---
language:
  - en
license: cc-by-4.0
task_categories:
  - text-classification
tags:
  - topic-modeling
  - topic-modelling
  - text-clustering
  - document-representation
size_categories:
  - 10K<n<100K
configs:
  - config_name: noow
    default: true
    data_files:
      - split: corpus
        path: data/noow.csv
  - config_name: bow
    data_files:
      - split: corpus
        path: data/bow.csv
---

# WOS-11967

This TopicNet preparation contains 11,965 retained scientific abstracts from the WOS-11967 release, with 33 observed integer labels serialized as strings. The source has 11,967 rows. The source README declares 35 categories and seven parents, but Y.txt contains 33 observed labels (0–32). Names for those 33 labels are not inferred from other WOS subsets.


## Population, labels and splits

Raw duplicate identity is body text after casefold and whitespace collapse. There are no raw duplicate groups or cross-label conflicts. Two identical lemma-sequence groups remove two rows; five near-duplicate pairs join families without removing their members. Exported labels use only Y.txt, the authoritative global class. The original source hierarchy has these ambiguities: global label 6 has child values 2 and 3, label 10 has child values 2 and 3, and parent/child pair (2,3) maps to global labels 10 and 11. The original YL1.txt/YL2.txt files are not copied as separate text files; their values, together with Y.txt, are included for every retained document in data/hierarchy.json. YL2 is a local child code interpreted with its parent YL1, not a globally unique class. Y values are not repaired and class names are not invented.

Family groups combine raw/lemma duplicate relations, source URL/DOI relations where present, and body five-token shingle Jaccard similarity ≥0.9. Groups are assigned by iterative multilabel stratification with seeds 20260824 and 20260825; all retained members of a group stay together. The resulting disjoint, exhaustive research splits are **fit 7,182, calibration 2,391, validation 2,392**. The Hugging Face split `corpus` contains every retained document; it is not `fit`.

Retained class counts:

| Label | Documents |
| --- | ---: |
| 0 | 298 |
| 1 | 301 |
| 2 | 300 |
| 3 | 300 |
| 4 | 299 |
| 5 | 300 |
| 6 | 406 |
| 7 | 426 |
| 8 | 389 |
| 9 | 397 |
| 10 | 392 |
| 11 | 393 |
| 12 | 387 |
| 13 | 371 |
| 14 | 402 |
| 15 | 346 |
| 16 | 420 |
| 17 | 386 |
| 18 | 410 |
| 19 | 423 |
| 20 | 384 |
| 21 | 441 |
| 22 | 449 |
| 23 | 309 |
| 24 | 357 |
| 25 | 368 |
| 26 | 321 |
| 27 | 262 |
| 28 | 351 |
| 29 | 340 |
| 30 | 401 |
| 31 | 335 |
| 32 | 301 |


## Representations and preprocessing

The source adapter reads aligned X/Y/YL1/YL2 lines with zero-based row indices. IDs are `wos11967_<six-digit-row-index>`. The CSV body is the saved X.txt line without its line terminator; no title is present.

Each UTF-8 CSV has exactly `id,raw_text,vw_text` columns and lexicographically ordered document IDs. `raw_text` contains the source abstract. The VW string starts with the document ID followed by only `|@word`. There is no title namespace. Labels are separate from the VW string. `noow` preserves normalized token occurrences in order; `bow` counts the same tokens independently per namespace as `token:count`, with lexicographic token order. The full prepared token sequences are included.

NOOW means occurrence order after preprocessing, not untouched or reversible prose. English preprocessing uses Python 3.11.15, NLTK 3.9.2 and NLTK-data commit `550b6625bcef1f2abff2ff770a5a0d272c9c6b2a`: WordPunctTokenizer, lowercase per original span, lowercased input token length at least 2 before POS tagging and lemmatization, averaged_perceptron_tagger_eng, WordNet POS mapping J/V/N/R with noun fallback, English stopword removal and `lemma.isalpha()` filtering. Output lemmas are not filtered by length and can contain one character. Names are not masked.

## Reproduce from source

The shared standalone code is in the [SciNews preprocessing folder](https://huggingface.co/datasets/TopicNet/SciNews/tree/main/preprocessing). Follow its [setup and code-download instructions](https://huggingface.co/datasets/TopicNet/SciNews/blob/main/preprocessing/README.md) to download only that folder at a full commit revision and create the locked Linux x86-64 environment with Python 3.11.15. Access follows the SciNews repository's permissions. No parent project checkout, BigARTM or TopicNet installation is needed.

From the directory containing the downloaded `preprocessing/` folder, choose writable source storage and fresh output and scratch paths. First obtain the original `ReadMe.txt` from the [official version 6 release](https://data.mendeley.com/datasets/9rw3vkcfy4/6) and place it at `/path/to/sources/wos/ReadMe.txt`, preserving its exact bytes. Its required size and SHA-256 are listed in the shared instructions and checked by the command:

```bash
python preprocessing/prepare.py download --dataset wos-11967 --sources-dir /path/to/sources
python preprocessing/prepare.py build --dataset wos-11967 \
  --sources-dir /path/to/sources \
  --output-dir /path/to/rebuilt/WOS-11967/data \
  --scratch-dir /path/to/scratch/wos-11967
python preprocessing/prepare.py verify --dataset wos-11967 \
  --data-dir /path/to/rebuilt/WOS-11967/data
```

The download validates the manually supplied notice and provides pinned `wos/WebOfScience.zip` and NLTK resources under the source directory. All three WOS selectors share these sources; the selector above builds WOS-11967. Build and verify run offline after setup. The output contains only `noow.csv`, `bow.csv`, `labels.json`, `splits.json` and `hierarchy.json`; verification compares them with the recorded SHA-256 values. Builds refuse to overwrite outputs. Preserve failed output and scratch files, and use fresh paths for a retry. The shared code's MIT license does not change the source-text rights below.

