Another issue that the Siemens’ implementation brings to the forefront is the problem indealing with false positives and alarm floods, which affects all continuous auditing architectures Alles et al. 2006; Alles et al. 2008 . If analytical criteria are set too stringently there is the potential to flag a large number of transactions as potentially erroneous or fraudulent and the information overload for the auditor could be substantial. Leaving criteria too loosely bounded, on the other hand, could fail to detect a large percentage of erroneous or fraudulent transactions. Thus, methodologies for addressing the inherent information overload are needed to improve the viability of continuous auditing. For instance, Perols and Murthy’s 2009 work with information infusion may yield one such approach to mitigating the information load effects. Further exami- nation of information infusion capabilities, as well as other alternative approaches, seem worthy of research attention. More sophisticated analytics that apply artificial intelligence techniques or other mathematical algorithms may also improve the detection capability of continuous auditing systems and to reduce the overload of false positives. Work such as that by Jans et al. 2010 suggests that such refined capabilities are possible with advanced analytic techniques.
đang được dịch, vui lòng đợi..