initial

README.md

@@ -0,0 +1,171 @@
+---
+license: cc-by-nc-sa-4.0
+task_categories:
+- text-classification
+- token-classification
+language:
+- en
+multilinguality:
+- monolingual
+size_categories:
+- n<1K
+tags:
+- causality
+pretty_name: BioCause
+configs:
+- config_name: causality detection
+  data_files:
+  - split: train
+    path: causality-detection/train.parquet
+  - split: validation
+    path: causality-detection/dev.parquet
+  - split: test
+    path: causality-detection/test.parquet
+  features:
+  - name: index
+    dtype: string
+  - name: text
+    dtype: string
+  - name: label
+    dtype:
+      class_label:
+        names:
+          '0': uncausal
+          '1': causal
+- config_name: causal candidate extraction
+  data_files:
+  - split: train
+    path: causal-candidate-extraction/train.parquet
+  - split: validation
+    path: causal-candidate-extraction/dev.parquet
+  - split: test
+    path: causal-candidate-extraction/test.parquet
+  features:
+  - name: index
+    dtype: string
+  - name: text
+    dtype: string
+  - name: entity
+    sequence:
+      sequence: int32
+- config_name: causality identification
+  data_files:
+  - split: train
+    path: causality-identification/train.parquet
+  - split: validation
+    path: causality-identification/dev.parquet
+  - split: test
+    path: causality-identification/test.parquet
+  features:
+  - name: index
+    dtype: string
+  - name: text
+    dtype: string
+  - name: relations
+    list:
+    - name: relationship
+      dtype:
+        class_label:
+          names:
+            '0': no-rel
+            '1': causal
+    - name: first
+      dtype: string
+    - name: second
+      dtype: string
+train-eval-index:
+- config: causality detection
+  task: text-classification
+  task_id: text_classification
+  splits:
+    train_split: train
+    eval_split: test
+  col_mapping:
+    text: text
+    label: label
+  metrics:
+  - type: accuracy
+  - type: precision
+  - type: recall
+  - type: f1
+- config: causal candidate extraction
+  task: token-classification
+  task_id: token_classification
+  splits:
+    train_split: train
+    eval_split: test
+  metrics:
+  - type: accuracy
+  - type: precision
+  - type: recall
+  - type: f1
+- config: causality identification
+  task: text-classification
+  task_id: text_classification
+  splits:
+    train_split: train
+    eval_split: test
+  metrics:
+  - type: accuracy
+  - type: precision
+  - type: recall
+  - type: f1
+---
+
+> [!NOTE]
+> This repository integrates the BioCause corpus into hf datasets. Please find the original dataset
+> [here](https://github.com/Luisiglm/BioCause). The data is sourced from the
+> [CREST](https://github.com/phosseini/CREST) aggregation. Please see the [citations](#citations) at the end of this README.
+
+## Dataset Description
+
+- **Homepage:** https://github.com/Luisiglm/BioCause
+- **Paper:** [BioCause: Annotating and Analysing Causality in the Biomedical Domain](https://doi.org/10.1186/1471-2105-14-2)
+
+# Usage
+## Causality Detection
+```py
+from datasets import load_dataset
+dataset = load_dataset("thagen/BioCause", "causality detection")
+```
+
+## Causal Candidate Extraction
+```py
+from datasets import load_dataset
+dataset = load_dataset("thagen/BioCause", "causal candidate extraction")
+```
+
+## Causality Identification
+```py
+from datasets import load_dataset
+dataset = load_dataset("thagen/BioCause", "causality identification")
+```
+
+# Citations
+
+The BioCause paper by [Mihaila et al., 2013](https://doi.org/10.1186/1471-2105-14-2):
+```bib
+@article{mihaila:2013,
+  title = {{{BioCause}}: {{Annotating}} and Analysing Causality in the Biomedical Domain},
+  shorttitle = {{{BioCause}}},
+  author = {Mihaila, Claudiu and Ohta, Tomoko and Pyysalo, Sampo and Ananiadou, Sophia},
+  year = {2013},
+  journal = {BMC Bioinform.},
+  volume = {14},
+  pages = {2},
+  doi = {10.1186/1471-2105-14-2}
+}
+```
+
+CREST by [Hosseini et al., 2021](https://arxiv.org/abs/2103.13606) &mdash; whose aggregation we used to source the BioCause data:
+```bib
+@article{hosseini:2021,
+  title = {Predicting {{Directionality}} in {{Causal Relations}} in {{Text}}},
+  author = {Hosseini, Pedram and Broniatowski, David A. and Diab, Mona T.},
+  year = {2021},
+  journal = {CoRR},
+  volume = {abs/2103.13606},
+  eprint = {2103.13606},
+  archiveprefix = {arXiv}
+}
+```

conversion_script.py

@@ -0,0 +1,28 @@
+#!/usr/bin/env python3
+
+"""
+Run this script as ./conversion_script.py to convert the BioCause entries from
+the CREST v2 aggregation file to HF-compatible parquet files.
+"""
+
+# 1) Install dependencies:
+#    pip install git+https://github.com/TheMrSheldon/causality-toolkit.git
+# 2) Source file (fetched automatically via pandas):
+#  - https://raw.githubusercontent.com/phosseini/CREST/master/data/crest_v2.xlsx
+
+from pathlib import Path
+
+from ctk.data.constants import Task
+from ctk.data.conversion import CREST2HF, CRESTSource
+
+converter = CREST2HF(
+    "https://raw.githubusercontent.com/phosseini/CREST/master/data/crest_v2.xlsx",
+    Path.cwd(),
+    prefix="biocause",
+    filters={"source": CRESTSource.BioCause},
+)
+
+for split in ["train", "dev", "test"]:
+    converter.convert(Task.CausalityDetection, split)
+    converter.convert(Task.CausalCandidateExtraction, split)
+    converter.convert(Task.CausalityIdentification, split)