Clustering Methods Based on Stay Points and Grid Density for Hotspot Detection
Bibliographic Data
| ID | 22033698 |
|---|---|
| Authors | Xiaohan Wang (0000-0002-3978-1946, Anhui Institute of Information Technology), Zepei Zhang (Anhui Institute of Information Technology), Yonglong Luo (0000-0003-4987-0376, Anhui Institute of Information Technology, corresponding author) |
| Year | 2022 |
| Volume | 11 |
| Issue | 3 |
| Pages | 190 |
| Publication date | 2022-03-11 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | ISPRS International Journal of Geo-Information (JOURNAL) |
| Journal identifiers | ISSN: 2220-9964 • E-ISSN: 2220-9964 |
| Publisher | MDPI AG (PUBLISHER • IT) |
| DOI | 10.3390/ijgi11030190 |
| OpenAlex | W4221058158 |
| Language | EN |
| Citations received | 3 |
| References cited | 31 |
With the widespread use of GPS equipment, a large amount of mobile location data is recorded, and urban hotspot areas extracted from GPS data can be applied to location-based services, such as tourist recommendations and point of interest positioning. It can also provide decision support for the analysis of population migration distribution and land use and planning. However, taxi GPS location data has a large amount of data and sparse points. How to avoid the influence of noise and efficiently detect hotspots in cities have become urgent problems to be solved. This paper proposes a clustering algorithm based on stay points and grid density. Firstly, a filtering pre-processing algorithm using stay points classification and stay points thresholds is proposed, so the influence of stop points is avoided. Then, the data space is divided into rectangular grid cells; each grid cell is determined to be a dense or non-dense grid according to the defined density threshold, and the cluster boundary points and noise points are judged in the non-dense grid cells to avoid normal sampling points being treated as noise. Finally, the associated dense grids are connected into clusters. The sampling points mapped to the grid cells are the elements in the clusters. Our method is more efficient than the DBSCAN algorithm because the grid cells are calculated. The superiority of the proposed algorithm in terms of clustering accuracy and time efficiency is verified in the real data set compared to traditional algorithms
Algorithm · Cluster analysis · Correlation clustering · CURE data clustering algorithm · Data mining · DBSCAN · Geodesy · Geography · Global Positioning System · Grid · k-medians clustering · Point of interest · Population · Computer Science · Data Management and Algorithms · Geographic Information Systems Studies · Human Mobility and Location-Based Analysis · Artificial Intelligence
| Unique citing works | 3 |
|---|---|
| Citations per year | 1 |
| Citation span | 2023 - 2025 (3) |
| Citation velocity | recent |
| Highly cited | No |
| Citation types | Neutral: 3 |