Information Retrieval Lab

Postal Address:
Lab. 4.2, Fac. Informática, Universidade da Coruña,
Campus de Elviña s/n, 15071, A Coruña, SPAIN

Phone contact:
Voice: +34 981 167 000 ext. 1276
Fax: +34 981 167 160

Email:

• NowOnWeb can be defined as a NewsIR system that deals with the on-line news sources. It supplies with an effective and efficient approach to show news articles, about a specific topic, to the user in a comfortable way.


• The system was component-based designed and comprise a crawler to obtain the web-pages, an indexer to maintain the incremental index whit a temporal window, a news recognition and extraction module that enables the dynamic adding of sources, a news grouping component that uses novelty and redundancy detection approaches, and a summariser, among others.


• As commercial product NowOnWeb offers:
  • A web service that serves the news about the user information needs.
  • A flexible and adaptable application to the needs of the users and their areas.
  • An efficient and scalable product with effective results.
  • A component based software with modules that allows the reuse in different applications.
  • A great product to the press areas of the companies and institutions that also can be adapted to other fields as technology surveillance or vertical search.

• Lexisla can be defined as a LegalIR system that crawls, segments and index official legal bulletins. It is currently under development phase with the support of Fundación Calidade


• The system is composed by different modules such as specific crawler, a document analyser component, an indexer and a advanced search interface.


• As commercial product Lexisla offers:
  • Flexible multisource segmentation algorithm
  • Efficient document analysis and processing
  • Automatic content updating
  • Advanced local browsing over the bulletins
  • Advanced search capabilities

• The CCT this is a development carried out by the IRLab in collaboration with the English Department. Indeed the application came up because the need of the Muste Group of has a system to manage and exploit its linguistic corpus.


• The objective is help linguists to extract and condense valuable information for their research. But the application was not designed tied to the Coruña Corpus and it supports any xml-formatted corpus being, in this sense, an application that could be widely used.


• As commercial product The CCT offers:
  • Linguistic corpus management, not only documents as text but also author information and styled document rendering.
  • Treatment and validation of TEI encoded documents with support for non-standard characters. It supplies information about the format errors in order to allow the correction by the linguists.
  • Intra-documental and collection basic search by single terms.
  • Concordance generation (key-word in context) of all the term appearances and location in the document.
  • Prefix, suffix and regular expressions search, which is very useful for the linguistic work.
  • Phrase search with term distance specification in order to search for linguistic structures.
  • Generation of types and tokens lists in document and collection level to allow statistical study of the terms occurrences.