Terrier

3.5 represents a significant update over the previous 3.0 version of Terrier, including: Document-at-a-time (DAAT) retrieval for large indices; refactored tokenisation for enhanced multi-language support; upgraded Hadoop support to version 0.20; synonym support in query language and retrieval; out-of-the box indexing support for query-biased summaries and an improved example Web-based interface; new, 2nd generation DFR models as well as other recent effective information-theoretic effective models; and many more JUnit tests (now 300+).

Terrier 3.0 is a major update, including field-based models (such as BM25F), and term dependence proximity models (Markov Random Fields). Index changes have made the platform more scalable. Hadoop MapReduce indexing is improved. A JSP Web-based interface is now included.

This is a substantial update, which includes new
support for Hadoop, primarily a Hadoop Map Reduce
indexing system, allowing large collections of
documents to be indexed in a highly distributed
fashion. Also included are various minor
improvements, including improved support for the
IIT CDIP1 (TREC Legal track) collection, and
various bug fixes. This is intended to be the
ultimate release in the 2.x series.

This is a minor update that contains some bugfixes, and some minor improvements. Support for indexing various test collections has been improved (CLEF and TREC Legal track), and the flexibility of the settings of some applications such as the Desktop Search and Interactive Terrier has been enhanced. This release includes a filesystem abstraction layer, which allows various types of files to be accessed through a uniform API. For example, indexing an HTTP Web page is as easy as indexing a local document. Moreover, a notable indexing bug affecting only the Windows platform was resolved.

This is a major update that integrates a new faster indexing architecture contributed by the University of A Coruna (Spain). Other enhancements include the new Divergence From Randomness weighting model (DFRee) from Fondazione Ugo Bordoni (Italy), which provides robust performance without the need for any parameter tuning.