Analysis of taxi dataset to categorize city rankings

Abstract

The city administration values the district attraction rating since it can aid in extracting the desirability of the location and hence support the officials in making smart city development decisions. Traditional urban planning tactics mostly rely on Gross Domestic Product, rate of employment, number of people per unit area, and district statistics gleaned through surveys and questionnaires, among other factors. As a point of reference, such knowledge becomes less and less helpful over time. The volume of urban data is growing at an exponential rate. Furthermore, these tactics suffer from a fatal flaw: they are unsuccessful. Independent representations of a district's appeal, as well as inter-district interactions, are not taken into account. It is now feasible to use urban data efficiently for urban planning thanks to advances in urban computing. To that end, this paper proposes PageRank, a district attractiveness rating algorithm based on taxi big data, which is the first to do so. A working software is constructed for visualization motives. To begin, the total area is split into numerous parts using the k-means algorithm.