Query-Load Balancing by Caching Search Results in Peer-to-Peer Information Retrieval Networks
Tigelaar, A. S. & Hiemstra, D.
In Proceedings of DIR 2011, Amsterdam, The Netherlands (pp. 28-31).

View in Repository

Abstract
For peer-to-peer web search engines it is important to keep the delay between receiving a query and providing search results within an acceptable range for the end user. How to achieve this remains an open challenge. One way to reduce delays is by caching search results for queries and allowing peers to access each others cache. In this paper we explore the limitations of search result caching in large-scale peer-to-peer information retrieval networks by simulating such networks with increasing levels of realism. We find that cache hit ratios of at least thirty-three percent are attainable.

Presented at Dutch-Belgian Information Retrieval 2011 Workshop, February 4th 2011, Amsterdam, The Netherlands.

Query-Based Sampling using Snippets
Tigelaar, A. S. & Hiemstra, D.
In Proceedings of LSDS-IR 2010, Geneva, Switzerland (pp. 9-14).

View in Repository

Abstract
Query-based sampling is a commonly used approach to model the content of servers. Conventionally, queries are sent to a server and the documents in the search results returned are downloaded in full as representation of the server’s content. We present an approach that uses the document snippets in the search results as samples instead of downloading the entire documents. We show this yields equal or better modeling performance for the same bandwidth consumption depending on collection characteristics, like document length distribution and homogeneity. Query-based sampling using snippets is a useful approach for real-world systems, since it requires no extra operations beyond exchanging queries and search results.

Presented at the Large-Scale Distributed Systems for Information Retrieval Workshop on July 23rd in Geneva, Switzerland.

More Photos

Query-Based Sampling using Only Snippets
Tigelaar, A. S. & Hiemstra, D.
Poster NWO SIREN 2009, Enschede, The Netherlands.

Matching Queries to Frequently Asked Questions: Search Functionality for the MRSA Web-Portal
Tigelaar, A. S. & Akker, R. op den & Verhoeven, F.
In Proceedings of DIR 2009, Enschede, The Netherlands (pp. 26-33).

View in Repository

Abstract
As part of the long-term EUREGIO MRSA-net project a system was developed which enables health care workers and the general public to quickly find answers to their questions regarding the MRSA pathogen. This paper focuses on how these questions can be answered using Information Retrieval (IR) and Natural Language Processing (NLP) techniques on a Frequently-Asked-Questions-style (FAQ) database.

Presented at the Dutch-Belgian Information Retrieval Workshop 2009