Browsing by Author "Kamaleswaran, Rishikesan"

Now showing 1 - 2 of 2

CBPsp: complex business processes for stream processing
(2011-04-01) Kamaleswaran, Rishikesan; McGregor, Carolyn
This thesis presents the framework of a complex business process driven event stream processing system to produce meaningful output with direct implications to the business objectives of an organization. This framework is demonstrated using a case study instantiating the management of a newborn infant with hypoglycaemia. Business processes defined within guidelines, are defined at build-time while critical knowledge found in the definition of business processes are used to support their enactment for stream analysis. Four major research contributions are delivered. The first contribution enables the definition and enactment of complex business processes in real-time. The second contribution supports the extraction of business process using knowledge found within the initial expression of the business process. The third contribution allows for the explicit use of temporal abstraction and stream analysis knowledge to support enactment in real-time. Finally, the last contribution is the real-time integration of heterogeneous streams based on Service-Oriented Architecture principles.
A dynamic visual analytics framework for complex temporal environments
(2016-08-01) Kamaleswaran, Rishikesan; McGregor, Carolyn
Introduction: Data streams are produced by sensors that sample an external system at a periodic interval. As the cost of developing sensors continues to fall, an increasing number of data stream acquisition systems have been deployed to take advantage of the volume and velocity of data streams. An overabundance of information in complex environments have been attributed to information overload, a state of exposure to overwhelming and excessive information. The use of visual analytics provides leverage over potential information overload challenges. Apart from automated online analysis, interactive visual tools provide significant leverage for human-driven trend analysis and pattern recognition. To facilitate analysis and knowledge discovery in the space of multidimensional big data, research is warranted for an online visual analytic framework that supports human-driven exploration and consumption of complex data streams. Method: A novel framework was developed called the temporal Tri-event parameter based Dynamic Visual Analytics (TDVA). The TDVA framework was instantiated in two case studies, namely, a case study involving a hypothesis generation scenario, and a second case study involving a cohort-based hypothesis testing scenario. Two evaluations were conducted for each case study involving expert participants. This framework is demonstrated in a neonatal intensive care unit case study. The hypothesis generation phase of the pipeline is conducted through a multidimensional and in-depth one subject study using PhysioEx, a novel visual analytic tool for physiologic data stream analysis. The cohort-based hypothesis testing component of the analytic pipeline is validated through CoRAD, a visual analytic tool for performing case-controlled studies. Results: The results of both evaluations show improved task performance, and subjective satisfaction with the use of PhysioEx and CoRAD. Results from the evaluation of PhysioEx reveals insight about current limitations for supporting single subject studies in complex environments, and areas for future research in that space. Results from CoRAD also support the need for additional research to explore complex multi-dimensional patterns across multiple observations. From an information systems approach, the efficacy and feasibility of the TDVA framework is demonstrated by the instantiation and evaluation of PhysioEx and CoRAD. Conclusion: This research, introduces the TDVA framework and provides results to validate the deployment of online dynamic visual analytics in complex environments. The TDVA framework was instantiated in two case studies derived from an environment where dynamic and complex data streams were available. The first instantiation enabled the end-user to rapidly extract information from complex data streams to conduct in-depth analysis. The second allowed the end-user to test emerging patterns across multiple observations. To both ends, this thesis provides knowledge that can be used to improve the visual analytic pipeline in dynamic and complex environments.