Similarity Calculation of Environmental Complaint Events Based on Improved FOA

Abstract

For the treatment of environmental complaint information, it is essential to judge the similarity between current report events and historical report cases. However, complaint events are complicated, so it is hard to calculate the similarity between them. In general, the event similarity is usually a comprehensive result based on the similarity of constituent elements. Therefore, an event similarity calculation method based on an improved fruit fly optimization algorithm (IFOA) is proposed in this paper to solve the problem of environmental complaint events similarity computation. Firstly, the fruit fly optimization algorithm (FOA) is modified to solve the issues of fixed search radius and low population diversity. Secondly, the similarity degree set is constructed by calculating the similarity degree of each component between two events, which is taken as the sample data. Then, the generalized regression neural network (GRNN) is applied to establish the similarity calculation model. Finally, the parameter of the model is optimized by IFOA. Experimental results show that this method has high accuracy and can meet the actual demand.