histlearn/notas-comunidade-ptbr

Community Notes BR: Enriched Portuguese Dataset for NLP and Text Mining

Community Notes BR is a curated Portuguese-language subset of X Community Notes enriched for Natural Language Processing (PLN), text mining, and research on collaborative misinformation moderation. It combines note-level metadata, topic and macrotheme labels, named entities, source-domain extraction, Matrix Factorization scoring fields, an optional Universal Dependencies syntax layer, and automatic quality-audit tables.

The dataset does not redistribute tweet text. It contains Community Notes text and identifiers released by the official Community Notes public data program; use tweetId for hydration when your own legal and API access conditions allow it.

Associated preprint: Davi Machado da Rocha, Community Notes no Brasil: consenso, polarização e aceitabilidade algorítmica (SciELO Preprints, 2026). https://doi.org/10.1590/SciELOPreprints.15996

Snapshot: official Community Notes public data from 2026-04-07. Upstream source: https://communitynotes.x.com/guide/en/under-the-hood/download-data Coverage: 2021-10-13 to 2026-04-05. Notes: 142,448 Portuguese notes, selected with fastText lid.176 confidence >= 0.80. Tweets: 90,822 unique tweet IDs. Authors: 25,483 anonymized Community Notes participant IDs. Entities: 421,912 post-processed entity rows, with 21 observed canonical entity types. A structural correction released in August 2026 removed 88,440 rows: model spans that fell inside URL_DOMINIO spans (already captured by the deterministic layer) and spans whose offsets did not recover their own text. See Structural correction below. Topics: 898 valid fine-grained topics, plus BERTopic outliers (topic_id = -1), organized into 38 macrothemes. Syntax layer: 292,275 automatic UD sentences and 5,599,679 token rows. Current documentation state: updated after the entity post-processing and quality-audit rounds of June 2026 and the publication of the associated SciELO preprint in August 2026.

Resumo em português

Este é um subconjunto curado em língua portuguesa das Community Notes da X (critério linguístico, por fastText; o recorte não geolocaliza autores nem publicações), enriquecido para pesquisa em PLN, mineração de texto e moderação colaborativa. Ele inclui 142.448 notas, 90.822 tweets distintos, 421.912 entidades nomeadas pós-processadas, 898 tópicos válidos, 38 macrotemas, campos de pontuação por fatoração matricial e uma camada opcional de sintaxe Universal Dependencies. A versão atual também publica tabelas de auditoria em quality_audit/, criadas para tornar explícitos os problemas e limites de offsets de entidades, alinhamento com UD e integridade entre notas, tópicos e entidades.

As anotações são automáticas e não passaram por revisão humana linha a linha. As tabelas de auditoria devem ser usadas como parte do dataset, especialmente quando a análise depende de spans, offsets ou alinhamento entre entidades e sentenças.

---

Files

File or folder	Rows	Size	Granularity / role
notes_pt.parquet	142,448	42.4 MB	one row per Portuguese Community Note
entities.parquet	421,912	57.2 MB	one row per extracted entity
quality_audit/entities_removed_2026-08.parquet	88,440	11.0 MB	rows removed by the structural correction, with the rule that removed each
topic_metadata.parquet	898	78.3 KB	one row per valid BERTopic cluster
macrotheme_metadata.parquet	38	9.4 KB	one row per macrotheme
DATA_DICTIONARY.md	-	-	explicit data dictionary with column-level descriptions
taxonomy.json	-	5.0 KB	entity extraction taxonomy and thresholds
syntax_ud_bosque/sentences/*.parquet	292,275	22.8 MB	automatic UD sentence spans
syntax_ud_bosque/tokens/*.parquet	5,599,679	89.7 MB	automatic UD token-level annotations
syntax_ud_bosque/notes_udpipe_bosque.conllu.zst	-	42.3 MB	compressed CoNLL-U export
quality_audit/	-	90.7 MB	automatic validation tables
quality_audit/human_audit/	-	0.6 MB	single-annotator human audit: gold decisions, topic answer key, sampling manifest and analysis scripts
cross_modal_validation/	-	64.6 KB	macrotheme/entity/domain consistency previews
source_code/	-	2.3 MB	sanitized construction and analysis notebooks

Primary join key: noteId between notes_pt.parquet, entities.parquet, the UD files, and the audit tables. tweetId is kept for tweet hydration and cross-reference, but tweet text is not included.

The Dataset Viewer is intentionally limited to the main analysis tables and the optional UD syntax layer. The automatic quality-audit tables remain available under quality_audit/ in the repository files.

