HealthVer is ready.

README.md

@@ -17,7 +17,7 @@ task_categories:
 - text-classification
 task_ids:
 - fact-checking
-pretty_name: CovidFact
+pretty_name: HealthVer
 dataset_info:
   features:
   - name: claim_id
@@ -36,17 +36,20 @@ dataset_info:
     sequence: int32
   splits:
   - name: train
-    num_bytes: 1547185
+    num_bytes: 9490482
+    num_examples: 5292
+  - name: validation
+    num_bytes: 1707997
     num_examples: 940
   - name: test
-    num_bytes: 523542
-    num_examples: 317
+    num_bytes: 1620257
+    num_examples: 903
   download_size: 3610222
-  dataset_size: 2070727
+  dataset_size: 12818736
 ---
 
 
-# Dataset Card for "covidfact_entailment"
+# Dataset Card for "healthver_entailment"
 
 ## Table of Contents
 
@@ -54,16 +57,15 @@ dataset_info:
   - [Dataset Summary](#dataset-summary)
 - [Dataset Structure](#dataset-structure)
   - [Data Fields](#data-fields)
-  - [Data Splits](#data-splits)
 
 ## Dataset Description
 
-- **Repository:** <https://github.com/asaakyan/covidfact>
+- **Repository:** <https://github.com/sarrouti/HealthVe>
 - **Point of Contact:** [David Wadden](mailto:[email protected])
 
 ### Dataset Summary
 
-COVID-FACT is a dataset of claims about COVID-19. For this version of the dataset, we follow the preprocessing from the MultiVerS modeling paper https://github.com/dwadden/multivers, verifying claims against abstracts of scientific research articles. Entailment labels and rationales are included.
+HealthVer is a dataset of public health claims, verified against scientific research articles. For this version of the dataset, we follow the preprocessing from the MultiVerS modeling paper https://github.com/dwadden/multivers, verifying claims against full article abstracts rather than individual sentences. Entailment labels and rationales are included.
 
 ## Dataset Structure
 
@@ -76,9 +78,3 @@ COVID-FACT is a dataset of claims about COVID-19. For this version of the datase
 - `abstract`: A list of `strings`, one for each sentence in the abstract.
 - `verdict`: The fact-checking verdict, a `string`.
 - `evidence`: A list of sentences from the abstract which provide evidence for the verdict.
-
-### Data Splits
-
-|      |train|validation|
-|------|----:|---------:|
-|claims| 919 |       340|

healthver_entailment.py

@@ -7,19 +7,18 @@ import json
 
 
 _CITATION = """\
-@article{Saakyan2021COVIDFactFE,
-  title={COVID-Fact: Fact Extraction and Verification of Real-World Claims on COVID-19 Pandemic},
-  author={Arkadiy Saakyan and Tuhin Chakrabarty and Smaranda Muresan},
-  journal={ArXiv},
-  year={2021},
-  volume={abs/2106.03794},
-  url={https://api.semanticscholar.org/CorpusID:235364036}
+@inproceedings{Sarrouti2021EvidencebasedFO,
+    title={Evidence-based Fact-Checking of Health-related Claims},
+    author={Mourad Sarrouti and Asma Ben Abacha and Yassine Mrabet and Dina Demner-Fushman},
+    booktitle={Conference on Empirical Methods in Natural Language Processing},
+    year={2021},
+    url={https://api.semanticscholar.org/CorpusID:244119074}
 }
 """
 
 
 _DESCRIPTION = """\
-COVID-FACT is a dataset of claims about COVID-19. For this version of the dataset, we follow the preprocessing from the MultiVerS modeling paper https://github.com/dwadden/multivers, verifying claims against abstracts of scientific research articles. Entailment labels and rationales are included.
+HealthVer is a dataset of public health claims, verified against scientific research articles. For this version of the dataset, we follow the preprocessing from the MultiVerS modeling paper https://github.com/dwadden/multivers, verifying claims against full article abstracts rather than individual sentences. Entailment labels and rationales are included.
 """
 
 _URL = "https://scifact.s3.us-west-2.amazonaws.com/longchecker/latest/data.tar.gz"
@@ -29,8 +28,8 @@ def flatten(xss):
     return [x for xs in xss for x in xs]
 
 
-class CovidFactEntailmentConfig(datasets.BuilderConfig):
-    """builderconfig for covidfact"""
+class HealthVerEntailmentConfig(datasets.BuilderConfig):
+    """builderconfig for healthver"""
 
     def __init__(self, **kwargs):
         """
@@ -38,19 +37,19 @@ class CovidFactEntailmentConfig(datasets.BuilderConfig):
         Args:
             **kwargs: keyword arguments forwarded to super.
         """
-        super(CovidFactEntailmentConfig, self).__init__(
+        super(HealthVerEntailmentConfig, self).__init__(
             version=datasets.Version("1.0.0", ""), **kwargs
         )
 
 
-class CovidFactEntailment(datasets.GeneratorBasedBuilder):
-    """TODO(covidfact): Short description of my dataset."""
+class HealthVerEntailment(datasets.GeneratorBasedBuilder):
+    """TODO(healthver): Short description of my dataset."""
 
-    # TODO(covidfact): Set up version.
+    # TODO(healthver): Set up version.
     VERSION = datasets.Version("0.1.0")
 
     def _info(self):
-        # TODO(covidfact): Specifies the datasets.DatasetInfo object
+        # TODO(healthver): Specifies the datasets.DatasetInfo object
 
         features = {
             "claim_id": datasets.Value("int32"),
@@ -75,7 +74,6 @@ class CovidFactEntailment(datasets.GeneratorBasedBuilder):
             # builder.as_dataset.
             supervised_keys=None,
             # Homepage of the dataset for documentation
-            homepage="https://scifact.apps.allenai.org/",
             citation=_CITATION,
         )
 
@@ -90,18 +88,20 @@ class CovidFactEntailment(datasets.GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
-        # TODO(scifact): Downloads the data and defines the splits
+        # TODO(healthver): Downloads the data and defines the splits
         # dl_manager is a datasets.download.DownloadManager that can be used to
         # download and extract URLs
         archive = dl_manager.download(_URL)
         for path, f in dl_manager.iter_archive(archive):
             # The claims are too similar to paper titles; don't include.
-            if path == "data/covidfact/corpus_without_titles.jsonl":
+            if path == "data/healthver/corpus.jsonl":
                 corpus = self._read_tar_file(f)
                 corpus = {x["doc_id"]: x for x in corpus}
-            elif path == "data/covidfact/claims_train.jsonl":
+            elif path == "data/healthver/claims_train.jsonl":
                 claims_train = self._read_tar_file(f)
-            elif path == "data/covidfact/claims_test.jsonl":
+            elif path == "data/healthver/claims_dev.jsonl":
+                claims_validation = self._read_tar_file(f)
+            elif path == "data/healthver/claims_test.jsonl":
                 claims_test = self._read_tar_file(f)
 
         return [
@@ -114,6 +114,15 @@ class CovidFactEntailment(datasets.GeneratorBasedBuilder):
                     "split": "train",
                 },
             ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                # These kwargs will be passed to _generate_examples
+                gen_kwargs={
+                    "claims": claims_validation,
+                    "corpus": corpus,
+                    "split": "validation",
+                },
+            ),
             datasets.SplitGenerator(
                 name=datasets.Split.TEST,
                 # These kwargs will be passed to _generate_examples