metadata

language:
  - pt
license: mit
task_categories:
  - token-classification
pretty_name: Buscapé
dataset_info:
  - config_name: default
    features:
      - name: tokens
        sequence: string
      - name: srl_frames
        list:
          - name: frames
            sequence: string
          - name: verb
            dtype: string
    splits:
      - name: train
        num_bytes: 210904
        num_examples: 709
    download_size: 47061
    dataset_size: 210904
  - config_name: flatten
    features:
      - name: tokens
        sequence: string
      - name: verb
        dtype: string
      - name: frames
        sequence: int64
    splits:
      - name: train
        num_bytes: 236352
        num_examples: 709
    download_size: 46664
    dataset_size: 236352
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
  - config_name: flatten
    data_files:
      - split: train
        path: flatten/train-*
tags:
  - semantic role labelling
  - srl

Buscapé Sample annotated for Semantic Role Labelling

Propbank-Br Corpora Buscapé Sample

The Propbank-Br is a project that aims to annotate corpora with semantic role labels for the purpose of creating training datasets for automated semantic role classifiers. The annotation scheme is quite similar to that of the English Propbank (Palmer et al., 2005), with language-specific differences taken into account. The set of semantic roles was designed to facilitate automatic learning. The annotation is done on syntactic trees generated by the Palavras parser (Bick, 2000).

This particular sample was annotated for the purpose of evaluating semantic role classifiers. It contains 840 instances annotated with semantic role labels on syntactic trees generated by the Palavras parser (Bick, 2000). The instances were extracted from the Buscapé corpus (Hartmann et al. 2014), a corpus of user reviews on products. The syntactic trees in the sample were not reviewed by humans, and were annotated using two annotators for each sentence (double-blind annotation).

Data Treatment at LIAAD

131 propositions were excluded in this revision of the dataset. These included propositions with verb index annotation errors or no verb annotations, and propositions with more than one label for a word. Additionally, we removed arguments labeled as "AM-MED" or "AM-PIN" because there is no mention of these labels in the annotation guides, and we removed any propositions with flags "WRONGSUBCORPUS", "LATER" or "REEXAMINE", since, according to the guide, these indicate something wrong with the sentence that prevents its annotation.

Annotated by: PROSA
Language: Portuguese

Dataset Sources

Link: http://www.nilc.icmc.usp.br/semanticnlp/index.php?id=index&id_sub=principal&dir_sub=includes/projects/propbankbr&dir=includes/projects/propbankbr&lang=pt-br
Paper: Hartmann, N. S.; Avanço. L.; Balage, P. P.; Duran, M. S.; Nunes, M. G. V.; Pardo, T.; Aluísio, S. (2014). A Large Opinion Corpus in Portuguese - Tackling Out-Of-Vocabulary Words. In: Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC 2014).

Citation

BibTeX:

@inproceedings{hartmann-etal-2014-large,
    title = "A Large Corpus of Product Reviews in {P}ortuguese: Tackling Out-Of-Vocabulary Words",
    author = "Hartmann, Nathan  and
      Avan{\c{c}}o, Lucas  and
      Balage, Pedro  and
      Duran, Magali  and
      das Gra{\c{c}}as Volpe Nunes, Maria  and
      Pardo, Thiago  and
      Alu{\'\i}sio, Sandra",
    editor = "Calzolari, Nicoletta  and
      Choukri, Khalid  and
      Declerck, Thierry  and
      Loftsson, Hrafn  and
      Maegaard, Bente  and
      Mariani, Joseph  and
      Moreno, Asuncion  and
      Odijk, Jan  and
      Piperidis, Stelios",
    booktitle = "Proceedings of the Ninth International Conference on Language Resources and Evaluation ({LREC}'14)",
    month = may,
    year = "2014",
    address = "Reykjavik, Iceland",
    publisher = "European Language Resources Association (ELRA)",
    url = "http://www.lrec-conf.org/proceedings/lrec2014/pdf/413_Paper.pdf",
    pages = "3865--3871",
    abstract = "Web 2.0 has allowed a never imagined communication boom. With the widespread use of computational and mobile devices, anyone, in practically any language, may post comments in the web. As such, formal language is not necessarily used. In fact, in these communicative situations, language is marked by the absence of more complex syntactic structures and the presence of internet slang, with missing diacritics, repetitions of vowels, and the use of chat-speak style abbreviations, emoticons and colloquial expressions. Such language use poses severe new challenges for Natural Language Processing (NLP) tools and applications, which, so far, have focused on well-written texts. In this work, we report the construction of a large web corpus of product reviews in Brazilian Portuguese and the analysis of its lexical phenomena, which support the development of a lexical normalization tool for, in future work, subsidizing the use of standard NLP products for web opinion mining and summarization purposes.",
}

APA:

Hartmann, N. S., Avanço, L. V., Balage, P. P., Duran, M. S., Nunes, M. das G. V., Pardo, T. A. S., & Aluísio, S. M. (2014). A large corpus of product reviews in Portuguese: tackling out-of-vocabulary words. In Proceedings. Paris: ELRA. Recuperado de http://www.lrec-conf.org/proceedings/lrec2014/pdf/413_Paper.pdf

Dataset Card Authors

Rita Lopes | https://huggingface.co/rita443 | [email protected]