Exploring the Terrier information Retrieval Platform for Web Search of Documents Written in Macedonian

Abstract

Terrier is a modular and scalable platform for rapid development of Information Retrieval (IR) systems. This paper presents a short overview of the Terrier architecture and describes ways in which it can be extended for more effective indexing and searching of documents written in Macedonian language. Although Terrier supports out of the box search in a few non-English languages, the Macedonian language poses some specific challenges, especially when search of Web content is involved. An integrated search platform is developed for the purpose of this research extending the text retrieval engine with a more advanced content filtering capabilities. Some of the proposed methods can be easily applied to other non-English languages.