TurkLink EL — Turkish Wikidata Entity Linking
A complete pilot: learned mention detector, MiniLM bi-encoder, FAISS HNSW catalog,
BERTurk cross-encoder and validation-fitted confidence. Trained primarily on the
TurkLink corpus. Source code is included in source/; there is no remote model code.
from huggingface_hub import snapshot_download
root = snapshot_download("frcturus/turklink-el")
# Install the included library: pip install /path/to/downloaded/snapshot/source
from turklink import EntityLinker
linker = EntityLinker.from_pretrained(root, device="cpu")
print(linker.link("Apple İstanbul'da yeni mağazasını açtı."))
Install the library without cloning model weights:
pip install https://huggingface.co/frcturus/turklink-el/resolve/main/turklink_el-0.1.0-py3-none-any.whl
The full bundle includes a multi-GB index. For embeddings or reranking alone, download retriever, reranker, or mention detector. Allow roughly 10 GB of disk and at least 12 GB of RAM for the complete bundle. Linux CPU/CUDA and macOS CPU are supported; see the included source README for the macOS OpenMP workaround if your installed wheels conflict.
Actual example prediction
Apple→ Apple Inc. (Q312), confidence 0.577, span[0, 5)İstanbul→ İstanbul (Q406), confidence 0.853, span[6, 14)
Measured pipeline results
| Dataset | Mentions | Accuracy / micro-F1 | R@1 | R@5 | R@10 | R@32 | Catalog coverage |
|---|---|---|---|---|---|---|---|
| mewsli-9-tr | 5811 | 0.8226 | 0.7567 | 0.8845 | 0.9029 | 0.9248 | 0.9955 |
| mewsli-x-tr-dev | 262 | 0.9008 | 0.7863 | 0.9389 | 0.9542 | 0.9656 | 0.9962 |
| mewsli-x-tr-test | 1215 | 0.8593 | 0.7778 | 0.8963 | 0.9185 | 0.9342 | 0.9951 |
| test | 2000 | 0.8020 | 0.7500 | 0.8695 | 0.8900 | 0.9050 | 0.9985 |
| validation | 2000 | 0.8335 | 0.7725 | 0.8900 | 0.9105 | 0.9335 | 0.9980 |
These are measured results of this sampled pilot, not the TurkLink paper's results. Gold spans are supplied for the table. Accuracy equals micro-F1 because each mention has one gold and one predicted Q-ID. The catalog has 3,958,456 Q-IDs; missing gold entities count as misses. Candidate retrieval never inserts the correct answer.
Mewsli results, when present, are Turkish supervised domain transfer against this catalog, not the original zero-shot XTREME-R protocol/candidate descriptions. TurkLink end-to-end metrics are sparse-hyperlink proxies and are in the JSON reports; they must not be presented as exhaustive NER accuracy.
Raw-text end-to-end proxy
| Dataset | Windows | EL precision | EL recall | EL micro-F1 | Mention micro-F1 |
|---|---|---|---|---|---|
| test | 300 | 0.2286 | 0.6429 | 0.3372 | 0.4052 |
| validation | 300 | 0.2073 | 0.6559 | 0.3150 | 0.3669 |
These exact-span measurements include mention detection. Unannotated but correct mentions are counted as false positives by the sparse hyperlink reference.
Data and reproducibility
TurkLink, Akdaş & Tantuğ
(2026), DOI 10.1016/j.procs.2026.01.041.
Dataset revision: c8d7fe7bdd0ae934268d30ef64f6444940bdf6dc. Official split membership is preserved;
sampling is by document. The catalog additionally includes 449 missing training
entities recovered via Wikidata tr/mul/en label fallback; its frozen snapshot is
in wikidata-supplement.jsonl in the full bundle. No test labels were used for
this repair. Selected train/validation/test documents:
28,310 / 2,350 /
2,353. Seed: 42.
Full training settings, actual effective counts and validation histories are in
training.json for each component. The bundle contains source, manifests,
evaluation reports and a reproducible full-corpus recipe. The full recipe was
not run for this pilot. GPU floating-point kernels and parallel HNSW construction
may differ slightly across machines.
Limitations and license
Turkish case and apostrophe features are shared between training and inference; this is not a full morphological disambiguator. Mention supervision is weak and not exhaustive. The corpus, automatic annotations, translated descriptions and historical catalog can contain errors. Performance can degrade on new domains, new entities, common nouns and ambiguous mentions. Confidence is calibrated for a Q-ID given a supplied span, not for mention correctness, and may shift by domain. Abstention is threshold-based; this is not a trained NIL detector.
Corpus-derived artifacts and these checkpoints are distributed under CC BY-SA 4.0, with TurkLink attribution. Source code is Apache-2.0. Pretrained model licenses/notices remain applicable: MiniLM (Apache-2.0), BERTurk (MIT). See NOTICE and included licenses. No user text was sent to an external LLM during training.