Use the Genetic Algorithm to Optimize the Alpha Network of the RETE Algorithm
Daoqu Geng, Chang Liu, Xingchuan Lan, Hanwen He · 2021
The RETE algorithm greatly improves the efficiency of the production reasoning system by sharing rule conditions and saving temporary matching results, making it one of the most efficient production reasoning algorithms. However, with the increasing size of data, frequent changes of business information, and the emergence of incomplete data and fuzzy logic, the reasoning time and the storage consumption of the RETE algorithm in the reasoning process are increasing. Many studies are devoted to improving the RETE algorithm to reduce the inference time and the occupation of storage resources. In this article, we propose a method that using the genetic algorithm (GA) to optimize the Alpha network of the RETE algorithm. Experimental results show that genetic algorithm has an optimization effect on the rule matching system in reasoning time and memory consumption.