5, January 2020. | Retrieval Number: E6578018520/2020©BEIESP | DOI: 10.35940/ijrte.E6578.018520
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In today’s scenario where data volumes are growing on enormous speed over cloud or internet, we want to limit this growing data size. This can be achieved by data processing methods where data processing can be done in parallel. To make the data processing done in parallel, various clustering sampling methodologies are in use such as Slink, DBSCAN, and Optics and so on. The power accomplished by various methodologies which already exist will be focusing to the preservation of three-dimensional surroundings such as grid tree, grid files, quad tree and tree like k-d-tree, etc. This all compartmentalization constructions are generally done in static way which is a fix way. Since this data volume size is very big, this results in a high cost of information sharing and clustering. Hence through this research work we want to analyze various clustering algorithms both on static level and at dynamic level. For doing this we are majorly comparing the dynamic distribution using DBSCAN and DD-Rtree algorithm by proposing a DD-Rtree will help us to preserver the spatial vicinity. In addition, DD-Rtree is not static but more than that it is dynamic, i.e. it will create build the data as we progress with clustering. DD-Rtree methodologies are based on R-Tree concepts which analyses the data at dynamic random way. We tend to compare DD-RTree’s information distribution norm with one of the clustering system recently published, DBSCAN. On the side of the potential of DBSCAN formula, we tend to distinguish the potential of queries managed by these compartmentalization structures. Numerous applications requires such kind of implementation at dynamic level of spatial database system such as satellite images, X-Ray crystallography, metrological department or other such atomic equipment’s spatial datasets. Our research work will help to implements spatial data dynamically using DDR-tree mechanism.
Keywords: Data Dissemination, KNN, Spatial Data Mining, Density Clustering.
Scope of the Article: Data Mining.