Business Rule Management (BRM) is a means to make decision-making within organizations explicit and manageable. BRM functions within the context of an Enterprise Architecture (EA). The aim of EA is to enable the organization to achieve its strategic goals. Ideally, BRM and EA should be well aligned. This paper explores through study of case study documentation the BRM design choices that relate to EA and hence might influence the organizations ability to achieve a digital business strategy. We translate this exploration into five propositions relating BRM design choices to EA characteristics.
From the article: Abstract Business rules (BR’s) play a critical role in an organization’s daily activities. With the increased use of BR (solutions) and ever increasing change frequency of BR’s the interest in modifiability guidelines that address the manageability of BR’s has increased as well. A method of approach to improve manageability and modifiability is to utilize architectures to structure BR’s. In current literature three different methods to structure business rules can be identified: 1) the rule family-oriented approach, 2) the fact-oriented approach and, 3) the decision-oriented approach. Scientific research comparing the ability to modify business rules in each of the three architectural candidates is limited. The goal of this research is to evaluate which architectural candidate and underlying architectural structures allow for the best modifiability. We sought to do so by applying design science research for the creation of the architectural candidates and by conducting semi-structured interviews to identify the case-specific productivity scores. By applying an Architecture-Level Modifiability Analysis using eight years of historical data from the British National Health Service each architectural candidate is evaluated with regards to its modifiability. Results of the analysis reveal that the rule family-oriented architecture scores best on modifiability, followed by the fact-oriented architecture, and lastly the decision-oriented architecture. The results of this study provide a foundation for further research on the application and evaluation of business rule architectures.
LINK
We present a novel architecture for an AI system that allows a priori knowledge to combine with deep learning. In traditional neural networks, all available data is pooled at the input layer. Our alternative neural network is constructed so that partial representations (invariants) are learned in the intermediate layers, which can then be combined with a priori knowledge or with other predictive analyses of the same data. This leads to smaller training datasets due to more efficient learning. In addition, because this architecture allows inclusion of a priori knowledge and interpretable predictive models, the interpretability of the entire system increases while the data can still be used in a black box neural network. Our system makes use of networks of neurons rather than single neurons to enable the representation of approximations (invariants) of the output.
LINK
