def _get_gender_dict()

in src/morphological_filtering.py [0:0]

• 21 lines of code
• 14 McCabe index (conditional complexity)

    def _get_gender_dict() -> Dict[str, str]:
        """Get the gender dictionary (mapping of word to gender label) for the supported genders for DE."""
        gender_dict = {}
        fem_morphs, msc_morphs = set(), set()
        with open(GERMAN_MORPH_DICT, 'r') as morphs:
            for line in morphs:
                # read in each word and its corresponding gender
                line = line.strip().split()
                if len(line) > 1:
                    tag = line[1].split(',')
                    if len(tag) > 2:
                        if tag[0] == 'NN':
                            if tag[1] == 'fem':
                                fem_morphs.add(line[0].lower())
                            elif tag[1] == 'masc':
                                msc_morphs.add(line[0].lower())
        # if a word occurs with both feminine and masculine gender, exclude it
        for word in fem_morphs:
            if word not in msc_morphs:
                gender_dict[word] = FEM_LABEL
        for word in msc_morphs:
            if word not in fem_morphs:
                gender_dict[word] = MSC_LABEL
        return gender_dict