## Loading and metadata

The default configuration is `noow`; `bow` contains aggregated counts. Both expose the complete population as the Hugging Face split `corpus`. Labels and the stored fit/calibration/validation memberships are auxiliary files, joined by document ID. `data/labels.json` directly maps each string ID to a list of string labels; `data/splits.json` directly maps each split name to an ordered ID list. `data/hierarchy.json` directly maps each ID to {"Y": "...", "YL1": "...", "YL2": "..."}, with all source codes preserved as strings. Join these mappings by ID.

```python
from datasets import load_dataset

repo = "TopicNet/WOS-11967"
corpus = load_dataset(repo, "noow", split="corpus")
bow = load_dataset(repo, "bow", split="corpus")
```

For local use, obtain the complete package including actual Git LFS contents. Set `HF_DATASETS_OFFLINE=1` and `HF_HUB_OFFLINE=1` before Python starts and choose a writable cache directory:

```python
import json
from pathlib import Path
from datasets import load_dataset

package = Path("/path/to/WOS-11967")
corpus = load_dataset(str(package), "noow", split="corpus",
                      cache_dir="/writable/cache")
bow = load_dataset(str(package), "bow", split="corpus",
                   cache_dir="/writable/cache")
with (package / "data/labels.json").open(encoding="utf-8") as handle:
    labels = json.load(handle)
with (package / "data/splits.json").open(encoding="utf-8") as handle:
    splits = json.load(handle)
with (package / "data/hierarchy.json").open(encoding="utf-8") as handle:
    hierarchy = json.load(handle)
membership = {doc_id: split for split, ids in splits.items() for doc_id in ids}
annotated = corpus.map(lambda row: {"labels": labels[row["id"]],
                                   "source_hierarchy": hierarchy[row["id"]],
                                   "research_split": membership[row["id"]]})
fit = annotated.filter(lambda row: row["research_split"] == "fit")
```

For remote metadata, download `data/labels.json`, `data/splits.json` and `data/hierarchy.json` with `huggingface_hub.hf_hub_download(repo, filename, repo_type="dataset")`, then read the same direct mappings. Private repositories require authorized authentication.

## Related datasets and overlap

Related preparations are [TopicNet/WOS-5736](https://huggingface.co/datasets/TopicNet/WOS-5736) and [TopicNet/WOS-46985](https://huggingface.co/datasets/TopicNet/WOS-46985), alongside [TopicNet/WOS-11967](https://huggingface.co/datasets/TopicNet/WOS-11967). They are not independent datasets. Comparing original source abstracts after casefold and whitespace collapse gives 5,736 shared abstract identities between WOS-5736 and WOS-46985, 11,268 between WOS-11967 and WOS-46985, and 1,708 between WOS-5736 and WOS-11967. These counts describe source abstracts, not final retained populations. WOS-11967 is not entirely contained in WOS-46985 under this comparison. Splits are generated independently for each preparation; combining subsets for training and evaluation can leak shared documents across splits.

## Files and integrity

The package contains this card, `.gitattributes`, [SHA256SUMS](SHA256SUMS), two CSV representations and [labels](data/labels.json)/[splits](data/splits.json)/[hierarchy](data/hierarchy.json) metadata, plus the verbatim source README. Run `sha256sum -c SHA256SUMS` from the package root to check the listed files.

## Rights, limitations and citation

The [Mendeley release, version 6](https://data.mendeley.com/datasets/9rw3vkcfy4/6) declares CC BY 4.0; its DOI is [10.17632/9rw3vkcfy4.6](https://doi.org/10.17632/9rw3vkcfy4.6). The verbatim [SOURCE_README.txt](SOURCE_README.txt) retains Copyright (c) 2017 Kamran Kowsari and the permission notice requiring the copyright and permission notices in copies or substantial portions. Preserve those notices and cite the source. The release license is not a blanket guarantee that abstracts or other third-party content have no additional rights.

Source selection, historical coverage, class imbalance and automatic preprocessing affect the labels and topic distributions. Family grouping reduces known duplicate leakage, but some semantic duplicates may remain. The abstracts may contain personal information and outdated claims. Assess the corpus's suitability for your application before relying on it for decisions.

Kamran Kowsari, Donald E. Brown, Mojtaba Heidarysafa, Kiana Jafari Meimandi, Matthew S. Gerber and Laura E. Barnes (2017), [*HDLTex: Hierarchical Deep Learning for Text Classification*](https://arxiv.org/abs/1709.08267), IEEE ICMLA. DOI: [10.1109/ICMLA.2017.0-134](https://doi.org/10.1109/ICMLA.2017.0-134).

When citing this preparation, identify its retained population and repository revision as well as the source publication.