medical-bios.py

# coding=utf-8
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Medical BIOS"""

import json
import os
import textwrap

import datasets


MAIN_CITATION = """https://aclanthology.org/2023.emnlp-main.427/"""
_DESCRIPTION = """NA"""
MAIN_PATH = 'https://huggingface.co/datasets/coastalcph/medical-bios/resolve/main'


class MedicalBIOSConfig(datasets.BuilderConfig):
    """BuilderConfig for Medical BIOS."""

    def __init__(
        self,
        label_classes,
        url,
        data_url,
        citation,
        **kwargs,
    ):
        """BuilderConfig for Medical BIOS.
        Args:
          label_classes: `list`, list of label classes
          url: `string`, url for the original project
          data_url: `string`, url to download the zip file from
          data_file: `string`, filename for data set
          citation: `string`, citation for the data set
          url: `string`, url for information about the data set
          **kwargs: keyword arguments forwarded to super.
        """
        super(MedicalBIOSConfig, self).__init__(version=datasets.Version("1.0.0", ""), **kwargs)
        self.label_classes = label_classes
        self.url = url
        self.data_url = data_url
        self.citation = citation


class XAIFairness(datasets.GeneratorBasedBuilder):
    """Fairlex: A multilingual benchmark for evaluating fairness in legal text processing. Version 1.0"""

    BUILDER_CONFIGS = [
        MedicalBIOSConfig(
            name="standard",
            description=textwrap.dedent(
                """\
                The dataset is based on the Common Crawl. Specifically, De-Arteaga et al. identified online
                biographies, written in English, by filtering for lines that began
                with a name-like pattern (i.e., a sequence of two capitalized words)
                followed by the string “is a(n) (xxx) title,” where title is 
                an occupation from the BLS Standard Occupation Classification system.
                This version of the dataset comprises English biographies labeled with occupations.
                We also include a subset of biographies labeled with human rationales.
                """
            ),
            label_classes=['psychologist', 'surgeon', 'nurse', 'dentist', 'physician'],
            data_url=os.path.join(MAIN_PATH, "bios.zip"),
            url="https://github.com/microsoft/biosbias",
            citation=textwrap.dedent(
                """\
            @inproceedings{eberle-etal-2023-rather,
                title = "Rather a Nurse than a Physician - Contrastive Explanations under Investigation",
                author = "Eberle, Oliver  and
                  Chalkidis, Ilias  and
                  Cabello, Laura  and
                  Brandl, Stephanie",
                booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
                year = "2023",
                publisher = "Association for Computational Linguistics",
                url = "https://aclanthology.org/2023.emnlp-main.427",
            }"""
            ),
        ),
        MedicalBIOSConfig(
            name="rationales",
            description=textwrap.dedent(
                """\
                The dataset is based on the Common Crawl. Specifically, De-Arteaga et al. identified online
                biographies, written in English, by filtering for lines that began
                with a name-like pattern (i.e., a sequence of two capitalized words)
                followed by the string “is a(n) (xxx) title,” where title is 
                an occupation from the BLS Standard Occupation Classification system.
                This version of the dataset comprises English biographies labeled with occupations.
                We also include a subset of biographies labeled with human rationales.
                """
            ),
            label_classes=['psychologist', 'surgeon', 'nurse', 'dentist', 'physician'],
            data_url=os.path.join(MAIN_PATH, "bios.zip"),
            url="https://github.com/microsoft/biosbias",
            citation=textwrap.dedent(
                """\
            @inproceedings{eberle-etal-2023-rather,
                title = "Rather a Nurse than a Physician - Contrastive Explanations under Investigation",
                author = "Eberle, Oliver  and
                  Chalkidis, Ilias  and
                  Cabello, Laura  and
                  Brandl, Stephanie",
                booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
                year = "2023",
                publisher = "Association for Computational Linguistics",
                url = "https://aclanthology.org/2023.emnlp-main.427",
            }"""
            ),
        ),
    ]

    def _info(self):
        if self.config.name == "standard":
            features = {"text": datasets.Value("string"), "label": datasets.ClassLabel(names=self.config.label_classes)}
        else:
            features = {"text": datasets.Value("string"), "label": datasets.ClassLabel(names=self.config.label_classes),
                        "foil": datasets.ClassLabel(names=self.config.label_classes),
                        "words": datasets.Sequence(datasets.Value("string")),
                        "rationales": datasets.Sequence(datasets.Value("int8")),
                        "contrastive_rationales": datasets.Sequence(datasets.Value("int8")),
                        "annotations": datasets.Sequence(datasets.Sequence(datasets.Value("int8"))),
                        "contrastive_annotations": datasets.Sequence(datasets.Sequence(datasets.Value("int8")))}
        return datasets.DatasetInfo(
            description=self.config.description,
            features=datasets.Features(features),
            homepage=self.config.url,
            citation=self.config.citation + "\n" + MAIN_CITATION,
        )

    def _split_generators(self, dl_manager):
        data_dir = dl_manager.download_and_extract(self.config.data_url)
        if self.config.name == 'standard':
            return [
                datasets.SplitGenerator(
                    name=datasets.Split.TRAIN,
                    # These kwargs will be passed to _generate_examples
                    gen_kwargs={
                        "filepath": os.path.join(data_dir, f"train.jsonl"),
                        "split": "train",
                    },
                ),
                datasets.SplitGenerator(
                    name=datasets.Split.TEST,
                    # These kwargs will be passed to _generate_examples
                    gen_kwargs={
                        "filepath": os.path.join(data_dir, "test.jsonl"),
                        "split": "test",
                    },
                ),
                datasets.SplitGenerator(
                    name=datasets.Split.VALIDATION,
                    # These kwargs will be passed to _generate_examples
                    gen_kwargs={
                        "filepath": os.path.join(data_dir, f"validation.jsonl"),
                        "split": "val",
                    },
                ),
            ]
        else:
            return [
                datasets.SplitGenerator(
                    name=datasets.Split.TEST,
                    # These kwargs will be passed to _generate_examples
                    gen_kwargs={
                        "filepath": os.path.join(data_dir, "test_rationales.jsonl"),
                        "split": "test",
                    },
                ),
            ]

    def _generate_examples(self, filepath, split):
        """This function returns the examples in the raw (text) form."""
        with open(filepath, encoding="utf-8") as f:
            for id_, row in enumerate(f):
                data = json.loads(row)
                example = {
                    "text": data["text"],
                    "label": data["title"]
                }
                if self.config.name == "rationales":
                    example["foil"] = data["foil"]
                    example["words"] = data["words"]
                    example["rationales"] = data["rationales"]
                    example["contrastive_rationales"] = data["contrastive_rationales"]
                    example["annotations"] = data["annotations"]
                    example["contrastive_annotations"] = data["contrastive_annotations"]
                yield id_, example