A Framework for Uncertainty-Aware Visual Analytics in Big Data

Visual analytics has become an important tool for gaining insight on big data. Numerous statistical tools have been integrated with visualization to help analysts understand big data better and faster. However, data is inherently uncertain, due to sampling error, noise, latency, approximate measurement or unreliable sources. It is very important and vital to quantify and visualize uncertainties for analysts to improve the results of decision making process and gain valuable insights during analytic process on big data. In this paper, we propose a new framework to support uncertainty in the visual analytics process through a fuzzy self-organizing map algorithm running in MapReduce framework for parallel computations on massive amounts of data. This framework uses an interactive data mining module, uncertainty modeling and knowledge representation that supports insertion of the user’s experience and knowledge for uncertainty modeling and visualization in the big data.

• Type: Journal
• Year: 2015
• Where: AIC
• Authors: Amin Karami
• Keywords: AIC 2015