The Dataset Viewer has been disabled on this dataset.

FENRIX PM Decision News

Professor-facing financial news product organized by article meaning for portfolio-manager decision research.

Current release scope (teammate slice only)

This repository currently contains the verified teammate product slice ingested from the July 2026 teammate event package. It does not yet include the full 37.57M-row warehouse silver lake routed into the product tree.

Metric Value
Accepted rows 44,904
Date coverage 2019-07-01 to 2026-06-30
Parquet files 656
Data/News Parquet files 572
Data/Structured Parquet files 84

Sources in this release

Guardian, IMF, FRED, Federal Register, GDELT, Census, EIA, BEA, FDA, CFPB, FDIC, and BLS news rows from the teammate package.

What is not included

  • Raw zip, jsonl, or txt intake files
  • The full warehouse-backed product build (37,571,131 silver rows) — planned as a separate staging build
  • API keys or local filesystem paths in exported Parquet

Folder layout

Folder Content
Data/News/Economic_Macro/ Macro releases, rates, GDP, global macro
Data/News/Sector_Industry/ Energy, sector themes
Data/News/Single_Stock/ Company-specific Guardian and tagged rows
Data/News/Multi_Company/ Multi-company context
Data/News/Market_Context/ Regulatory, banking, geopolitical, unclassified context
Data/Structured/ Official structured economic/regulatory series
Reports/ Teammate ingest reconciliation and QA reports
Docs/ Product notes and professor guidance

Verification notes

Use the Hugging Face API file tree and Reports/Teammate_* files as the source of truth for this release. Do not rely on the dataset viewer preview if it shows stale smoke-test samples from an earlier staging state.

Required QA reports:

  • Reports/Teammate_Ingest_Summary.csv
  • Reports/Teammate_Source_Row_Counts.csv
  • Reports/Teammate_Date_Coverage.csv
  • Reports/Teammate_Product_Path_Counts.csv
  • Reports/Teammate_Article_Group_Counts.csv
  • Reports/Teammate_Full_Text_Join_Report.csv
  • Reports/Teammate_Duplicate_Report.csv
  • Reports/Teammate_Secret_Scan_Report.csv
  • Reports/Teammate_Guardian_Ingest_Summary.csv
  • Reports/Teammate_Staging_Upload_Manifest.csv
  • Reports/Professor_Summary_Teammate_Update.md
  • Docs/Teammate_Data_Notes.md

Quick load

import pandas as pd
from huggingface_hub import hf_hub_download

path = hf_hub_download(
    repo_id="Scottswi/fenrix-pm-decision-news",
    filename="Data/News/Economic_Macro/Global_Macro/Imf/year=2020/month=01/part-imf-2020-01-00000.parquet",
    repo_type="dataset",
)
df = pd.read_parquet(path)
print(df.columns.tolist())
print(len(df))

Schema (41 columns)

Key fields: article_id, article_group, article_type, topic, market_scope, published_at, event_date, title, description, body, body_available, tickers, companies, sectors, macro_tags, canonical_url, publisher, source_id, license_class, redistribution_status, provenance_uri, source_record_id, source_package, ingested_at_utc.

License and redistribution

Sources have mixed terms. Use row-level license_class and redistribution_status before external sharing. Guardian and some news rows may be restricted_or_unknown or require non-commercial review. Official macro/regulatory structured rows are generally public-domain with citation.

Technical warehouse

The full technical news lake (37.57M sanitized silver rows) remains at:

https://huggingface.co/datasets/Scottswi/fenrix-financial-news-lake

Product staging builds are validated at:

https://huggingface.co/datasets/Scottswi/fenrix-pm-decision-news-staging

Citation

@dataset{fenrix_pm_decision_news_teammate_2026,
  author = {FENRIX},
  title = {FENRIX PM Decision News — Teammate Product Slice},
  year = {2026},
  publisher = {Hugging Face},
  howpublished = {\\url{https://huggingface.co/datasets/Scottswi/fenrix-pm-decision-news}},
  note = {44,904 rows; 2019-07 to 2026-06; teammate structured + Guardian package}
}
Downloads last month
1,372