The core dataset is released under CC0. The optional syntax_ud_bosque/ layer was generated with UDPipe using the portuguese-bosque-ud-2.17-251125 model, whose license is reported by the API as CC BY-NC-SA; see syntax_ud_bosque/udpipe_bosque_metadata.json.

---

Schema

For a complete column-level data dictionary, including keys, Parquet types, controlled values and interpretation notes, see DATA_DICTIONARY.md. The summary below highlights the main analysis tables.

notes_pt.parquet

Column	Description
noteId	Primary key, the official Community Notes note identifier.
tweetId	The tweet the note refers to.
note_author_participant_id	Anonymized Community Notes participant ID.
created_at, year_month	Note creation time in UTC and monthly convenience key.
summary	Note text as published in the Community Notes data.
summary_clean	Whitespace-normalized and URL-stripped text used by embeddings and UD parsing.
classification_algoritmo	Original Community Notes note classification, such as MISINFORMED_OR_POTENTIALLY_MISLEADING or NOT_MISLEADING.
idioma, confianca_idioma	fastText language label and confidence. All retained rows have idioma = pt and confidence >= 0.80.
status_final, consenso	Latest Community Notes status and compact status class (CRH, CRNH, NMR, Outro).
n_ratings, n_notas_no_tweet	Rating count for the note and number of notes attached to the same tweet.
coreNoteIntercept, coreNoteFactor1	Re-estimated with the official Matrix Factorization scoring code over the Portuguese subgraph only. Not the platform's official global scores. Null when a note has insufficient ratings for scoring.
topic_id, topic_name, contexto_central, categoria_ampla	Fine-grained topic ID and LLM-generated topic metadata. topic_id = -1 denotes BERTopic outliers.
macrotheme_id, macrotheme_label	Hierarchical macrotheme ID and LLM-generated label.
stratum	Hydration sample stratum (top_notas, bat_signals, aleatorio); null for unsampled notes.

entities.parquet

Column	Description
entity_id	Unique ID of the extraction, formatted as <row_id>_<j>. No duplicates are present in the current release.
noteId, tweetId	Foreign keys to the note and tweet.
texto_entidade	Surface form extracted from the note/extraction context.
texto_canonico	Normalized form after lowercasing, unidecode and dictionary-based canonicalization where applicable.
tipo_entidade	Canonical Portuguese entity type. The current table has 21 observed types; taxonomy.json documents extraction rules and regex aliases.
label_modelo	Original GLiNER label or regex extractor name.
inicio, fim	Generated character offsets from extraction. Treat them as automatic spans, not as gold offsets against summary_clean; use quality_audit/entity_offset_audit.parquet before span-level analysis.
score	GLiNER confidence or high-confidence regex score.
fonte_extracao	Extraction source: GLiNER or one of the regex extractors for URLs, dates, percentages, monetary values, legal references or court-process numbers.
contexto	Local context window around the entity.
papel_no_texto	mencao or fonte_ou_evidencia.
dominio	Canonicalized domain for URL/domain entities.
topic_id, macrotheme_id	Denormalized topic and macrotheme IDs copied from the source note for analysis convenience.

syntax_ud_bosque/

The optional syntax layer is automatic UDPipe output over notes_pt.summary_clean.

Folder	Description
sentences/part-*.parquet	Sentence-level spans, text and sentence IDs.
tokens/part-*.parquet	Token-level CoNLL-U fields plus noteId and sentence metadata.
notes_udpipe_bosque.conllu.zst	Compressed CoNLL-U export with note-level comments.

quality_audit/

File	Rows	Description
entity_offset_audit.parquet	510,352	Checks whether texto_entidade and offsets validate against notes_pt.summary_clean.
ud_sentence_tokenization_audit.parquet	292,275	Checks whether UD sentence text can be located in summary_clean, and whether token counts and token IDs are coherent.
entity_ud_alignment.parquet	510,352	Aligns entities with UD sentence spans when offsets validate.
notes_topic_integrity_audit.parquet	142,448	Validates note-level topic and macrotheme references.
entities_topic_integrity_audit.parquet	510,352	Checks whether entity topic/macrotheme IDs match their source note.
quality_audit_summary.parquet / .csv	27	Compact count table for all audit statuses.

quality_audit/human_audit/ holds the human-audit artifacts (JSON and Python, not Parquet). See Human audit (single annotator) under Quality Assessment.

