Sashakt Pathak, Arushi Agarwal, Ankita Ankita, M. Gurve
{"title":"Restricted Randomness DBSCAN : A faster DBSCAN Algorithm","authors":"Sashakt Pathak, Arushi Agarwal, Ankita Ankita, M. Gurve","doi":"10.1145/3474124.3474204","DOIUrl":null,"url":null,"abstract":"Data Mining is the process of extracting useful and accurate information or patterns from large databases using different algorithms and methods of machine learning. To analyze the data, Clustering is one of the methods in which similar data is grouped together and DBSCAN clustering algorithm is the one, which is broadly used in numerous practical applications. This paper presents a more efficient density based clustering algorithm, which has the ability to discover cluster faster than the existing DBSCAN algorithm. The efficiency is achieved by restricting the randomness of choosing points from the dataset. Our proposed algorithm named Restricted Randomness DBSCAN (RR DBSCAN) is compared with conventional DBSCAN algorithm over 9 datasets on the basis of Silhouette Coefficient, Time taken in formation of clusters and accuracy. The results show that RR DBSCAN performs better than traditional DBSCAN in terms of accuracy and time taken to form clusters.","PeriodicalId":144611,"journal":{"name":"2021 Thirteenth International Conference on Contemporary Computing (IC3-2021)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 Thirteenth International Conference on Contemporary Computing (IC3-2021)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3474124.3474204","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Data Mining is the process of extracting useful and accurate information or patterns from large databases using different algorithms and methods of machine learning. To analyze the data, Clustering is one of the methods in which similar data is grouped together and DBSCAN clustering algorithm is the one, which is broadly used in numerous practical applications. This paper presents a more efficient density based clustering algorithm, which has the ability to discover cluster faster than the existing DBSCAN algorithm. The efficiency is achieved by restricting the randomness of choosing points from the dataset. Our proposed algorithm named Restricted Randomness DBSCAN (RR DBSCAN) is compared with conventional DBSCAN algorithm over 9 datasets on the basis of Silhouette Coefficient, Time taken in formation of clusters and accuracy. The results show that RR DBSCAN performs better than traditional DBSCAN in terms of accuracy and time taken to form clusters.