| In a Big Data environment, organizations and individuals benefit from increased prediction accuracy and real-time data analysis results. Currently, Big Data continues to increase in complexity, particularly in respect of the uncertainty of the environments in which decisions are made, and many business decisions need to be made under a range of data uncertainties, e.g., ambiguous data or unlabelled data. This issue results in the business decision-making process itself becoming more dynamic, since big data distributions change over time, and the change is irreducible. Decision-makers must react quickly to insights to take full advantage of such dynamic data environments. More importantly, many organizational decisions need to be made across multiple dimensions, multiple data sources and/or multiple data types, such as text data or image data, and sometimes in time-critical situations. The increasing developments in decision data presentation, decision data analytics, data-based knowledge discovery and data-driven decision support systems have simultaneously resulted in a challenging need to deal with issues of data uncertainty. Thus, to effectively handle data uncertainties and fully use Big Data in decision support systems, there is an urgent and powerful requirement to develop new methodologies and techniques. |