---

Construction Methodology

1. Community Notes ETL

The five official Community Notes tables (notes, noteRatings, noteStatusHistory, userEnrollment, batSignals) were downloaded by snapshot date, hash-verified, checked for schema drift across shards and materialized in DuckDB. Bronze tables preserve the source fields as strings, silver tables add typed columns and snake_case names, and gold tables export compressed Parquet files.

2. Portuguese filter

Each note's summary_clean was classified with fastText lid.176.bin. Notes with idioma = pt and confianca_idioma >= 0.80 were retained, yielding 142,448 notes. A 300-note stratified sample across confidence bands was visually inspected and looked correct, but this was a quick sanity check rather than a formal human evaluation.

3. Topic and macrotheme modeling

Notes were embedded with intfloat/multilingual-e5-large-instruct using a Portuguese instruction prompt oriented toward the factual matter under dispute. BERTopic was fitted with UMAP and HDBSCAN parameters selected by a DBCV grid search. Outliers were reduced conservatively by embedding similarity.

Each valid topic was labeled by Marco-Mini-Instruct with grammar-constrained JSON output. Macrothemes were obtained by hierarchical aggregation over topic embeddings and labeled in a second LLM pass using topic descriptions rather than only keywords. A procedural-content filter was used to separate substantive topics from notes about Community Notes mechanics, satire without a target, and similar meta-platform content.

4. Named entity recognition

Entity extraction combines:

• GLiNER (urchade/gliner_multi-v2.1) with a Portuguese taxonomy for people, political actors, parties, public bodies, courts, security forces, media outlets, platforms, public programs, organizations, locations and political events.
• Regex extractors for structured patterns such as domains/URLs, dates, percentages, monetary values, legal references and judicial process numbers.
• Canonical dictionaries for public programs, public agencies, parties, platforms and recurring media/domain variants.
• Overlap resolution favoring higher-precision regex outputs and canonical types.

The current release includes automatic post-processing after the first extraction. It removed generic role nouns and extraction markers, corrected known media/platform/party type conflicts, applied a POS/UD-assisted filter to single-token false positives, and refined the separation among PESSOA, PESSOA_PUBLICA and ATOR_POLITICO. The post-processing was conservative and automatic; it was not a human line-by-line adjudication.

Post-processing history for entities.parquet:

Stage	Rows	Notes
Initial extraction	536,373	GLiNER + regex + canonicalization.
Structural/entity cleanup	521,992	Removed extraction markers, malformed standalone URL tokens, generic role nouns and a few systematic type conflicts.
POS/UD-assisted cleanup	510,352	Removed 11,640 single-token false positives using morphology/POS and lexicon rules.
Person-type refinement	510,352	Reclassified 1,282 public/personal names as ATOR_POLITICO and 126 non-political public figures away from ATOR_POLITICO.

Current entities.parquet SHA-256: 6c4f87b6d41d839b610043668d47c5ba95b7a3bf42ebb4b2ba786a0e5be53388.

5. Matrix Factorization scoring

The official Twitter/X communitynotes Matrix Factorization scoring code was run on the PT subset with minimal compatibility patches for the smaller note universe. The release includes coreNoteIntercept and coreNoteFactor1, restricted to Portuguese notes. These fields are null for notes with too few ratings.

Note on the audit tables. The parquet files under quality_audit/ were computed against the pre-correction revision and therefore still report 510,352 rows. They remain published as the evidence trail that motivated the correction; for the current entity layer, use entities.parquet (421,912 rows).

6. Syntax and audit layers

The UD syntax layer was generated automatically with UDPipe Bosque over summary_clean. The quality_audit/ tables were added after the post-processing rounds to document offset validity, entity-to-UD alignment and topic/entity referential integrity. These tables are intended to make downstream filtering explicit instead of silently hiding uncertain spans.

---

Source Code, Reproducibility and Exploration

The source_code/ folder contains sanitized notebooks used to construct and analyze the dataset:

Notebook	Purpose
01_etl_snapshot.ipynb	upstream snapshot ETL and typed analytical tables
02_portuguese_filter.ipynb	Portuguese-language filtering and ratings/status joins
03_topic_modeling_macrothemes.ipynb	embeddings, topic modeling, LLM labels and macrothemes
04_named_entity_recognition.ipynb	GLiNER, regex extractors and canonical dictionaries
05_matrix_factorization_scoring.ipynb	reproduction of Matrix Factorization scoring fields
06_helpfulness_classification.ipynb	exploratory helpfulness/consensus classification experiments
07_udpipe_bosque_enrichment_colab.ipynb	optional Colab pipeline for UDPipe Bosque annotations

Notebook outputs and embedded credentials were removed before publication. Full reproduction requires the upstream Community Notes data, local credentials where needed, optional X API access for hydration, and sufficient CPU/GPU resources for the modeling stages. The notebooks are transparent research code, not a one-command production package.

