Measuring inter-city connectivity in an urban agglomeration based on multi-source data

A comprehensive understanding of inter-city connectivity is important for regional planning. However, most studies adopted only one single data source for measurements, which is incomplete since each source has its own limitations. There are biases and uncertainties in the connectivity results when using different data sources. To address this problem, our study proposed a novel method that could combine the advantages of multi-source data. First, we measured inter-city connectivities using several datasets individually, and then analyzed each city's node strength based on the connectivities. Next, the performance of each dataset was validated according to several correlation analyses between the node strength and various socio-economic metrics. Based on these validations, we used the genetic algorithm to search for the optimal weights for combination. Only those datasets with higher weights were retained for further calculation. The final connectivity result is more reasonable than any single result according to the validation. For the first time, this study compares different data sources related to inter-city connectivity, and combines their advantages based on selective weighted combination. The results are expected to provide strong support for large-scale regional planning. In addition, the proposed method could be further applied to other large areas for analyzing inter-city connectivities.