
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/">
  <dc:language>eng</dc:language>
  <dc:format>PDF/A (144 lista)</dc:format>
  <dc:format>4030838 bytes</dc:format>
  <dc:title xml:lang="eng">Treniranje strukturnih klasifikatora za različite funkcije gubitaka sa primenom na probleme klasifikovanja sekvenci : doktorska disertacija</dc:title>
  <dc:contributor>Todorović, Branimir 1967-</dc:contributor>
  <dc:contributor>Ćirić, Miroslav 1964-</dc:contributor>
  <dc:contributor>Stanimirović, Predrag 1959-</dc:contributor>
  <dc:contributor>Stanković, Miomir</dc:contributor>
  <dc:contributor>Stoimenov, Leonid</dc:contributor>
  <dc:date>2015</dc:date>
  <dc:rights>http://creativecommons.org/licenses/by-nc-nd/2.0/at/legalcode</dc:rights>
  <dc:type>info:eu-repo/semantics/bachelorThesis</dc:type>
  <dc:description xml:lang="eng">This thesis presents algorithms for training structured classifiers over different loss functions. It introduces a new primal subgradient method for the optimization of average sum loss, several extensions of max-margin classifiers to the k-best case, a sequential dual method for the structured ramp loss optimization, It considers different decoding algorithms over semirings and presents an organization of a two-structured-model committee in order to improve the results. It includes theoretical analysis of introduced algorithms, as well as experimental results on sequence labelling problems in natural language processing.</dc:description>
  <dc:description xml:lang="eng">Umnoženo za odbranu. Univerzitet u Nišu, Prirodno-matematički fakultet, 2015. Bibliografija: str. 87-93. Izvod ; Abstract. artificial intelligence Datum odbrane: 06.07.2015.</dc:description>
  <dc:identifier>https://phaidrani.ni.ac.rs/o:1009</dc:identifier>
  <dc:identifier>cobiss:1025057769</dc:identifier>
  <dc:identifier>thesis:2252</dc:identifier>
  <dc:creator>Mančev, Dejan, 1985-</dc:creator>
</oai_dc:dc>