Papers by Attila Novák
Much Ado About Nothing – Identification of Zero Copulas in Hungarian Using an NMT Model (2020.lrec-1)
Copied to clipboard
| Challenge: | Zero copulas are the phenomenon that nominal predicates lack an explicit verbal copule in default present tense 3rd person indicative cases. |
| Approach: | They propose a tool that can identify and mark the location of zero copulas in Hungarian clauses that contain nominal predicates at the right position. |
| Outcome: | The proposed tool can identify and mark the location of zero copulas, i.e. where an overt copulan would appear in the non-default cases. |
E-magyar – A Digital Language Processing System (L18-1)
Copied to clipboard
Tamás Váradi, Eszter Simon, Bálint Sass, Iván Mittelholcz, Attila Novák, Balázs Indig, Richárd Farkas, Veronika Vincze
| Challenge: | e-magyar is a free, open, modular text processing pipeline for Hungarian . existing tools were overhauled to operate in the pipeline with a uniform encoding and run in the same Java platform. |
| Approach: | e-magyar is a free, open, modular text processing pipeline for Hungarian . it was created by a collaborative effort by the language technology community . the system is aimed at a broad range of users, from language developers to researchers . |
| Outcome: | The proposed tool is open source and available for download on the HFST framework. |
NerKor+Cars-OntoNotes++ (2022.lrec-1)
Copied to clipboard
| Challenge: | In this paper, we present an upgraded version of the Hungarian NYTK-NerKor named entity corpus . it contains twice as many annotated spans and 7 times as many distinct entity types as the original version. |
| Approach: | They present an upgraded version of the Hungarian NYTK-NerKor named entity corpus with an extended OntoNotes 5 annotation scheme. |
| Outcome: | The enhanced version of the corpus contains twice as many annotated spans and 7 times more distinct entity types than the original version. |
CBOW-tag: a Modified CBOW Algorithm for Generating Embedding Models from Annotated Corpora (2020.lrec-1)
Copied to clipboard
| Challenge: | Using word2vec, we train distributional semantic models that predict a word from the context or vice versa. |
| Approach: | They propose a modified version of the CBOW algorithm implemented in the fastText framework that includes the representation of original word forms and their annotation at the same time. |
| Outcome: | The proposed model can answer questions such as What do we eat?, What can we do with a skeleton?, etc. |
Cross-Lingual Generation and Evaluation of a Wide-Coverage Lexical Semantic Resource (L18-1)
Copied to clipboard
| Challenge: | Neural word embedding models are not interpretable for humans by themselves . we present a method that assigns explicit symbolic semantic features to words . |
| Approach: | They propose a method that assigns explicit symbolic semantic features to words in an embedding model . they use a finite list of terms to make the model interpretable for humans . |
| Outcome: | The proposed method is shown to be very efficient for word embedding models . it can be applied across languages and can be used as a searchable semantic annotation . |