Workshop: Machine learning and language processes

MEGACRO (funded by the Croatian Science Foundation and supported by the University of Zagreb) organizes a workshop Machine learning and language processes for doctoral and postdoctoral students.

The workshop will be held on Friday, 6 December 2019 in the room A 302 (9:30-14:30) at the Faculty of Humanities and Social Sciences in Zagreb.

Workshop lectures:

Petar Milin, PhD University of Birmingham (project Out of our Minds)

https://outofourminds.bham.ac.uk/

Artificial Neural Networks against Natural Language Ailments
Maja Anđel, PhD Faculty of Humanities and Social Sciences, Zagreb

Leabra – biologically plausible modelling algorithm
Petar Milin, PhD University of Birmingham (project Out of our Minds)

https://outofourminds.bham.ac.uk/

A royal way to a headache: Complex research phenomena require complex statistical models (A gentle introduction to Mixed Effects Modeling)

 

 102 total views

SCLC-2019

Professor Anita Peti-Stantić will present a paper on Informativity of pronomial Clitics in Croatian prepared with Mirjana Tonković, PhD, at the 16th conference of the Slavic Cognitive Linguistics Association  (SCLA).
Professor Mateusz-Milan Stanojević (a member of the MEGACRO research group) will be presenting a paper on Premodification by adjectives in Croatian verbal idioms, prepared with professor Jelena Parizoska.
The conference is going to be held at Harvard University (Cambridge/Boston, Massachusetts) 12-14 October 2019.
Conference programme is available here

 73 total views

ESCOP: Psycholinguistic estimates and lexical decision times for 1000 Croatian nouns

Mirjana Tonković will present a poster at the 21st  ESCOP conference (conference organized by the European Society for Cognitive Psychology, 25- 28 September 2019, Tenerife). She will present the topic of Psycholinguistic estimates and lexical decision times for 1000 Croatian nouns  prepared in collaboration with Francesca Dumančić, Denis Vlašiček and two MEGACRO researchers, Gordana Keresteš and Anita Peti-Stantić.

The list of all poster presentations is available here

Conference programme is available here

 77 total views

Croatian psycholinguistic database expanded by 3000 words

Croatian psycholinguistic database, created and developed as a part of the MEGACRO project (HRZZ-IP-2016-06-1210) has been expanded by 3000 new words – nouns, verbs and adjectives. The two new sampling sources were the Croatian frequency dictionary (for 1500 words) and the most frequently used textbooks for the fourth, fifth and sixth grade (1500 words from the textbooks for Croatian language, Natural Sciences, History, Geography and Mathematics).

University students gave their ratings of concreteness, imageability, subjective frequency and age of acquisition for these new 3000 words, just as it was the case for the first 3000 words included in the database.

Additional data available for each word also include word class, word length, gender, animacy/inanimacy for nouns and word frequency from the Croatian web Corpus hrWaC

One can search through the database here.

If you want your search to include only the first 3000 words added to the database, the first drop-down menu should be set to “Baza riječi” and the third box should contain the number “1“. 

If you want your search to include only the second 3000 words added to the database, the first drop-down menu should be set to “Baza riječi”, and the third box should contain the number “2“.

If you wish to search through all 6000 words currently in the CPD, simply choose the option “SVE RIJEČI”.

 98 total views