Public exploration interfaces:

• Static project catalog: https://communitynotesbr.netlify.app/
• Argument/entity explorer: https://explorador-argumentos.netlify.app/

---

Quantitative Description

Volume by status

consenso	Source status	n_notes	%
NMR	NEEDS_MORE_RATINGS	116,741	82.0
CRH	CURRENTLY_RATED_HELPFUL	15,857	11.1
CRNH	CURRENTLY_RATED_NOT_HELPFUL	4,423	3.1
Outro	null or unmapped	5,427	3.8

Topic coverage

Status	n_notes	%
Valid topic and macrotheme	94,889	66.6
BERTopic outlier (topic_id = -1)	47,162	33.1
Missing topic ID	397	0.3

Entities by type

tipo_entidade	n
URL_DOMINIO	174,582
PESSOA	58,543
LOCAL	49,022
ORGANIZACAO	46,835
VEICULO_MIDIA	43,292
ATOR_POLITICO	33,617
PLATAFORMA_DIGITAL	27,889
ORGAO_PUBLICO	16,740
DATA	13,292
PARTIDO	9,016
ESTATISTICA	7,519
EVENTO_POLITICO	7,347
PROGRAMA_PUBLICO	5,732
VALOR_MONETARIO	4,224
ORGAO_SEGURANCA	4,196
ORGAO_JUDICIARIO	2,828
LEI_NORMA	1,789
OPERACAO_POLICIAL	1,170
PESSOA_PUBLICA	1,159
PROCESSO_JUDICIAL	888
FONTE_CITADA	630

126,527 notes (89.1%) have at least one entity. The current release has a mean of 3.33 entities per annotated note and 2.97 entities per note overall. Person mentions (PESSOA + ATOR_POLITICO + PESSOA_PUBLICA) total 78,196.

Missing data

Field family	Missingness	Interpretation
coreNoteIntercept / coreNoteFactor1	27.3%	Notes with too few ratings are not scored by the MF algorithm.
topic_id	397 rows (0.3%)	Notes outside the topic-model run.
topic_name, contexto_central, categoria_ampla	47,559 rows (33.4%)	BERTopic outliers plus missing topic IDs.
macrotheme_id, macrotheme_label	47,559 rows (33.4%)	Same scope as topic labels.
status_final	5,427 rows (3.8%)	Status empty or unmapped in the snapshot.
stratum	138,393 rows (97.2%)	Only 4,055 notes were selected for tweet-hydration sampling.

---

Quality Assessment

No exhaustive human validation was performed. The release provides automatic checks, transparent failure-mode reporting, and one stratified human audit of a sample, described immediately below.

Human audit (single annotator)

A stratified sample was audited by hand by one annotator, who is also the dataset author. The decisions, the answer key, the sampling manifest and the analysis scripts are published under quality_audit/human_audit/, against canonical revision d4f438b3acba66a56b0124f5ebc57e2a48dd76a6 and sampling seed 42.

Layer	n	Strict sample precision	Wilson 95% CI
Model layer (GLiNER)	659	80.6%	[77.4, 83.4]
Deterministic layer (regex)	357	99.4%	[98.0, 99.8]
Combined sample	1,016	87.2%	[85.0, 89.1]

Of the 128 model-layer errors, 120 are over-extractions -- spans that should not exist -- and 8 are type errors; no boundary or canonical-form error was found. Precision varies sharply across types: ATOR_POLITICO scored 114/114, FONTE_CITADA 16.1%. With n around 30 per type the interval exceeds 13 percentage points, so per-type values are indicative rather than estimates. The deterministic subsample was not stratified by rule route: 314 of its 357 cases are URL_DOMINIO, so its aggregate must not be read as equal validation of all six routes.

A topic-intruder task over 148 topics stratified by volume decile reached 96.6% [92.3, 98.5] detection against 25% chance. The intruder was drawn from a different macrotheme and topic glosses were visible, so the result measures broad cross-macrotheme separability under those cues, not blind fine-topic coherence.

What this is not. There is a single annotator, who is also the author: this is self-assessment, not independent audit, and no inter-annotator reliability is estimated. The design oversamples rare types, so the aggregate is not a prevalence-weighted corpus estimate. The audit starts from already-extracted spans, so it measures precision and says nothing about recall.

