Ph.D. Research:

Peer-to-peer networks, sensor networks and the web are instances of dynamic data networks, the next generation of the distributed database systems with large scale, autonomous nodes, and dynamic data-set, node-set, and topology. Despite the complexity of the new physical level at these databases, they are expected to provide a superset of the traditional database functionality while maintaining the same user view.

I study approaches to extend the data management techniques (and particularly, query processing methods) from the database literature to adapt to the complex physical level of these new database systems. I believe modeling these distributed database systems upon autonomic complex systems such as social networks is a key solution to enable the adaptation process. With my thesis, titled "Approximate Query Answering in Unstructured Peer-to-Peer Databases", I particularly focus on developing sample-based distributed query processing techniques to answer both set-valued and aggregate SQL queries in peer-to-peer databases.

Keywords: Distributed Data Management, Peer-to-Peer Databases, Sensor Databases, Approximate Query Answering, Sampling, Similarity Search, Complex Systems, Percolation Theory, Small-World

Projects:

• Digest: A generic testbed for evaluation of the distributed query processing algorithms in peer-to-peer databases
• Kerena: A peer-to-peer streaming system
• WSPDS (Web Services Peer-to-peer Discovery Service): A peer-to-peer discovery service for Web Services
• SNDB (Sensor Network Database): A remote environment-monitoring system enabled by a network of sensors embedded in the environment
  
  

Publications:

• Farnoush Banaei-Kashani and Cyrus Shahabi, Self-sufficient and adaptive sampling for aggregation in peer-to-peer databases, submitted for review.
• Farnoush Banaei-Kashani and Cyrus Shahabi, Partial query processing in peer-to-peer databases, submitted for review.
• Farnoush Banaei-Kashani and Cyrus Shahabi, Continuous aggregate queries in peer-to-peer databases, submitted for review.
• Farnoush Banaei-Kashani and Cyrus Shahabi, Partial Selection Query in Peer-to-Peer Databases, Proceedings of the 22nd International Conference on Data Engineering (ICDE'06), April 2006.
• Cyrus Shahabi and Farnoush Banaei-Kashani. Modeling Peer-to-Peer Data Networks under Complex System Theory, International Journal of Computational Science and Engineering (IJCSE), Accepted for Publication.
• Farnoush Banaei-Kashani and Cyrus Shahabi, SWAM: A Family of Access Methods for Similarity Search in Querical Data Networks, Information and Knowledge Management (CIKM'04), Washington D.C., November 2004.
• Farnoush Banaei-Kashani, Ching-Chien Chen, and Cyrus Shahabi, WSPDS: Web Services Peer-to-peer Discovery Service, International Symposium on Web Services and Applications(ISWS'04), Nevada, June 2004.
• Farnoush Banaei-Kashani and Cyrus Shahabi, Searchable Querical Data Networks, International Workshop on Databases, Information Systems and Peer-to-Peer Computing in conjunction with VLDB'03 , Berlin, Germany, September, 2003
• Farnoush Banaei-Kashani, Cyrus Shahabi, Brief Announcement: Efficient Flooding in Power-Law Networks, Twenty-Second ACM Symposium on Principles of Distributed Computing (PODC'03) , Boston, Massachusetts, July, 2003
• Farnoush Banaei-Kashani and Cyrus Shahabi, Criticality-based Analysis and Design of Unstructured Peer-to-Peer Networks as Complex Systems, Third International Workshop on Global and Peer-to-Peer Computing (GP2PC) in conjunction with CCGrid'03 , Tokyo, Japan, May, 2003.
• Cyrus Shahabi, Farnoush Banaei-Kashani, Decentralized Resource Management for a Distributed Continuous Media Server, IEEE Transactions on Parallel and Distributed Systems, Vol. 13, No. 7, pp. 710-727, July 2002.

 

M.S. Research:

Projects:

• InSite: Web Usage Mining (WUM) for web persnalization

Publications:

• Cyrus Shahabi, Farnoush Banaei-Kashani, Efficient and Anonymous Web Usage Mining for Web Personalization, , To appear in INFORMS Journal on Computing - Special Issue on Data Mining, 2002.
• Cyrus Shahabi, Farnoush Banaei-Kashani, A Framework for Efficient and Anonymous Web Usage Mining Based on Client-Side Tracking, , Book Chapter: Lecture Notes in Computer Science - Mining Log Data Across A11 Customer Touch Points (WebKDD'01), VOL. 2356, 2001.
• Cyrus Shahabi, Farnoush Banaei-Kashani, Yi-Shin Chen, Dennis McLeod, Yoda: An Accurate and Scalable Web-based Recommendation System, Sixth International Conference on Cooperative Information Systems (CoopIS'01) , Trento, Italy, September 2001.
• Cyrus Shahabi, Farnoush Banaei-Kashani, Jabed Faruque, and Adil Faisal, Feature Matrices: A Model for Efficient and Anonymous Web Usage Mining , 12th International Workshop on Database and Expert Systems Applications (DEXA - ECWeb'01), Germany, September 2001.
• Cyrus Shahabi, Farnoush Banaei-Kashani, and Jabed Faruque, A Reliable, Efficient, and Scalable System for Web Usage Data Acquisition, WebKDD'01 Workshop in conjunction with the ACM-SIGKDD 2001, San Francisco, CA, August, 2001.

Services:

• USENIX 2006, Co-reviewer
• SIGMOD 2005, Co-reviewer
• SIGCOMM 2005, Co-reviewer
• CIKM 2004, Session Chair
• SIGMOD 2004, Co-reviewer
• SIGCOMM 2003, Co-reviewer
• CIKM 2003, Co-reviewer
• SIGMETRICS 2002, Student Volunteer
• CIKM 2002, Co-reviewer
• MM 2001, Co-reviewer

Contact Information:

E-mail address:	banaeika@usc.edu
Contact Numbers:	Office (RTH-323): (213) 821-1462 FAX: (213) 821-1739
Campus Address :	University of Southern California Computer Science Department SAL 300 Los Angeles, CA 90089-0781