The concept of context awareness has been in existence since the 1990s. Though initially applied exclusively in computer science, over time it has increasingly been adopted by many different application domains such as business, health and military. Contexts change continuously because of objective reasons, such as economic situation, political matter and social issues. The adoption of big data analytics by businesses is facilitating such change at an even faster rate in much complicated ways. The potential benefits of embedding contextual information into an application are already evidenced by the improved outcomes of the existing context-aware methods in those applications. Since big data is growing very rapidly, context awareness in big data analytics has become more important and timely because of its proven efficiency in big data understanding and preparation, contributing to extracting the more and accurate value of big data. Many surveys have been published on context-based methods such as context modelling and reasoning, workflow adaptations, computational intelligence techniques and mobile ubiquitous systems. However, to our knowledge, no survey of context-aware methods on big data analytics for business applications supported by enterprise level software has been published to date. To bridge this research gap, in this paper first, we present a definition of context, its modelling and evaluation techniques, and highlight the importance of contextual information for big data analytics. Second, the works in three key business application areas that are context-aware and/or exploit big data analytics have been thoroughly reviewed. Finally, the paper concludes by highlighting a number of contemporary research challenges, including issues concerning modelling, managing and applying business contexts to big data analytics.

中文翻译：

---

针对业务应用程序的大数据分析中的上下文意识调查

自1990年代以来一直存在情境意识的概念。尽管最初仅在计算机科学中应用，但随着时间的流逝，它逐渐被许多不同的应用领域所采用，例如商业，卫生和军事。由于客观原因，例如经济形势，政治问题和社会问题，背景不断变化。企业采用大数据分析正在以非常复杂的方式以更快的速度促进这种变化。将上下文信息嵌入到应用程序中的潜在好处已经被那些应用程序中现有的上下文感知方法的改进结果所证明。由于大数据增长非常迅速，大数据分析中的上下文感知已变得更加重要和及时，这是因为其在大数据理解和准备方面的行之有效的效率，有助于提取大数据的准确性。关于基于上下文的方法，例如上下文建模和推理，工作流适应，计算智能技术和移动无处不在的系统，已经发表了许多调查。但是，据我们所知，迄今为止，尚未发布有关企业级软件支持的业务应用程序的大数据分析的上下文感知方法的调查。为了弥合这一研究差距，本文首先介绍了上下文的定义，其建模和评估技术，并强调了上下文信息对于大数据分析的重要性。第二，已经彻底审查了上下文感知和/或利用大数据分析的三个关键业务应用领域中的工作。最后，本文通过重点介绍了许多当代研究挑战，包括与建模，管理业务环境并将其应用于大数据分析的问题。