SloNLP 2016

česká verze

SloNLP is a Slovakoczech workshop focused on Natural Language Processing and Computational Linguistics. Its main aim is to encourage cooperation among NLP researchers in Czechia and Slovakia; we therefore also invite Master and Doctoral students of computational linguistics to participate.

The topics of the workshop include: automatic speech recognition (ASR), automatic natural language analysis and generation (morphology, syntax, semantics...), dialogue systems, machine translation (MT), information retrieval (IR), practical applications of NLP technologies, and other topics of computational linguistics.

We also welcome work-in-progress papers with preliminary results, negative results papers, and future research suggestion papers.

The workshop is part of the ITAT conference, which takes place from 15th to 19th September 2016 in SOREA Hutník I. in Tatranské Matliare (1.5 km from Tatranská Lomnica; map).

Proceedings

proceedings coverSloNLP papers are part of the main ITAT proceedings, which are published online in CEUR, as well as in a printed form on Amazon. The suggested way of citing the papers using BibTeX is the following (see Biblio entry for a human-readable form):

@inproceedings{slonlp2016:XXXauthornameXXX,
  booktitle = {Proceedings of the 16th {ITAT}: Slovensko{\v{c}}esk{\'{y}} {NLP} workshop (Slo{NLP} 2016)},
  title = {XXX paper title XXX},
  editor = {Bro{\v{n}}a Brejov{\'{a}}},
  author = {XXX paper authors XXX},
  year = {2016},
  publisher = {CreateSpace Independent Publishing Platform},
  organization = {Comenius University in Bratislava, Faculty of Mathematics, Physics and Informatics},
  address = {Bratislava, Slovakia},
  venue = {{SOREA} Hutn{\'{i}}k I.},
  series = {{CEUR} Workshop Proceedings},
  volume = {1649},
  pages = {XXXpagefrom-pagetoXXX},
  isbn = {978-1537016740},
  issn = {1613-0073},
}

Programme

Saturday 17th September: common programme with the ITAT conference

  • 17:45-18:45 Jiří Materna (Seznam.cz): Aplikace strojového učení ve fulltextovém vyhledávání (SloNLP invited talk)
  • 19:00-          Conference banquet

Sunday 18th September: SloNLP workshop

  • 08:55-09:00 Rudolf Rosa: Opening
  • 09:00-09:20 Ladislav Lenc, Tomas Hercig: Neural Networks for Sentiment Analysis in Czech (paperslides)
  • 09:20-09:40 Ondřej Plátek, Petr Bělohlávek, Vojtěch Hudeček, Filip Jurčíček: Recurrent Neural Networks for Dialog State Tracking (paperslides)
  • 09:40-10:00 Tomáš Jelínek: Error rates and agreements of parsers: two experiments with ensemble parsing of Czech (paperslides)
  • 10:00-10:20 David Mareček: 12 years of Unsupervised Dependency Parsing (paperslides)
  • 10:20-10:30 Jaroslava Hlaváčová: Kolik potřebujeme slovních druhů? (discussion group)
     
  • 10:30-11:00 Coffee break
     
  • 11:10-11:30 Katrin Přikrylová, Vladislav Kuboň, Kateřina Veselovská: Logical vs. Natural Language Conjunctions in Czech: A Comparative Study (paperslides)
  • 11:30-11:40 Josef Chaloupka: Automatic Symbol Processing for Language Model Building in Slavic Languages (paperslidesdemo)
  • 11:40-12:00 Alexandr Rosen: Building and using corpora of non-native Czech (paperslidesdemocorpus)
  • 12:00-12:20 Rudolf Rosa: Czechizator - Čechizátor (paperslidesdemo)
  • 12:20-12:30 László Kovács, Erika Baksa-Varga, Daniel Hládek: Lexicon-based Post Correction of OCR Errors

SloNLP attendees can also attend other ITAT talks (see the conference programme). We wish to draw your attention especially to the following ones:

  • Saturday 17/9 11:00-11:30 Martin Kopp: How to Mimic Humans, Guide for Computers (WCIDM workshop talk)
  • Saturday 17/9 12:00-12:30 Tomáš Šabata: Modeling and Clustering the Behavior of Animals using Hidden Markov Models (WCIDM workshop talk)
  • Sunday 18/9 17:00-17:30 Martin Plátek, Karel Oliva: Redukční analýza A-stromů s minimalistickými omezeními (ITAT main track talk)
  • Sunday 18/9 17:30-18:00 Vladislav Kuboň, Markéta Lopatková, Tomáš Hercig: Searching for a Measure of Word Order Freedom (ITAT main track talk)

Accepted Papers

  • Ladislav Lenc and Tomas Hercig: Neural Networks for Sentiment Analysis in Czech
  • Ondřej Plátek, Petr Bělohlávek, Vojtěch Hudeček and Filip Jurčíček: Recurrent neural networks for dialog state tracking
  • Tomáš Jelínek: Error rates and agreements of parsers: two experiments with ensemble parsing of Czech
  • David Mareček: 12 years of Unsupervised Dependency Parsing
  • Katrin Přikrylová, Vladislav Kuboň and Kateřina Veselovská: Logical vs. Natural Language Conjunctions in Czech: A Comparative Study
  • Alexandr Rosen: Building and using corpora of non-native Czech
  • Rudolf Rosa: Czechizator - Čechizátor
  • Josef Chaloupka: Automatic Symbol Processing for Language Model Building in Slavic Languages

Instructions for Authors

  • paper in English (or Slovak or Czech)
  • 4-8 pages including references, in two columns (latex style, or Word style if you really must)
  • submit through EasyChair
  • each paper will be reviewed by 2 or 3 reviewers
  • if you cross-submit your paper to multiple conferences/journals and it gets accepted and published elsewhere, you must withdraw it from SloNLP before submitting the camera-ready version at the latest
  • oral presentation of the paper in English (or Slovak or Czech)
  • the papers will be published in the proceedings of ITAT (with an ISBN), which should be indexed in the Scopus database

Deadlines

  • until   5/6 29/5 abstract submission (50 - 200 words) -- extended deadline
  • until   8/6 5/6 paper submission -- extended deadline
  • until   3/7 announcement of accept/reject/conditional accept decisions
  • until 15/7 camera-ready submission (until 13/7 for conditionally accepted papers)
  • until 31/7 early registration
  • 17/9 welcome reception and invited talk
  • 18/9 workshop

All deadlines are 23:59 AoE.

Program Committee

Past Workshops