Partner Organizations:

Other collaborators:

We have collaborated with Prof. V. Kalogeraki (UC Riverside), Prof. S.
Lonardi (UC Riverside), Prof. A. Delis (Polytechnic Univ.), Prof.
Kollios (Boston Univ.), Prof. Sudipto Guha (U.Penn.), and with
industrial labs, more specifically with Dr. N. Koudas from AT&T
Research, and Dr. E. Hoel from ESRI.

Activities and findings:

Research and Education Activities: 

Spatiotemporal data appears in many real-life applications
(global change, surveillance, transportation etc.) Together
with regular attributes such data contains topological as well
as temporal attributes. This combination creates novel interesting
problems. Moreover, spatiotemporal data is usually presented
in 'streams' which drastically affects the data processing methods.
We propose general exploratory techniques that will allow the user
not only to verify specific hyphotheses, but more importantly,
to understand the underlying process that controls the changes
recorded in the spatiotemporal datasets. In particular:

1. We will first address the problem of performing on-line analysis
on spatiotemporal streams. This part of the project will provide
the appropriate tools for the major effort of this proposal,
i.e., understanding patterns of change.

2. We will then address the problem of understanding patterns
of change in specific spatiotemporal applications, namely:
(i) epidemiological, (ii) environmental and (iii) surveillance
applications. In each application we will first define low-level
analysis primitives. Such primitives are simpler to identify in
each application area. We will then allow combinations of such
promitives in order to describe complex (high-level) patterns of
change.

Findings:

(i) In [SIGKDD 2003] we have presented new techniques for online
analysis of multi-dimensional dta streams.
(ii) In [IPDPS 2003] and [CIKM 2002] we initiated a research study on
improving the performance of peer-to-peer systems using data anlysis
techniques and taking into account the physical and the virtual
topology of the P2P network.
(iii) In [ISJ 2005] and in [CISE 2004] we have extended our techniques
for improving the performance of P2P systems and gave novel apporaches
for building trully decentralized keyword search engines on top of
flexible and dymanic P2P networks. 
(iv) In [ISJ 2003] we present techniques for computing temporal
aggregates over streams.
(v) In [SSTD 2003] we present techniques for on-line analysis of
spatiotemporal datasets for discovering regions of interest to the
user.
(vi) In [EDBT 2004] we present any-time techniques for clustering time
series.
(vii) In [VLDB 2004] we present new indexing techniques for indexing
trajectories representing human motion.
(viii) In [VLDB 2003] we present techniques for efficient answering of
optimization queries over parametric aggregation constraits.
(ix) In [SSDBM 2003] we present a thorough comparison of selectivity
estimation techniques for spatiotemporal data.
(x) In [SSDBM 2004a] and [VLDBJ 2005] we present techniques for
indexing and mining spatiotemporal data.
(xi) In [SSDBM 2004b]we describe how to support spatiotemporal data
services in a large-scale environment.
(xii) In [TIME 2005] and [DCC 2005] we give compression techniques for
data streams, and we apply these techniques to sensor networks.

Training and Development:

Demetrios Zeinalipour-Yiazti earned his MS. degree in 2003, and his
Ph.D. degree in 2005. 
Marios Hadjieleftheriou earned his Ph.D. degree in 2004.

Outreach Activities:

The findings of this project will be reported in a web page:
www.cs.ucr.edu/~dg/change.html

Journal Publications:

M. Hadjieleftheriou, G. Kollios, D. Gunopulos, V.J. Tsotras, "On-Line Discovery of Dense Areas in Spatio-temporal Databases", Proc. of SSTD 2003, vol. , (2003), p. 1. Published

S. Guha, D. Gunopulos, N. Koudas, D. Srivastava, M. Vlachos, "Efficient Approximation Of Optimization Queries Under Parametric Aggregation Constraints", Proc. of 29th VLDB, Berlin, Germany, vol. , (2003), p. 1. Accepted

