---
title: "Improving Search Using Indexing: A Study with Temporal CSPs"
authors:
  - "Nikos Mamoulis"
  - "Dimitris Papadias"
doi: ""
url: "https://ijcai.org/proceedings/1999/63"
pdf: "https://ijcai.org/proceedings/1999/0063.pdf"
venue: "Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence"
year: 1999
pages: "436-440"
track: "Constraint Satisfaction"
keywords:
---

# Improving Search Using Indexing: A Study with Temporal CSPs

Nikos Mamoulis, Dimitris Papadias

*Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence*, pages 436-440.

## Abstract

Most studies concerning constraint satisfaction problems (CSPs) involve variables that take v~l ues from small domains. This paper deals with an alternative form of temporal CSPs; the number of variables is relatively small and the domains are large collections of intervals. Such situations may arise in temporal databases where several types of queries can be modeled and processed as CSPs. For these problems, systematic CSP algorithms can take advantage of temporal indexing to accelerate search. Directed search versions of chronological backtracking and forward checking are presented and tested . Our results show that indexing can drastically improve search performance.