Terrier is a highly flexible, efficient, effective, and robust search engine with state-of-the-art indexing and retrieval functionality. It provides an ideal platform for the rapid development of large-scale retrieval applications.

The open source version of Terrier is a transparent platform for research and experimentation in text retrieval. The research put into Terrier constantly expands towards new branches of the wider information retrieval field, making Terrier an ideal, strong, modular, and state-of-the-art platform for developing, assessing, and evaluating new concepts and ideas.

Terrier is written in Java and was used for Web and Enterprise search, Desktop, Intranet and Vertical search engines, as well as for developing and evaluating novel large-scale text information retrieval techniques and applications.

For more information, please consult the Terrier wiki.

The indexing of the MAREC data needs a pre-processing step to transform the MAREC XML format into an XML format supported by Terrier.

The documents of the MAREC collection are full-text indexed. Furthermore, some fields like abstract, title, or description are indexed as well to allow searching only in these parts of the documents.