Central Concerns and Questions

Traditional database systems are designed to answer transient, snapshot queries over persistent data. However, the evolution of wireless communications, positioning devices (e.g., GPS) and sensor technologies has recently given rise to a new data processing model. In this model, multiple long-running queries require continuous evaluation as the data dynamically change. These queries are called continuous, and their efficient processing is central in many emerging applications.

Emerging Ideas and Initiatives

We study continuous queries that span several domains. One of them is location-based services, where the objective is to design systems that monitor spatial relationships [1, 2, 3]. As an example, consider a system that receives the locations of taxis and of clients asking for service, and continuously reports to (the driver of) each free taxi who/where are its 10 closest clients as both the taxis and the clients move frequently and unpredictably. The main challenge here is that results need to be reported/updated as fast as possible. The dynamic and time-critical nature of the problem necessitates the design of efficient processing algorithms.

Another application domain is online decision support systems [4]. Consider a server which receives a continuous feed with stock market data, including their price and trend indicators. Users connected to the server specify their investment criteria, expressed as functions (over the price and trend indicators) that score each available stock. While prices and indicators change continuously, the server needs to report to each user the 20 stocks that score highest according to his/her criteria. Again, there is a need for fast processing in order to facilitate timely user decisions.

A third domain where continuous queries are important is text filtering applications. For instance, a security analyst who monitors email traffic for potential terror threats would register several continuous text queries to identify recent emails that most closely fit certain threat profiles (e.g., emails that mention names of explosives or possible biological weapons). As another example, an investment manager who is interested in a portfolio of industries and companies would monitor newsflashes from his information provider (e.g., Reuters, Bloomberg, etc.) to identify those that are relevant to his portfolio. Words related to the industries of interest can be formulated as continuous text queries over the newsflashes.

Selected Publications

[1] Kyriakos Mouratidis, Marios Hadjieleftheriou, Dimitris Papadias. Conceptual Partitioning: An Efficient Method for Continuous Nearest Neighbor Monitoring. ACM Conference on Management of Data (SIGMOD), 2005.

[2] Kyriakos Mouratidis, Dimitris Papadias, Spiridon Bakiras, Yufei Tao. A Threshold-based Algorithm for Continuous Monitoring of k Nearest Neighbors. IEEE Transactions on Knowledge and Data Engineering (TKDE), 17(11), 1451-1464, 2005.

[3] Kyriakos Mouratidis, Man Lung Yiu, Dimitris Papadias, Nikos Mamoulis. Continuous Nearest Neighbor Monitoring in Road Networks. Very Large Data Bases Conference (VLDB), 2006.

[4] Kyriakos Mouratidis, Spiridon Bakiras, Dimitris Papadias. Continuous Monitoring of Top - k Queries over Sliding Windows. ACM Conference on Management of Data (SIGMOD), 2006.

[5] Kyriakos Mouratidis, HweeHwa Pang. An Incremental Threshold Method for Continuous Text Search Queries. IEEE International Conference on Data Engineering (ICDE), short paper, 2009.

[6] Leong Hou U, Kyriakos Mouratidis, Nikos Mamoulis. Continuous Spatial Assignment of Moving Users. Very Large Data Bases Journal (VLDBJ), 19(2), 141-160, 2010.

Projects, Presentations and Posters

1. Mouratidis, K. Continuous Monitoring of Spatial Queries. Encyclopedia of Database Systems, Ling Liu and M. Tamer Özsu (editors), Springer, 2008.
2. Mouratidis, K. Continuous Spatial Queries. Lecture notes for "IS410: Advanced Data Management".
3. Mouratidis, K. Continuous Monitoring of Top-k Queries over Sliding Windows. Presentation slides of [4] at SIGMOD 2006.
4. Mouratidis, K. An Incremental Threshold Method for Continuous Text Search Queries. Poster of [5] at ICDE 2009.
5. Mouratidis K. Continuous Spatial Assignment. Presentations slides of [6].

External Collaborations and Industry Linkages

1. Research Mentor: Christos FALOUTSOS, Professor, Carnegie Mellon University
2. Collaborator: Nikos MAMOULIS, Associate Professor, Department of Computer Science, University of Hong Kong
3. Visitor/Collaborator: Man Lung YIU, Assistant Professor at Department of Computer Science, Aalborg University