Reproduce from quality_audit/human_audit/:

python analyze_entities.py
python analyze_topics.py gold_topics_cross_macro_single_annotator.json

Deterministic checks

The ETL uses file hashes, schema-drift checks, referential integrity checks and snapshot-to-snapshot comparisons. In the current release, entities.entity_id has zero duplicates and every entity row matches the topic and macrotheme IDs of its source note.

Entity offset audit

quality_audit/entity_offset_audit.parquet checks entities.parquet against notes_pt.summary_clean.

Audit status	n_entities
exact	238,770
normalized_exact	45
mismatch	11,384
out_of_bounds	260,153

Important interpretation: many offsets are not relative to summary_clean. Several URL/evidence spans appear to have been extracted from a richer extraction text where links were inserted into or appended to the note. The audit therefore flags these rows instead of rewriting offsets automatically. For span-level tasks, prefer rows with exact or normalized_exact, or inspect the shifted/repeated-text categories in the audit table.

Entity-to-UD alignment

Alignment status	n_entities
aligned_to_ud_sentence	237,738
entity_offset_not_validated_against_summary_clean	271,537
entity_crosses_or_exceeds_sentence_span	626
no_sentence_covering_offset	257
no_valid_ud_sentence_span	194

This alignment is useful for POS/morphology-aware filtering and syntactic analyses, but it should not be treated as a gold span annotation.

UD sentence/token audit

Audit	Status	n_sentences
Sentence location	exact_in_order	292,149
Sentence location	not_found	126
Token row count	matches_n_conllu_rows	292,275
Token ID sequence	sequential	292,275

Topic integrity

Audit	Status	n
Note topic reference	topic_exists	94,889
Note topic reference	no_topic_cluster	47,162
Note topic reference	missing_topic_id	397
Entity topic reference	topic_matches_note	510,352
Entity macrotheme reference	macrotheme_matches_note	510,352

Cross-modal validation

The cross_modal_validation/ folder contains compact tables comparing macrotheme labels with empirical entity-type and domain signatures. These are diagnostic previews rather than formal validation. They are useful for spotting residual procedural topics, overly broad macrothemes and labels whose entity/domain distribution is weakly concentrated.

---

Limitations

• No exhaustive human adjudication. Topic labels, entity types, offsets and UD parses are automatic, and the post-processing rules are conservative but still rule-based. Only a stratified sample was checked by hand, by a single annotator who is also the author (quality_audit/human_audit/): it estimates precision, carries no inter-annotator reliability and says nothing about recall.
• Tweet text is not redistributed. This follows X terms and the Community Notes public-data design. Use tweetId for hydration when permitted.
• Snapshot-bound status. Community Notes statuses evolve. The fields here reflect the 2026-04-07 snapshot.
• Portuguese filter tradeoff. The fastText threshold favors precision and may exclude short, code-mixed or unusual Portuguese notes.
• Topic outliers. 33.4% of notes have no valid topic/macrotheme label because they are BERTopic outliers or were outside the topic run.
• Entity offsets require care. Offsets are not uniformly valid against summary_clean; consult quality_audit/ for span-level work.
• Automatic syntax layer. UD annotations may contain tokenization, lemmatization, POS-tagging, morphological or dependency errors, especially in informal text, sarcasm, named entities and short notes.
• NER taxonomy is domain-specific. Types such as ATOR_POLITICO, PESSOA_PUBLICA, ORGAO_PUBLICO and VEICULO_MIDIA are designed for Brazilian Community Notes research and may not transfer directly to generic NER benchmarks.

---

Structural correction (August 2026)

A structural audit of the entity layer, run while preparing the associated journal submission, found two systematic defects in the previous revision and removed 88,440 of its 510,352 rows.

R1 — model spans overlapping a URL_DOMINIO span (88,300 rows). The NER model was reading URL slugs as running text: globo extracted from g1.globo.com/... was recorded a second time as a media outlet, brazil-covid-data as a LOCAL, caixa-encerra-emissao-de-cartoes-black as an ORGANIZACAO. Because URL content is already captured by the deterministic regex layer, these rows duplicated material that was already counted. The duplication was highly concentrated:

Type	Before	Inside URL	After
VEICULO_MIDIA	43,292	32,121 (74.2%)	11,170
PLATAFORMA_DIGITAL	27,889	15,052 (54.0%)	12,835
PESSOA	58,543	13,395 (22.9%)	45,135
ORGANIZACAO	46,835	10,284 (22.0%)	36,542
LOCAL	49,022	9,650 (19.7%)	39,368
ATOR_POLITICO	33,617	1,568 (4.7%)	32,045

R2 — spans whose offsets did not recover their own text (140 rows). Unusable for any span-level work. These are the residual cases previously flagged in the audit tables.

