News Dataset

Access real-time financial news articles with AI-powered sentiment analysis for any stock ticker. FinBrain’s news dataset aggregates headlines from major financial publications and scores each article for market sentiment, giving you a comprehensive view of the news landscape around your positions.

What’s Included

The News dataset provides:

News Article Headlines: Full headline text for each article with publication dates
Source Attribution: Publication name for every article (e.g., Reuters, Bloomberg, Yahoo Finance)
Direct Article URLs: Links to the original source article
AI Sentiment Scores: Normalized score from -1 (bearish) to +1 (bullish), nullable when unavailable
Coverage: All FinBrain tracked markets including US equities, crypto, forex, and more

Coverage

Detail	Description
Markets	All FinBrain markets
Update Frequency	Real-time / multiple times daily
History	Rolling recent articles
API Endpoint	`/v2/news/{symbol}`

Understanding News Sentiment

Score Range	Interpretation	Signal
Above 0.5	Strongly positive	Bullish sentiment
0.1 to 0.5	Mildly positive	Slightly bullish
-0.1 to 0.1	Neutral	No strong signal
-0.5 to -0.1	Mildly negative	Slightly bearish
Below -0.5	Strongly negative	Bearish sentiment

Note: The sentiment field can be null when no score is available for an article. Always handle this case in your code.

from finbrain import FinBrainClient

fb = FinBrainClient(api_key="YOUR_API_KEY")

df = fb.news.ticker("AAPL", as_dataframe=True)
print(df)

import requests

API_KEY = "YOUR_API_KEY"
BASE_URL = "https://api.finbrain.tech/v2"
headers = {"Authorization": f"Bearer {API_KEY}"}

# Get news for AAPL
response = requests.get(f"{BASE_URL}/news/AAPL", headers=headers)
result = response.json()

for article in result["data"]["articles"]:
    sentiment = article["sentiment"] or "N/A"
    print(f"{article['date']} | {article['source']} | {sentiment} | {article['headline']}")

For complete endpoint details, parameters, and response schema, see the News API Reference.

Use Cases

Filter High-Sentiment News Articles

Surface only the most positive or negative news articles for a ticker:

from finbrain import FinBrainClient

fb = FinBrainClient(api_key="YOUR_API_KEY")

df = fb.news.ticker("TSLA", as_dataframe=True)

# Filter articles with strong sentiment
bullish_news = df[df["sentiment"].notna() & (df["sentiment"] > 0.5)]
bearish_news = df[df["sentiment"].notna() & (df["sentiment"] < -0.5)]

print(f"Strongly bullish articles: {len(bullish_news)}")
for _, a in bullish_news.iterrows():
    print(f"  [{a['sentiment']:.2f}] {a['headline']}")

print(f"\nStrongly bearish articles: {len(bearish_news)}")
for _, a in bearish_news.iterrows():
    print(f"  [{a['sentiment']:.2f}] {a['headline']}")

Aggregate Daily Sentiment from News

Compute an average sentiment score per day from individual article scores:

from finbrain import FinBrainClient

fb = FinBrainClient(api_key="YOUR_API_KEY")

df = fb.news.ticker("NVDA", as_dataframe=True)

# Filter out articles without sentiment scores
scored = df[df["sentiment"].notna()].copy()

# Group by date (index) and compute daily average
daily_avg = scored.groupby(scored.index).agg(
    avg_sentiment=("sentiment", "mean"),
    article_count=("sentiment", "count")
)

for date, row in daily_avg.sort_index().iterrows():
    print(f"{date}: avg sentiment = {row['avg_sentiment']:.3f} ({int(row['article_count'])} articles)")

Track News Volume and Sentiment Trend

Monitor how news volume and sentiment shift over time for early signals:

from finbrain import FinBrainClient

fb = FinBrainClient(api_key="YOUR_API_KEY")

df = fb.news.ticker("AMZN", as_dataframe=True)

# Compute daily volume and average sentiment
daily = df.groupby(df.index).agg(
    count=("headline", "count"),
    avg_sentiment=("sentiment", "mean")
)

for date, row in daily.sort_index().iterrows():
    sent_str = f"{row['avg_sentiment']:.3f}" if row["avg_sentiment"] == row["avg_sentiment"] else "N/A"
    print(f"{date}: {int(row['count'])} articles, avg sentiment: {sent_str}")

News API Reference - Endpoint details, parameters, and response schema
News Sentiment Dataset - Aggregated daily sentiment scores
Screener API Reference - Screen tickers by sentiment and other criteria