Aller au contenu

Données ouvertes

Bibliothèque

Jeux de données ouverts, entièrement documentés — interrogeables ici, et lisibles par n’importe quel LLM.

Les titres et les descriptions proviennent des sources de données, en anglais.

8 jeux de données

sentiment

  • Business & Consumer Confidence Signals (derived)

    Global business & consumer confidence signals (sentiment momentum, pessimism streaks, business-consumer gap)

    Monthly economic-sentiment signals derived from the OECD's standardised business and consumer confidence indicators (amplitude adjusted, long-term average = 100; 56 countries and aggregates): 30-month annualized change volatility, 3-month momentum, 3-sigma anomaly flags vs a trailing 12-month baseline, naive-drift 1-month forecasts, a per-month cross-series volatility rank, the business-minus-consumer confidence gap (the demand-mismatch gauge), consecutive-months-below-100 pessimism streaks and 24-month sentiment z-scores. All rows carry canonical country_code so they join cleanly with other macro datasets. Derived keyless from the local raw oecd_cli catalog snapshots; underlying data: OECD (commercial re-use requires prior permission).

    • sentiment
    • consumer-confidence
    • business-confidence
    • surveys
    lignes
    52 267
    Qualité
    96
    Mis à jour
    22 sept. 2026
    À jour
    Licence
    Usage commercial interdit

  • Crypto Fear & Greed Signals (derived)

    Crypto Fear & Greed signals (crypto sentiment regimes)

    Daily crypto-sentiment signals from the alternative.me Crypto Fear and Greed Index (2018-02 ->, 0–100): 7-day and 30-day momentum, 30-day change volatility, 3-sigma anomaly flags, naive-drift forecasts, extreme-fear (<=25) and extreme-greed (>=75) regime flags, a 90-day sentiment z-score, and the provider's own classification. The crypto-sentiment companion to fng-signals (CNN stock-market Fear & Greed) — crypto runs its own sentiment cycle, often leading equities. Keyless JSON API; commercial redistribution not explicitly granted (marked unclear). Global market, rows labeled WLD. Raw series: alternative.me (tobieskrambs GmbH).

    • crypto
    • bitcoin
    • sentiment
    • fear-and-greed
    lignes
    3 156
    Qualité
    92
    Mis à jour
    26 sept. 2026
    Vieillissant
    Licence
    Licence incertaine

  • Economic Policy Uncertainty Signals (derived)

    Economic policy uncertainty signals (US daily + global monthly, z-score regime gauges, US-vs-world divergence)

    Value-added policy-uncertainty signals derived from FRED's mirror of the Baker/Bloom/Davis Economic Policy Uncertainty indices (policyuncertainty.com): daily US EPU (1985 -> present) and monthly global EPU (1997 -> present). Each row carries change volatility, ~3-month momentum, year-over-year change, a trailing-1y uncertainty z-score with an elevated-uncertainty flag, 3-sigma anomaly flags, naive-drift 1-month forecasts, a per-date cross-series volatility rank, and the US-vs-world uncertainty divergence gauge. All computation is local pandas/numpy; no paid models or APIs.

    • uncertainty
    • policy
    • sentiment
    • risk
    lignes
    15 597
    Qualité
    92
    Mis à jour
    26 sept. 2026
    Vieillissant
    Licence
    Licence incertaine

  • Crypto Fear & Greed Signals (derived)

    Crypto fear & greed index signals (sentiment momentum, extreme-regime flags, contrarian z-scores)

    Daily market-sentiment signals derived from the alternative.me Crypto Fear & Greed Index (0-100, keyless free API, history from 2018-02-01): 7-day and 30-day sentiment momentum, 30-day change volatility, 3-sigma anomaly flags, extreme fear/greed regime flags (<=15 / >=85), consecutive-days regime streaks, a 90-day contrarian z-score and naive-drift 1-month forecasts. The sentiment companion to the price-based crypto market signals — no price data here. All rows are normalized to country_code WLD (World). Raw index: alternative.me (free keyless API).

    • sentiment
    • fear-and-greed
    • crypto
    • contrarian
    lignes
    3 156
    Qualité
    100
    Mis à jour
    26 sept. 2026
    Vieillissant
    Licence
    Usage commercial OK

  • GDELT global-news company tone (agent-curated)

    GDELT global-news company tone (daily)

    Daily enrichment signals scoring how the world's news talks about ~70 US and Canadian companies. Each day the 96 keyless GDELT 2.1 GKG fifteen-minute bulk files for the last complete UTC day are downloaded; articles are deduplicated on URL, their V2ENHANCEDORGANIZATIONS are resolved against a curated company dictionary (name, primary-listing ticker, HQ country, aliases — shared with the hn_company_sentiment series so curation cannot drift), and GDELT's own per-article V1.5TONE scores (computed from full article text) are aggregated per company and day. Per company: n_articles, n_sources, mean/median/std/min/max tone, pos_share / neg_share (fraction of articles with tone >= 1.0 / <= -1.0), mean_pos_words_pct / mean_neg_words_pct, dominant_theme, tone_signal (positive / negative / polarized / none), and news_tone_score = 100 * (55% min-max-normalized mean_tone + 25% min-max-normalized pos_share + 20% min-max-normalized (1 - neg_share)), ranked as tone_rank (1 = most positive; ties: n_articles desc, company name asc). Who joins this: a quant desk joins per-ticker daily global-news tone to their positions/returns table on ticker + as_of; an IR/comms team joins on company_name to their media-monitoring calendar. Primary key: (as_of, company_name); join keys: ticker, company_name, as_of, country_code (ISO alpha-3). Nullability: ticker is null for private companies (OpenAI, Anthropic, Stripe, ...); std_tone is null for single-article companies; dominant_theme is null when no themes were recorded. Caveats: news_tone_score is a within-snapshot relative measure, not comparable across days; GDELT tone is computed on machine-translated text for non-English articles; organization NER misses and corporate-suffix collisions are possible despite suffix stripping. No article text and no per-row URLs are stored. Sample use: filter ticker = 'TSLA' for Tesla's daily global-news tone, or order by tone_rank for today's most positively covered companies.

