Alternative data datasets
Browse by provider, category, marketplace, access, delivery and license
10-K filings for S&P 500 companies retrieved from EDGAR, stored as Parquet with the filing text attached (MIT). A focused corpus for the large-cap annual disclosure cycle: risk factors, MD&A and segment detail for every S&P 500 name.
View source information and access options.
View source information and access options.
View source information and access options.
Layout metadata produced by PatentHubLayoutV2 for a JusticeDAO repository holding patent and legal material. It references the CFR Title 37 annual edition for 2024 and carries the tag patent-legal-v2.0.0, with publication handled separately as an operator decision.
The Patent Legal IR GraphRAG release follows the Publicus retrieval format, providing dense corpus shards, BM25 documents with CID-keyed length and entry metadata, and BM25 postings arranged for inverted-index use.
Repository layout details generated by PatentHubLayoutV2 for JusticeDAO's patent and legal knowledge graph. The metadata references the CFR Title 37 2024 annual edition under a U.S. public-domain license and is tagged patent-legal-v2.0.0.
A vector and layout configuration for patent and legal texts produced via PatentHubLayoutV2, derived from Title 37 of the Code of Federal Regulations annual edition 2024 and marked as U.S. government public domain.
Datamule partnered with Teraflop AI and Eventual to publish the SEC-EDGAR dataset, which covers all major filings from the SEC EDGAR database and amounts to 590 GB across about 8 million samples and 43 billion tokens. Collection relied on the datamule-python library and John Friedman's official datamule API, tools designed in Python for gathering and handling SEC filing data.
View source information and access options.
The Future of Patent Corpus Management One Document, Fully Decoded - Without Anyone Reading It The signal you cannot get from a search interface The highest-value competitor signal in a patent portfolio is not the invention. It is the amount of money and urgency a company committed to it. That signal is never in an abstract, and no search interface surfaces it. It comes out of this corpus for ever
Machine-readable extraction of financial and operating figures drawn from publicly filed corporate documents.
The patents-publications-dataset is a large-scale tabular collection of patent publication records distributed in Parquet format. Hosted by publisher labofsahil on the Hugging Face Hub under an unstated license, it is sized in the 100M–1B row range.
Privacy-protected statistics on Florida holders of dormant property interests, broken out by holding size, loan-to-value ratio, portfolio scale, risk profile, and tier.
Information on roughly 95,000 Florida property investors, showing a median equity of $1.6 million, an 11% LTV, and no transactions in the past year.
Segmentation of Florida's 1.96 million real estate investors according to portfolio size, trading activity, equity position, investment approach, and risk exposure.
Approximately 20.8 million US real estate investors can be profiled by state, portfolio size, equity held, and recent transaction activity.
View source information and access options.
View source information and access options.
View source information and access options.
View source information and access options.
View source information and access options.
Metadata for hundreds of thousands of clinical trials in English, French and Spanish, structured in Parquet with study-level fields such as phase, condition, sponsor and status (Apache 2.0). Trial starts and phase transitions lead clinical development, so this works as an early indicator of biotech pipeline activity.
Minimal information is provided on this clinical-trials-embedded dataset, with additional details needed.
A Comprehensive Dataset on Real Estate Asking Prices and Property Features
Unlocking Real Estate Insights: Analyzing, Visualizing, and Predict with ML
A corpus of around 1.3 million U.S. patent filings paired with human-authored abstractive summaries, organized into nine Cooperative Patent Classification categories ranging from human necessities to textiles and paper.
The clinical_trials dataset distributes ClinicalTrials.gov registry records in Parquet format for offline analysis. It contains a single training split of roughly 150,000 rows sourced from the public clinical-trial registry.
patent-desc-test is a small text dataset on Hugging Face Datasets that contains short identifier-style strings derived from patent document numbers. The dataset is published by user matthias-ehrlich and is broadly tagged as public records–adjacent text data.
A unified compilation of worldwide crude oil prices along with import, export, and net import series, blending daily and monthly observations from multiple public sources into one analytical file.
An analytical project examining the Paris Housing Dataset to uncover which property attributes most strongly affect asking prices. It involves cleaning, visualizing, and statistically modeling the records to answer targeted research questions.
2.8K+ venture funding rounds across 35 countries & 20 sectors with investor data
Standardized financial datasets that clean and structure company financial statements from filings for downstream analytics.
Number and valuation of new housing units authorized by building permits
AlphaAI's near-real-time parse of SEC EDGAR Form 4 filings, providing 20,832 individual insider transaction tranches and 8,538 grouped economic events for US public-company officers, directors, and 10% owners.
Parsed SEC EDGAR Form 4 insider trades, as tranches and scored events
View source information and access options.
Real estate listings in Madrid crawled from popular internet portals
View source information and access options.
Clean and structured apartment prices data from Greater Cairo real estate market
EDGAR filings of companies
An analysis-ready compilation of roughly 3,000 ClinicalTrials.gov studies registered between 2000 and 2025 that involve AI, machine learning, or digital-health tools, enriched with 30 LLM-derived variables covering use case, therapeutic area, sponsor composition, trial phase, deployment score, evidence strength, and responsible-AI keyword indicators.
A queryable DuckDB mirror of the AACT flat-file export of ClinicalTrials.gov, packaged via the clinicaltrials-database project and covering every registered trial across 48 tables totaling roughly 58 million rows.
google-patents-data-preview is a community-uploaded preview of Google Patents bibliographic records, distributed in Parquet format through the Hugging Face Hub. The single training split contains roughly 340,000 rows covering patent identifiers, classifications, and localized text fields.
Complete filing metadata and URLs for 8 core SEC forms across Q1-Q4 2024
Complete filing metadata and URLs for all 8-K material event disclosures in Q4
Complete filing metadata and URLs for 82,000+ annual report filings (July-Dec)
Complete filing metadata and URLs for all insider transaction filings
High-Conviction Insider Trading & Institutional Holding Signals
A point-in-time snapshot of single-family-home price records for November 2024 in CSV, tens of thousands of entries. Useful for housing-market analysis and cross-checks against other property-price series.
Targeted research and datasets focused on niche sectors, private firms and specialised investment topics.
Two Parquet files hold global patent publication records paired with quality indicators for Chinese invention patent families from 2003 to 2019, with each row representing a focal family and carrying cumulative measures such as semantic knowledge recombination, strict historical semantic novelty, and IPC-based robustness.
5,000-row sample: Canadian building permits, licences, planning & inspections
View source information and access options.
Over 1.3 million granted US patents from the BIGPATENT benchmark, each pairing its claims with an abstractive summary and full description text in English (CC BY 4.0, Parquet). Use it to track the technology areas where applicants are filing and to study how claim language has evolved over time.
View source information and access options.
A broad registry of business identifiers and name variants maintained for organizations operating in over 140 countries.
Comprehensive officer records spanning more than 140 jurisdictions to support corporate governance research and due diligence checks.
Corporate linkage and beneficial ownership intelligence spanning more than 140 regulatory jurisdictions worldwide.
Roster of present-day and past candidates running for U.S. federal and state-level political offices.