Analyzing The Field Of Big Data - 954 Words

Literature review: To address the question of how and what techniques has been used to manages this big amount of data or in the field of Big Data, I review some research papers and review articles in the field of Big Data. This paper provides the synthesis of those papers which I found relevant to this field. This paper will focus on the following things: †¢ What are the technologies being used in Big data? †¢ Which technology is suitable for which type of data? †¢ Current trends in Big Data field. Fig: Big Data Sources 4.1 Survey Paper: A survey on data stream clustering and classification Authors: Hai-Long Nguyen , Yew-KwongWoon , Wee-KeongNg Published online: 17 December 2014 Purpose: This paper presents a inclusive survey of the†¦show more content†¦Therefore, to randomly access these datasets, which is commonly assumed in traditional data mining, is really expensive. Findings and Learning’s: 1) There are some useful, open source software for data stream mining research:. †¢ WEKA: WEKA is the most popular data mining software for the academic environment. WEKA contains the collection of learning algorithms such as data preprocessing, association rules , classification, regression, clustering, and information visualization. †¢ Massive Online Analysis (MOA): This is based on the WEKA framework that is build and designed for data stream learning. †¢ RapidMiner: RapidMiner is another importantopen source software for data mining. 2) Some important clustering algorithms discussed in this paper to group massive data and can be useful to industries and organization: †¢ Partitioning methods: This algorithm groups dataset into q clusters, where q is a predefined parameter. †¢ It continuously reassigns objects from one group to another group so as to r to minimize its objective function. †¢ Hierarchical methods: In the hierarchical method the aim is to group data objects into a hierarchical tree of clusters. Hierarchical clustering methods can be further classified as either agglomerative or divisive, where the hierarchical decomposition is formed in a bottom up(merging) or top down(splitting) fashion respectively. †¢ Density based methods: Under this method we build up the