DSPI : Spatial Learned Index Leveraging Data‐Aware Pivots
- 26 March 2026
- journal article
- Published by Wiley in Transactions in GIS
- Vol. 30 (2)
Abstract
Spatial learned indexes construct index structures by learning the spatial distribution characteristics of data. Compared to traditional spatial indexes, they significantly reduce storage overhead and query costs. A widely adopted strategy in learned spatial indexes involves employing spatial pivots to model the global distribution of spatial data. Each data point is indexed by its distances to these pivots to enable efficient data partitioning and fast query processing. However, most existing methods rely on overly simplistic assumptions when selecting or modeling pivots. These assumptions limit their effectiveness when dealing with complex or heterogeneous spatial data. To address this, we propose a Distribution‐Aware Spatial Pivot Learned Index (DSPI). DSPI leverages a self‐attention mechanism to model the spatial distribution of data. It employs a Transformer neural network to jointly optimize pivot placement and pruning efficiency, enabling end‐to‐end query pruning. Within a given pivot search space, DSPI accurately locates query results using a lightweight function mapping model, optimized via second‐order derivatives and Akaike Information Criterion (AIC). Based on DSPI, we further design a range query algorithm with encoding‐based pruning and a k‐nearest neighbor (kNN) query algorithm with adaptive search radius. Experimental results on real‐world and synthetic datasets show that DSPI achieves state‐of‐the‐art efficiency: its index size is about 66.7% of LIMS and 71.4% of TBSI, while average query latency is about 2.5× faster than LIMS and 1.59× faster than TBSI. Under high‐dimensional data and large‐scale spatial queries, these advantages further widen. Moreover, DSPI maintains stable retrieval performance under dynamic updates.Keywords
Funding Information
- National Natural Science Foundation of China (42271466)
This publication has 20 references indexed in Scilit:
- TBSI: a Transformer-based spatial learned index for efficient construction and queryInternational Journal of Geographical Information Science, 2025
- $\text {SPL}^{{index}}$: A Spatial Polygon Learned Index2024
- A Learned Index for Exact Similarity Search in Metric SpacesIEEE Transactions on Knowledge and Data Engineering, 2022
- The “AI + R” - tree: An Instance-optimized R - tree2022
- Pivot selection algorithms in metric spaces: a survey and experimental studyThe VLDB Journal, 2021
- Learned Index for Spatial Queries2019
- Learning to Index for Nearest Neighbor SearchIEEE Transactions on Pattern Analysis and Machine Intelligence, 2019
- The Case for Learned Index Structures2018
- iDistanceACM Transactions on Database Systems, 2005
- Information Theory and an Extension of the Maximum Likelihood Principle1998