International Journal of Computer Networks and Applications (IJCNA) DOI: 10.22247/ijcna/2017/49227 Volume 4, Issue 5, September – October (2017) ISSN: 2395-0455 ©EverScience Publications 129 REVIEW ARTICLE A Framework for Effective Big data Analytics for Decision Support Systems Osama Islam 1 , Ahmed Alfakeeh 2 , Farrukh Nadeem 3 1, 2, 3 Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia. 1 oislam@stu.kau.edu.sa, 2 asalfakeeh@kau.edu.sa, 3 fabdullatif@kau.edu.sa Published online: 02 November 2017 Abstract – Supporting decision makers requires a good understanding of the various elements that affect the outcomes of a decision. Decision Support Systems have provided decision makers with such insights throughout its history of usage with varying degrees of success. The availability of data sources was a main limitation to what decision support systems can do. Therefore, with the advent of improved analytical methods for Big data sources new opportunities have emerged that can possibly enhance how decision makers analyze their problem and arrive at decisions using information systems. This paper analyzed current related works on both Big data and decision support systems to identify clear elements and factors relevant to the subject and identifying possible ways to enhance their joint usage. Finally, the paper proposes a framework that integrates the key components needed to ensure the quality and relevance of data being analyzed by decision support systems while providing the benefits of insights generated over time from past decisions and positive recommendations. Index Terms – Big data, Big data Analytics, Decision Support Systems, Information Systems. 1. INTRODUCTION The recent raise of large online data sources, commonly referred to as Big data, has presented new opportunities to improve decision making processes using advanced Decision Support Systems (DSS). However, to be able to properly examine the potential that Big data has on DSS a proper understanding of both DSS and the analysis of Big data is important, which is detailed in sections 3 and 4 respectively. In general, DSS started as a research topic in the latter half of the past century. Then over the years DSS toke a more prominent role within organizations in supporting decisions and analyzing data [1]. At the same time, DSS as an information system had a wide range of types and variations [2] that depended on its use case and expected benefits. Data- driven DSS [3] with its focus on analyzing data has taken more attention in recent years due to the increasing interest in processing very large datasets. These complex datasets, which are very large in size and not ease to process for further analysis using traditional data storage and manipulation techniques are commonly referred to as Big data [4]. Big data is defined by variety, volume, and velocity or as the 3 Vs, while some have added more Vs such as Veracity, Variability and Value for example. Big data alone has no value, which is why Big data analytics as a sub-field has taken the attention of the academic and commercial sectors, who all strive to obtain value from Big data [5]. The main aim of this analytics approach is to deal with processing very large sets of data input, the limited timeframe for processing streams of data, and the various data types and formats that must be analyzed. The use of Big data with DSS also faces some key issues, such as the limited availability of expert personnel in this new field, the high costs of the underlying technologies as they are still in the emerging stage, and the difficulty in customizing these new systems according to unique requires without major software development projects. However, some research areas are also exploring potential solutions for the challenges of Big data [6], while others propose future research into Big data and DSS [7]. This paper is structured as follows. Section 2 gives a general review of other work related to the research topic, while section 3 discusses what DSS is and how Big data environments have presented opportunities for better decisions using data-driven DSS. Section 4 then analyses Big data analytics and its potential uses by decision makers. After that Section 5 presents relevant best practices and trends in Big data analytics for use with DSS. Section 6 then details the proposed framework highlighting key components. Finally, we conclude and describe the future directions in section 7. 2. RELATED WORK As a research topic both DSS and Big data have a wide range of related research work, some have looked at how Big data can play an effective role in the decision support process, while others have looked at more specific factors, models and algorithms that utilize Big data analytics in decision making. For example, the research paper by [8], presented a theoretical examination of organizational and technical elements within the process of decision making, by exploring the interrelated relationships between Big data and business intelligence within the context of decision making. They discussed the potential