Arnavutça dilinde haber arama ve erişim sistemi

Throughout the examination of thesis we screened the Albanian language and its principles to permit information retrieval for community development and adaption of system in a specific language. We already know free text searches are highly dependent in use of language. As an outcome we modified Standard Analyzer for Albanian language where all Albanian stopwords were added inside the analyzer. We accomplished the objective by using open source text retrieval library named Lucene which requires full text indexing afterwards, searching capability. Exploiting Lucene, the distinction of modified standard analyzer with three different analyzers were made. For indexing data we exploited our collection of data which is collected by five different most frequented newspapers in Albanian speaking countries at specific time period followed by fifty topics. As a result of retrieval, each analyzer were judged by relevance and later were evaluated. We marked that retrieved relevant documents for Albanian results, are getting improved by adding stopword list in that language, besides we noticed that Italian analyzer performed better for retrieving Albanian documents compared with other languages. Dataset, topics, qrels and other stuff related to this research are available at this link https://github.com/berruqazimi/InformationRetrivealInAlbanian.