cs_reviews.py

# coding=utf-8
# Copyright 2023 Wojciech Kusa
#
# 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.
import json
from typing import List, Tuple, Dict

import bibtexparser
import datasets
import pandas as pd

from csmed.loader.bigbiohub import BigBioConfig
from csmed.loader.bigbiohub import Tasks
from csmed.loader.bigbiohub import text_features

_LANGUAGES = ["English"]
_PUBMED = True
_LOCAL = False

_CITATION = """\
@inproceedings{hannousse2022semi,
  title={A Semi-automatic Document Screening System for Computer Science Systematic Reviews},
  author={Hannousse, Abdelhakim and Yahiouche, Salima},
  booktitle={Mediterranean Conference on Pattern Recognition and Artificial Intelligence},
  pages={201--215},
  year={2022},
  organization={Springer}
}
"""

_DATASETNAME = "cs_reviews"
_DISPLAYNAME = "cs_reviews"

_DESCRIPTION = """\
Seven systematic reviews from the computer science domain.
"""

_HOMEPAGE = "https://github.com/hannousse/Semantic-Scholar-Evaluation/"
_LICENSE = "Unknown"

_URLS = {
    "reviews_content": "https://github.com/hannousse/Semantic-Scholar-Evaluation/archive/refs/heads/master.zip",
    "reviews_metadata": "https://github.com/aliromagnoli/Erasmus-Stage/archive/refs/heads/main.zip",
}

_SUPPORTED_TASKS = [Tasks.TEXT_CLASSIFICATION]

_SOURCE_VERSION = "1.0.0"
_BIGBIO_VERSION = "1.0.0"

_CLASS_NAMES = ["included", "excluded"]

REVIEWS = [
    "Alhammad-2018",
    "Ghasemi-2019",
    "Goulao-2016",
    "Guinea-2016",
    "Santos-2018",
    "Shahin-2017",
    "Yang-2016",
]


def prepare_cs_dataset(
    dataset_name: str,
    included_studies_repository: str,
    review_metadata_repository: str,
) -> pd.DataFrame:
    """
    :param dataset_name: name of the dataset
    :param included_studies_repository: path to the included studies repository
    :param review_metadata_repository: path to the review metadata repository
    :return:
    """

    included_studies_repository_path = f"{included_studies_repository}/Semantic-Scholar-Evaluation-master/Datasets for automatic screening of papers/{dataset_name}"

    included_file = f"{included_studies_repository_path}/{dataset_name}-Included.bib"
    excluded_file = f"{included_studies_repository_path}/{dataset_name}-Excluded.bib"
    review_metadata_file = f"{review_metadata_repository}/Erasmus-Stage-main/datasets/new_datasets/{dataset_name.lower()}.json"

    out_df = pd.DataFrame()
    for file, label in zip([included_file, excluded_file], [1, 0]):
        with open(file) as f:
            bib_database = bibtexparser.load(f)
            df = pd.DataFrame(bib_database.entries)
            df["label"] = label
            out_df = pd.concat([out_df, df], ignore_index=True)

    try:
        with open(review_metadata_file) as f:
            review_metadata = json.load(f)
    except json.decoder.JSONDecodeError:
        review_metadata = {}

    out_df["review_id"] = dataset_name
    return out_df


class CsReviewsDataset(datasets.GeneratorBasedBuilder):
    """Seven systematic reviews from the computer science domain."""

    SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
    BIGBIO_VERSION = datasets.Version(_BIGBIO_VERSION)

    BUILDER_CONFIGS = []

    for dataset_version in REVIEWS:
        BUILDER_CONFIGS.append(
            BigBioConfig(
                name=f"cs_reviews_{dataset_version}_source",
                version=SOURCE_VERSION,
                description=f"cs_reviews {dataset_version} source schema",
                schema="source",
                subset_id=f"cs_reviews_{dataset_version}",
            )
        )
        BUILDER_CONFIGS.append(
            BigBioConfig(
                name=f"cs_reviews_{dataset_version}_bigbio_text",
                version=BIGBIO_VERSION,
                description=f"cs_reviews {dataset_version} BigBio schema",
                schema="bigbio_text",
                subset_id=f"cs_reviews_{dataset_version}",
            )
        )

    BUILDER_CONFIGS.append(
        BigBioConfig(
            name="cs_reviews_all_source",
            version=SOURCE_VERSION,
            description="cs_reviews all source schema",
            schema="source",
            subset_id="cs_reviews_all",
        )
    )

    DEFAULT_CONFIG_NAME = "cs_reviews_all_source"

    def _info(self) -> datasets.DatasetInfo:

        if self.config.schema == "source":
            features = datasets.Features(
                {
                    "review_id": datasets.Value("string"),
                    "doc_id": datasets.Value("string"),
                    "title": datasets.Value("string"),
                    "abstract": datasets.Value("string"),
                    "label": datasets.ClassLabel(names=_CLASS_NAMES),
                }
            )
        elif self.config.schema == "bigbio_text":
            features = text_features
        else:
            raise ValueError(f"Unsupported schema {self.config.schema}")

        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            homepage=_HOMEPAGE,
            license=_LICENSE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager) -> List[datasets.SplitGenerator]:
        """Returns SplitGenerators."""
        data_dir = dl_manager.download_and_extract(_URLS)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={
                    "split": "train",
                    "data_dir": data_dir,
                },
            ),
        ]

    def _generate_examples(
        self,
        data_dir: dict[str, str],
        split: str,
    ) -> Tuple[int, Dict]:
        """Yields examples as (key, example) tuples."""

        review = "_".join(self.config.subset_id.split("_")[2:])

        uid = 0

        if review == "all":
            reviews = REVIEWS
        else:
            reviews = [review]

        df = pd.DataFrame()
        for r in reviews:
            df = pd.concat(
                [
                    df,
                    prepare_cs_dataset(
                        dataset_name=r,
                        included_studies_repository=data_dir["reviews_content"],
                        review_metadata_repository=data_dir["reviews_metadata"],
                    ),
                ]
            )

        for key, example in df.iterrows():

            title = example["title"]
            abstract = example["abstract"]
            label = example["label"]
            review_id = example["review_id"]
            document_id = example["ID"]
            text = f"{title}\n\n{abstract}"

            uid += 1
            if self.config.schema == "source":
                data = {
                    "review_id": review_id,
                    "doc_id": document_id,
                    "title": title,
                    "abstract": abstract,
                    "label": label,
                }
                yield str(uid), data

            elif self.config.schema == "bigbio_text":
                data = {
                    "id": str(uid),
                    "document_id": str(uid),
                    "text": text,
                    "labels": [label],
                }
                yield str(uid), data


if __name__ == "__main__":
    x = datasets.load_dataset(__file__, name="cs_reviews_all_source")
    print(x)