Sudipto Guha, Dimitrios Gunopulos, Nikos Koudas, "Correlating synchronous and asynchronous data-streams", Proc. ACM SIGKDD 2003, vol. , (2003), p. 1. Accepted

Donghui Zhang, Dimitrios Gunopulos, Vassilis J. Tsotras, Bernhard Seeger, "Temporal and spatio-temporal aggregations over data streams using multiple time granularities.", Information Systems, vol. 28, (2003), p. 61. Published

Vana Kalogeraki, Dimitrios Gunopulos, Demetrios Zeinalipour-Yazti, " A local search mechanism for peer-to-peer networks.", Proc. ACM CIKM, vol. , (2002), p. 1. Published

Vana Kalogeraki, Alex Delis, Dimitrios Gunopulos, "Peer-to-Peer Architectures for Scalable, Efficient and Reliable Media Services.", Proc. IPDPS 2003, vol. , (2003), p. 1. Published

M. Hadjieleftheriou, G. Kollios, and V.J. Tsotras, "Performance Evaluation of Spatiotemporal Selectivity Estimation Techniques", Statistical and Scientific Database Management (SSDBM), vol. , (2003), p. . Accepted

Jessica Lin, Michail Vlachos, Eamonn J. Keogh, Dimitrios Gunopulos, "Iterative Incremental Clustering of Time Series", Proceedings of the EDBT Conference, vol. , (2004), p. 106. Published

Themis Palpanas, Marc Cardle, Dimitrios Gunopulos, Eamonn J. Keogh, Victor B. Zordan, "Indexing Large Human-Motion Databases", Proceedings of VLDB, vol. , (2004), p. . Accepted

M. Hadjieleftheriou, E. Hoel, V. Tsotras, "SaIL: A Library for Efficient Application Integration of Spatial Indices", Proc. of the 16th International Conference on Scientific and Statistical Database Management (SSDBM), vol. , (2004), p. . Accepted

M. Hadjieleftheriou, V. Kriakov, Y. Tao, G. Kollios, A. Delis, V. Tsotras, "Spatio-Temporal Data Services in a Shared-Nothing Environment", Proc. of the 16th International Conference on Scientific and Statistical Database Management (SSDBM), vol. , (2004), p. . Accepted

D. Zeinalipour-Yiazti, V. Kalogeraki, D. Gunopulos, "Exploiting Locality for Scalable Information Retrieval in Peer-to-Peer Systems", Elsevier Information Systems Journal, vol. 30, (2005), p. 277. Published

M. Hadjieleftheriou, G. Kollios, D. Gunopulos, V. J. Tsotras, "Indexing Spatio-temporal Archives", VLDB Journal, vol. , (), p. . Accepted

D. Zeinalipour-Yazti, V. Kalogeraki and D. Gunopulos, "Information Retrieval Techniques for Peer-to-Peer Networks", IEEE CiSE Magazine, Special Issue on Web Engineering, IEEE Publications, vol. 4, (2004), p. 11. Published

S. Lin, D. Gunopulos, V. Kalogeraki, S. Lonardi, "A Data Compression Technique for Sensor Networks with Dynamic Bandwidth Allocation", 12th Int. Symp. On Temporal Representation and Reasoning (TIME 2005), vol. , (2005), p. 1. Published

S. Lin, S. Lonardi, D. Gunopulos, V. Kalogeraki, "Applying LVQ Techniques to Compress Historical Information in Sensor Networks", IEEE Data Compression Conference, March 29-31 2005, Snowbird, Utah, vol. , (2005), p. 1. Published

Book(s) of other one-time publications(s):

Other Specific Products:


Contributions:

Contributions within Discipline:

 We have developed new techniques foe analyzing data streams
(identifying correlations between streams), for allowing the user to
identify regions of interest in spatiotemporal datasets, and for
efficient dynamic compression of streams, with applications to sensor
networks.
We have also developed techniques for efficient searching and sharing
of resources in Peer-to-Peer systems.

 Our P2P keyword search techniques and the systems we built demonstrate
that it is possible to build trully distributed web search engines
that are robust and reliable.

We welcome comments on this system