Two consequences worth stating plainly. First, analyses centred on ATOR_POLITICO are barely affected (4.7% contamination), while counts of VEICULO_MIDIA and PLATAFORMA_DIGITAL in the previous revision were substantially inflated. Second, the removal eliminated every overlapping span pair in the corpus, so the entity layer can now be projected onto non-overlapping schemes such as BIO.

Every removed row is published in quality_audit/entities_removed_2026-08.parquet with the rule that removed it, so the previous revision can be reconstructed exactly.

Suggested Use Cases

• Error analysis and qualitative inspection of automatic NER output over politically salient entity categories. Still not a NER benchmark: the released layer is automatic. A 1,016-mention sample was audited by a single annotator (quality_audit/human_audit/), which estimates precision but provides no recall, no inter-annotator reliability and no exhaustive reference annotation, so the layer cannot be used to score NER systems.
• Topic-conditioned studies of source citation, media domains and evidence practices.
• Helpful-note classification using text, rating/status fields and MF scoring features.
• Descriptive studies of the locally recomputed matrix-factorization fields (coreNoteIntercept, coreNoteFactor1) together with topic labels and macrothemes. These fields were re-estimated with the public scoring code restricted to the Portuguese subgraph: they are not the platform's official global scores, and reading coreNoteFactor1 as a political-position or polarization measure requires external validation that this release does not provide.
• Syntax-aware studies combining entities with automatic UD POS/morphology features.
• Procedural-content detection and macrotheme-quality diagnostics.
• Public-interest research on Portuguese-language online discourse across the 2021-2026 Community Notes window. The predominant context is Brazilian, as inferred from topics and entities, but the selection criterion is linguistic and does not establish the nationality of authors or posts.

---

Citation / Como citar

If you use this dataset, please cite the associated preprint:

Community Notes no Brasil: consenso, polarização e aceitabilidade algorítmica. (2026). Em SciELO Preprints. https://doi.org/10.1590/SciELOPreprints.15996

@article{Machado_da_Rocha_2026,
  title     = {Community Notes no Brasil: consenso, polarização e aceitabilidade algorítmica},
  url       = {https://doi.org/10.1590/SciELOPreprints.15996},
  doi       = {10.1590/SciELOPreprints.15996},
  publisher = {FapUNIFESP (SciELO)},
  author    = {Machado da Rocha, Davi},
  year      = {2026},
  month     = aug
}

The dataset itself can be cited as:

@misc{communitynotesbr2026,
  title        = {Community Notes BR: An NLP-Enriched Portuguese Dataset for Collaborative Misinformation Moderation},
  author       = {Rocha, Davi Machado da},
  year         = {2026},
  howpublished = {Hugging Face dataset},
  url          = {https://huggingface.co/datasets/histlearn/notas-comunidade-ptbr},
  note         = {Dataset associated with the preprint Community Notes no Brasil: consenso, polarização e aceitabilidade algorítmica (SciELO Preprints, DOI: 10.1590/SciELOPreprints.15996)}
}

License

The original Community Notes public data is released by X under CC0. Derivative annotations in the core release are distributed under CC0 1.0 Universal to match the source license. Tweet text is not included. The optional UDPipe Bosque syntax layer should be interpreted together with the model/license metadata in syntax_ud_bosque/udpipe_bosque_metadata.json.

AI Usage Disclosure

This dataset includes outputs from machine-learning and generative-AI components as part of its construction:

• intfloat/multilingual-e5-large-instruct for embedding generation.
• Marco-Mini-Instruct for structured topic and macrotheme labels, constrained to parseable JSON.
• urchade/gliner_multi-v2.1 for named-entity extraction.
• UDPipe Bosque for automatic tokenization, POS tagging, morphology, lemmatization and dependency parsing.

Generative AI also assisted with parts of the prose in this datacard and the associated SciELO preprint. The main released annotations are exactly the pipeline outputs and post-processing rules documented above; they should be treated as automatic research annotations, not as human gold labels. The only human-produced labels in this repository are the audit decisions under quality_audit/human_audit/, which cover a sample and were made by a single annotator.

---

Maintainer: Davi Machado da Rocha (USP). Upstream source: Community Notes public data, snapshot 2026-04-07, X Corp under CC0.