Analyzing public participant data to evaluate citizen satisfaction and to prioritize their needs via K-means, FCM and ICA

In this article, information from Municipal District 1 in the city of Bojnourd was gathered, analyzed to prioritize urban needs and assessed for citizen satisfaction. Forty-three citizen needs were identified and categorized based on K-means clustering, the Fuzzy Clustering Method (FCM) and the Imperialist Competitive Algorithm (ICA). The three algorithms were also evaluated. RFM (recency, frequency, monetary) analysis was performed for classification. The clustering methods were then assessed and compared using three parameters: execution time, accuracy and simplicity. The results of the FCM and ICA clustering were similar, however, the execution time for FCM was less than for ICA. Considering the similarity of the results and the flexibility of FCM, it was concluded that, if the execution time was of primary importance, then the use of FCM was more appropriate. In contrast, if accuracy was a priority, ICA was preferred. Our results also showed that if simplicity and speed were required, the K-means algorithm was the best choice. Finally, subjects such as the quality of the asphalt, garbage collection and park development were of primary importance to Bojnourd citizens, therefore the municipality should pay special attention to these subjects.