Datasets:

Languages:
Vietnamese
ArXiv:
License:
uit_vsfc / uit_vsfc.py
holylovenia's picture
Upload uit_vsfc.py with huggingface_hub
cffd665 verified
raw
history blame
8.72 kB
# coding=utf-8
# Copyright 2022 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.
from pathlib import Path
from typing import Dict, List, Tuple
import datasets
from seacrowd.utils import schemas
from seacrowd.utils.configs import SEACrowdConfig
from seacrowd.utils.constants import Licenses, Tasks
_CITATION = """\
@inproceedings{van2018uit,
title={UIT-VSFC: Vietnamese students’ feedback corpus for sentiment analysis},
author={Van Nguyen, Kiet and Nguyen, Vu Duc and Nguyen, Phu XV and Truong, Tham TH and Nguyen, Ngan Luu-Thuy},
booktitle={2018 10th international conference on knowledge and systems engineering (KSE)},
pages={19--24},
year={2018},
organization={IEEE}
}
"""
_DATASETNAME = "uit_vsfc"
_DESCRIPTION = """\
This corpus consists of student feedback obtained from end-of-semester surveys at a Vietnamese university.
Feedback is classified into four possible topics: lecturer, curriculum, facility or others.
Feedback is also labeled as one of three sentiment polarities: positive, negative or neutral.
"""
_HOMEPAGE = "https://drive.google.com/drive/folders/1HooABJyrddVGzll7fgkJ6VzkG_XuWfRu"
_LANGUAGES = ["vie"]
_LICENSE = Licenses.UNKNOWN.value
_LOCAL = False
_URLS = {
"train": {
"sentences": "https://drive.google.com/uc?id=1nzak5OkrheRV1ltOGCXkT671bmjODLhP&export=download",
"sentiments": "https://drive.google.com/uc?id=1ye-gOZIBqXdKOoi_YxvpT6FeRNmViPPv&export=download",
"topics": "https://drive.google.com/uc?id=14MuDtwMnNOcr4z_8KdpxprjbwaQ7lJ_C&export=download",
},
"validation": {
"sentences": "https://drive.google.com/uc?id=1sMJSR3oRfPc3fe1gK-V3W5F24tov_517&export=download",
"sentiments": "https://drive.google.com/uc?id=1GiY1AOp41dLXIIkgES4422AuDwmbUseL&export=download",
"topics": "https://drive.google.com/uc?id=1DwLgDEaFWQe8mOd7EpF-xqMEbDLfdT-W&export=download",
},
"test": {
"sentences": "https://drive.google.com/uc?id=1aNMOeZZbNwSRkjyCWAGtNCMa3YrshR-n&export=download",
"sentiments": "https://drive.google.com/uc?id=1vkQS5gI0is4ACU58-AbWusnemw7KZNfO&export=download",
"topics": "https://drive.google.com/uc?id=1_ArMpDguVsbUGl-xSMkTF_p5KpZrmpSB&export=download",
},
}
_SUPPORTED_TASKS = [Tasks.SENTIMENT_ANALYSIS, Tasks.TOPIC_MODELING]
_SOURCE_VERSION = "1.0.0"
_SEACROWD_VERSION = "2024.06.20"
class UITVSFCDataset(datasets.GeneratorBasedBuilder):
"""This corpus consists of student feedback obtained from end-of-semester surveys at a Vietnamese university.
Feedback is classified into four possible topics: lecturer, curriculum, facility or others.
Feedback is also labeled as one of three sentiment polarities: positive, negative or neutral."""
SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
SEACROWD_VERSION = datasets.Version(_SEACROWD_VERSION)
SENTIMENT_LABEL_CLASSES = ["positive", "negative", "neutral"]
TOPIC_LABEL_CLASSES = ["lecturer", "training_program", "others", "facility"]
SEACROWD_SCHEMA_NAME = "text"
BUILDER_CONFIGS = [
SEACrowdConfig(
name=f"{_DATASETNAME}_sentiment_source",
version=SOURCE_VERSION,
description=f"{_DATASETNAME} source schema",
schema="source",
subset_id=_DATASETNAME,
),
SEACrowdConfig(
name=f"{_DATASETNAME}_topic_source",
version=SOURCE_VERSION,
description=f"{_DATASETNAME} source schema",
schema="source",
subset_id=_DATASETNAME,
),
SEACrowdConfig(
name=f"{_DATASETNAME}_sentiment_seacrowd_{SEACROWD_SCHEMA_NAME}",
version=SEACROWD_VERSION,
description=f"{_DATASETNAME} SEACrowd schema",
schema=f"seacrowd_{SEACROWD_SCHEMA_NAME}",
subset_id=_DATASETNAME,
),
SEACrowdConfig(
name=f"{_DATASETNAME}_topic_seacrowd_{SEACROWD_SCHEMA_NAME}",
version=SEACROWD_VERSION,
description=f"{_DATASETNAME} SEACrowd schema",
schema=f"seacrowd_{SEACROWD_SCHEMA_NAME}",
subset_id=_DATASETNAME,
),
]
DEFAULT_CONFIG_NAME = f"{_DATASETNAME}_source"
def _info(self) -> datasets.DatasetInfo:
if self.config.schema == "source":
features = datasets.Features(
{
"sentence": datasets.Value("string"),
"sentiment": datasets.ClassLabel(names=self.SENTIMENT_LABEL_CLASSES),
"topic": datasets.ClassLabel(names=self.TOPIC_LABEL_CLASSES),
}
)
elif self.config.name == f"{_DATASETNAME}_sentiment_seacrowd_{self.SEACROWD_SCHEMA_NAME}":
features = schemas.text_features(self.SENTIMENT_LABEL_CLASSES)
elif self.config.name == f"{_DATASETNAME}_topic_seacrowd_{self.SEACROWD_SCHEMA_NAME}":
features = schemas.text_features(self.TOPIC_LABEL_CLASSES)
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
homepage=_HOMEPAGE,
license=_LICENSE,
citation=_CITATION,
)
def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
data_dir = dl_manager.download(_URLS)
return [
datasets.SplitGenerator(
name=datasets.Split.TRAIN,
gen_kwargs={
"sentences_path": data_dir["train"]["sentences"],
"sentiments_path": data_dir["train"]["sentiments"],
"topics_path": data_dir["train"]["topics"],
"split": "train",
},
),
datasets.SplitGenerator(
name=datasets.Split.TEST,
gen_kwargs={
"sentences_path": data_dir["test"]["sentences"],
"sentiments_path": data_dir["test"]["sentiments"],
"topics_path": data_dir["test"]["topics"],
"split": "test",
},
),
datasets.SplitGenerator(
name=datasets.Split.VALIDATION,
gen_kwargs={
"sentences_path": data_dir["validation"]["sentences"],
"sentiments_path": data_dir["validation"]["sentiments"],
"topics_path": data_dir["validation"]["topics"],
"split": "dev",
},
),
]
def _generate_examples(self, sentences_path: Path, sentiments_path: Path, topics_path: Path, split: str) -> Tuple[int, Dict]:
"""Yields examples as (key, example) tuples."""
if self.config.schema == "source":
with open(sentences_path, encoding="utf-8") as sentences, open(sentiments_path, encoding="utf-8") as sentiments, open(topics_path, encoding="utf-8") as topics:
for key, (sentence, sentiment, topic) in enumerate(zip(sentences, sentiments, topics)):
yield key, {
"sentence": sentence.strip(),
"sentiment": int(sentiment.strip()),
"topic": int(topic.strip()),
}
elif self.config.name == f"{_DATASETNAME}_sentiment_seacrowd_{self.SEACROWD_SCHEMA_NAME}":
with open(sentences_path, encoding="utf-8") as sentences, open(sentiments_path, encoding="utf-8") as sentiments:
for key, (sentence, sentiment) in enumerate(zip(sentences, sentiments)):
yield key, {"id": str(key), "text": sentence.strip(), "label": int(sentiment.strip())}
elif self.config.name == f"{_DATASETNAME}_topic_seacrowd_{self.SEACROWD_SCHEMA_NAME}":
with open(sentences_path, encoding="utf-8") as sentences, open(topics_path, encoding="utf-8") as topics:
for key, (sentence, topic) in enumerate(zip(sentences, topics)):
yield key, {
"id": str(key),
"text": sentence.strip(),
"label": int(topic.strip()),
}