    • gdelt
    • news
    • sentiment
    • signals
    lignes
    26
    Qualité
    100
    Mis à jour
    27 sept. 2026
    À jour
    Licence
    Usage commercial OK

  • HN company-mention sentiment (agent-curated)

    HN company-mention sentiment (daily)

    Daily enrichment signals scoring how Hacker News talks about ~70 US and Canadian companies. Each day the official keyless HN search API (Algolia) supplies the current front page plus every story submitted in the trailing 24 hours; a curated company dictionary (name, primary-listing ticker, HQ country, aliases, link domains) resolves mentions on title + selftext + URL host, and a documented deterministic valence lexicon (negation and intensifier rules, no model, no network) scores each mentioning story. Per company and day: mention_count, story_ids, total_points (HN points collected — an attention proxy), mean_sentiment (unweighted mean story sentiment), weighted_sentiment (points-weighted), pos_share / neg_share (fraction of mentioning stories with sentiment > 0.12 / < -0.12), the top story's title, and buzz_score = 100 * (50% min-max-normalized mention count + 50% min-max-normalized log10(total points + 1)), ranked as attention_rank (1 = hottest; ties: mention count desc, company name asc). Who joins this: a quant desk or investor-relations team joins per-ticker daily tech sentiment to their positions/returns table on ticker + as_of; a comms team joins on company_name to their media-monitoring calendar. Primary key: (as_of, company_name); join keys: ticker, company_name, as_of, country_code (ISO alpha-3). Nullability: ticker is null for private companies (OpenAI, Anthropic, Stripe, ...); top_story_title is never null. Caveats: sentiment is computed from story titles + selftext only (comments are not fetched — that would multiply requests ~100x); story-level sentiment is a lexicon mean, not a model judgment; buzz_score is a within-snapshot relative measure, not comparable across days; HN skews toward software/AI topics, so hardware/retail names are sparser. Sample use: filter ticker = 'NVDA' for Nvidia's daily HN sentiment, or order by attention_rank for today's most-buzzed companies.

    • hacker-news
    • technology
    • signals
    • sentiment
    lignes
    23
    Qualité
    92
    Mis à jour
    27 sept. 2026
    À jour
    Licence
    Licence incertaine

  • Wikipedia AI Attention Signals (derived)

    Wikipedia AI-attention & hype signals

    Monthly revealed-attention signals from English-Wikipedia pageviews (2015 ->): attention paid to AI-revealing articles (Artificial intelligence, ChatGPT, Large language model, Generative artificial intelligence, Machine learning, Deep learning), with 3-month momentum, 12-month change volatility, 3-sigma anomaly flags, drift forecasts, 5-year attention z-scores, attention-spike flags, and a composite AI-hype index (cross-article mean z-score) with high/extreme hype flags. The technology-hype companion to wiki-economic-attention-signals (economic anxiety topics). All rows normalized to country_code WLD (global internet attention). Raw data: Wikimedia Foundation pageview statistics API.

    • wikipedia
    • pageviews
    • ai
    • attention-economics
    lignes
    607
    Qualité
    91
    Mis à jour
    22 sept. 2026
    À jour
    Licence
    Licence incertaine

  • Wikipedia Attention Signals (derived)

    Wikipedia economic-attention & fear signals

    Monthly revealed-attention signals from English-Wikipedia pageviews (2015 ->): attention paid to economically revealing articles (Recession, Inflation, Unemployment, Great Recession, Stock market crash, Layoff), with 3-month momentum, 12-month change volatility, 3-sigma anomaly flags, drift forecasts, 5-year attention z-scores, attention-spike flags, and a composite economic-fear index (cross-article mean z-score) with high/extreme fear flags. The revealed-attention companion to us-consumer-sentiment-signals (stated sentiment) and crypto-fear-greed-signals (market-implied fear). All rows normalized to country_code WLD (global internet attention). Raw data: Wikimedia Foundation pageview statistics API.

    • wikipedia
    • pageviews
    • attention-economics
    • fear-index
    lignes
    879
    Qualité
    91
    Mis à jour
    22 sept. 2026
    À jour
    Licence
    Licence incertaine

Utiliser votre propre clé d’IA

Une fois l’allocation gratuite du jour épuisée, les fonctions d’IA peuvent passer par votre propre compte fournisseur.

Conservée uniquement dans cet onglet (effacée à sa fermeture) et envoyée avec chaque requête d’IA. Nos serveurs l’utilisent pour cette requête et ne la stockent ni ne la journalisent jamais.