jerteh/SrpKor4Tagging

Corpus is created via mix of literary (⅓) and administrative (⅔) texts in Serbian. It is tagged for POS for 2 tagsets: Universal POS tagset and SrpLemKor tagset (made according to traditional, descriptive Serbian grammar) and lemmatized

It is constituted of a single jsonl file that can be loaded via:

from datasets import load_dataset
dataset = load_dataset("jerteh/SrpKor4Tagging")

Preview:

ds = dataset["train"][1389]
for x, y, z in zip(ds["token"], ds["ud"], ds["lemma"]):
    print(x, y, z)

Okrugle ADJ okrugao
mongolske ADJ mongolski
fizionomije NOUN fizionomija
behu AUX biti
ustupile VERB ustupiti
mesto NOUN mesto
licima NOUN lice
evropskijeg ADJ evropski
tipa NOUN tip
, PUNCT ,
prljavim ADJ prljav
, PUNCT ,
obradatelim ADJ obradateo
i CCONJ i
iscrpenim ADJ iscrpen
. PUNCT .

Citation:

@inproceedings{stankovic-etal-2020-machine,
    title = "Machine Learning and Deep Neural Network-Based Lemmatization and Morphosyntactic Tagging for {S}erbian",
    author = "Stankovic, Ranka  and
      {\v{S}}andrih, Branislava  and
      Krstev, Cvetana  and
      Utvi{\'c}, Milo{\v{s}}  and
      Skoric, Mihailo",
    editor = "Calzolari, Nicoletta  and
      B{\'e}chet, Fr{\'e}d{\'e}ric  and
      Blache, Philippe  and
      Choukri, Khalid  and
      Cieri, Christopher  and
      Declerck, Thierry  and
      Goggi, Sara  and
      Isahara, Hitoshi  and
      Maegaard, Bente  and
      Mariani, Joseph  and
      Mazo, H{\'e}l{\`e}ne  and
      Moreno, Asuncion  and
      Odijk, Jan  and
      Piperidis, Stelios",
    booktitle = "Proceedings of the Twelfth Language Resources and Evaluation Conference",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://aclanthology.org/2020.lrec-1.487",
    pages = "3954--3962",
    abstract = "The training of new tagger models for Serbian is primarily motivated by the enhancement of the existing tagset with the grammatical category of a gender. The harmonization of resources that were manually annotated within different projects over a long period of time was an important task, enabled by the development of tools that support partial automation. The supporting tools take into account different taggers and tagsets. This paper focuses on TreeTagger and spaCy taggers, and the annotation schema alignment between Serbian morphological dictionaries, MULTEXT-East and Universal Part-of-Speech tagset. The trained models will be used to publish the new version of the Corpus of Contemporary Serbian as well as the Serbian literary corpus. The performance of developed taggers were compared and the impact of training set size was investigated, which resulted in around 98{\%} PoS-tagging precision per token for both new models. The sr{\_}basic annotated dataset will also be published.",
    language = "English",
    ISBN = "979-10-95546-34-4",
}