view release on metacpan or  search on metacpan

lib/Lingua/Interset/Tagset/FA/Conll.pm  view on Meta::CPAN

        'intfeature' => 'number',
        'simple_decode_map' =>
        {
            'SING' => 'sing', # xdá, ksí, ansán, xdáwnd, nfr
            'PLUR' => 'plur'  # mrdm, ksání, dígrán, afrád, znán
        }
    );
    # VERB FORM, MOOD, TENSE AND ASPECT ####################
    # Here is a website that helps understand Persian verb forms: http://www.jahanshiri.ir/pvc/en/
    # Some of the examples below are from the website, some from the PDF documentation of the treebank and some directly from the data.
    # The transliterations of the three sources differ.
    $atoms{tma} = $self->create_atom
    (
        'surfeature' => 'tma',
        'decode_map' =>
        {
            # There are 12 past tenses in Persian. Even some (not all) periphrastic tenses have their dedicated tags because
            # the participating verb forms are put together in one token in the treebank. Progressive tenses with the auxiliary
            # verb dáštan (to have) are not tagged, the auxiliary verb is tokenized separately (but note that some other tenses,
            # that are not called progressive in Persian, partially cover the meaning of the English continuous/progressive tenses).
            # The remaining past tenses can be classified along three dimensions: