Research of dynamic rule engine in financial management software
Bing Hao Xu, Shi-Yi Xie · 2008
It applies a rule engine design scheme with dynamic programming algorithm and optimal policy in financial management system. The scheme adopts backward reasoning method, when the facts and rules change in a small way, the results of constant parts are kept by the algorithm, and it just needs to calculate the changing parts. Thus, the massive repeated calculating workload can be reduced, which enhances system performance greatly. When the facts and rules have big changing, the algorithm still retains a high efficiency. The testing result shows that the average response time to system is far lower than both of forward reasoning and ordinary backward reasoning method in same circumstances.