Big Data Analysis with Interactive Visualization using R packages

Abstract

Compared to the traditional data storing, processing, analyzing and visualization which have been performed, Big data requires evolutionary technologies of massive data processing on distributed and parallel systems, such as Hadoop system. Big data analytic systems, thus, have been popular to derive important decision making in various areas. However, visualization on analytic system faces various limitation due to the huge amount of data. This brings the necessity of interactive visualization techniques beyond the traditional static visualization. R has been used and improved for a big data analysis and mining tool. Also, R is supported with various and abundant packages for different targets with visualization. However interactive visualization packages are not easily found in the market. This paper compares and analyzes interactive web packages with visualization packages for R. This paper also proposes interactive web visualized analysis environment for big data with a combination of interactive web packages and visualization packages. In particular, Big data analysis techniques with sensed data are presented as the result by reflecting the decision view on sensing field.

Requirements of Data Visualisation Tools to Analyse Big Data: A Structured Literature Review

“…By providing a visual interface which the user can interact with, the user is able to intuitively comprehend the data, perceive the underlying patterns (Lugmayr et al 2017) and query the data visually, without the need for programming knowledge. The ability for a user to interact with a visualisation makes visual exploration of the dataset possible and this has the major advantage of combining both human and machine intelligence (Chen et al 2019) to uncover unexpected and interesting phenomena within the dataset (Cho et al 2014). The benefits of visual analytics are "visual perception, interactive exploration, improved understanding, informed steering and intuitive interpretation" (Liu 2019).…”

ROSY Application for Selecting R Packages that Perform ROC Analysis

“…Shiny aims to create applications to either be run through a web browser or locally in an R environment. While this tool is mostly aimed at statisticians and data analysts as a way to share their findings, it is mostly available to all areas thanks to its basis on R, making it far more accessible [7].…”

Real time data analysis and visualization for the breast cancer disease

“…Exploring data using different graphs enables humans' eyes to easily detect meaningful information, understand large datasets, identify meaningful patterns within data, recognize outliers and form hypothesis efficiently Unlike traditional data that are well organized and structured, big data such as streaming data, images, documents, videos and music are complex, unstructured and unorganized thus required advanced visualization techniques rather than the traditional static ones. Since that, traditional techniques for presenting the tremendous growing amount of data have some limitations, dynamic data visualization became a significant alternative to effectively communicate and display datasets in an accessible manner not only for experts but also for non-experts, who can perform complex statistical analysis using a "pointand-click interface", as well (Ellis and Merdian, 2015)(Agrawal et al, 2015) (Cho et al, 2014). Wang et al (2015), stated four primary steps for interactive data visualization: 1) Selecting: interactively choosing a subset of the whole dataset according to the user's interests.…”