AGRIS 2.0 Online

AGRIS version 2.0 was released on 20/12/2010 and it's available at:

http://agris.fao.org/

This is a new version based on the latest web and search technologies. Some new features are:

  • Apache Solr search engine: New core architecture and technology. Solr is the server-ization of Lucene (not a wrapper), that uses HTTP requests for indexing and querying documents. High performances and more functionalities.
  • STRUTS 2 web framework: The usage of Struts 2 allows a lot of functionalities, such as internationalization and concurrency management.
  • Sorting filters: Solr allows to simply introduce filters to sort the results of a query. The default is “sort by scoring descending”, where the scoring is computed by the engine that takes into account the term frequency of the fields of a document, the inverse of the number of documents in which the term appears, and the presence of optional terms and fields’ length.
  • AGROVOC field boosted: Actually, the term frequency of the “agrovoc” field is boosted, i.e. it is more relevant (100 to 1). Thus, documents matching a query in the AGROVOC field will receive a higher score.
  • Stars for displaying the score: Each result has a score that represents how the document matches the query-string. The score is displayed by the usage of (five) stars, that relate the score of the document with the maximum score for the current query.
  • More like this: “faceting” enhances search results with aggregated information over all of the documents found in the search. It is more or less analogous to SQL’s group by feature on a column with count(*). It requires a copy field in the index and analyzes this field for all of the documents found and counts occurrences of words indexed. Words with greater occurrence are the suggestions. Suggestions are clickable and produce new searches.
  • Spell checking: Just because spelling suggestion is offered does not mean that using it will yield matching documents, but it means that the individual correct word is in the index. A query, which is comprised of multiple words that are correctly spelled, yet whose query returns no results, will not result in spelling corrections
  • Extended search syntax
  • New advanced search: The query is built on the fly. Each query produced with the advanced search is coherent with the user necessity and that the users can navigate between search and advanced search correctly. Basic search is more powerful than search assistant: it allows also boolean searche
  • Error messages: Error messages when the user types a query with a wrong syntax or when an exception due to the index occurs
  • Internationalization: The user interface has been translated in various languages and the user can choose which language the UI should expose. Results are most probably in English, as the repository is in English.

Add comment

Log in or register to post comments