• Descrizione :

Set di dati ceco da dati a testo nel dominio dei ristoranti. Le rappresentazioni del significato dell'input contengono un tipo di atto di dialogo (informare, confermare ecc.), slot (cibo, area, ecc.) ei loro valori. È nato come traduzione del set di dati inglese San Francisco Restaurants di Wen et al. (2015).

Diviso Esempi
'test' 842
'train' 3.569
'validation' 781
  • Struttura delle caratteristiche :
    'delex_input_text': FeaturesDict({
        'table': Sequence({
            'column_header': string,
            'content': string,
            'row_number': int16,
    'delex_target_text': string,
    'input_text': FeaturesDict({
        'table': Sequence({
            'column_header': string,
            'content': string,
            'row_number': int16,
    'target_text': string,
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
delex_input_text CaratteristicheDict
delex_input_text/tabella Sequenza
delex_input_text/table/column_header Tensore corda
delex_input_text/tabella/contenuto Tensore corda
delex_input_text/table/row_number Tensore int16
delex_target_text Tensore corda
testo di input CaratteristicheDict
testo_input/tabella Sequenza
input_text/table/column_header Tensore corda
testo_input/tabella/contenuto Tensore corda
testo_input/tabella/numero_riga Tensore int16
testo_destinazione Tensore corda
  • Citazione :
        author = {Dušek, Ondřej and Jurčíček, Filip},
        title = {Neural {Generation} for {Czech}: {Data} and {Baselines} },
        shorttitle = {Neural {Generation} for {Czech} },
        url = {https://www.aclweb.org/anthology/W19-8670/},
        urldate = {2019-10-18},
        booktitle = {Proceedings of the 12th {International} {Conference} on {Natural} {Language} {Generation} ({INLG} 2019)},
        month = oct,
        address = {Tokyo, Japan},
        year = {2019},
        pages = {563--574},
        abstract = {We present the first dataset targeted at end-to-end NLG in Czech in the restaurant domain, along with several strong baseline models using the sequence-to-sequence approach. While non-English NLG is under-explored in general, Czech, as a morphologically rich language, makes the task even harder: Since Czech requires inflecting named entities, delexicalization or copy mechanisms do not work out-of-the-box and lexicalizing the generated outputs is non-trivial. In our experiments, we present two different approaches to this this problem: (1) using a neural language model to select the correct inflected form while lexicalizing, (2) a two-step generation setup: our sequence-to-sequence model generates an interleaved sequence of lemmas and morphological tags, which are then inflected by a morphological generator.},