H. H. Aung et al., "A∗DAX-DAT: A toolkit framework for big data analytics," TENCON 2017 - 2017 IEEE Region 10 Conference, Penang, Malaysia, 2017, pp. 639-644. doi: 10.1109/TENCON.2017.8227940
Abstract:
A*STAR Data Analytics Exchange ("A*DAX") is a scalable platform built upon open architecture and it is designed to facilitate the secure management and analytics of urban data. In this paper, we introduced A*DAX-DAT, which is the latest development of A*DAX. A*DAX-DAT (or DAT for short) is the analytics component of the A*DAX Platform. DAT aims to provide advance analytics functionalities to A*DAX in such a way that: (i) it is extensible, (ii) it is easy to use for both end-users and developers and (iii) it is efficient in handling data regardless of variety, volume and velocity. DAT consists of three components: (a) a set of framework specifications for analytics libraries, (b) dynamic analytics libraries following the aforementioned framework specifications and (c) tools to develop/use the dynamic libraries either in standalone mode or in client/server mode. This paper presents the design of DAT components with implementation details before it discusses sample use-cases of DAT. Finally, it concludes with discussions on future works on DAT.