Find the Tone of News Articles with News API

Text sentiment analysis tells you whether written content is positive, negative, or neutral. The API returns a classification and a confidence score.

Polarity tells you the direction (positive, negative, neutral). The sentiment score (0-1) tells you the strength. A score of 0.9 with positive polarity means strongly positive; 0.5 means mildly positive or mixed.

85%+ accuracy across general news categories, with training for business, politics, technology, and entertainment. Accuracy varies by domain—factual business news scores higher than opinion pieces with irony or sarcasm.

Yes, entity-level sentiment analysis shows how specific brands, people, or organizations are portrayed. An article might be neutral overall but positive about one company and negative about another.

The API recognizes common sarcasm patterns and contextual irony by analyzing surrounding context, source characteristics, and topic-specific language patterns. No system is perfect with nuanced language.

Many financial firms use news sentiment as one input for trading models. The API provides real-time sentiment data that can be correlated with market movements. Combine sentiment signals with other fundamental and technical analysis.

Yes, sentiment works in 60+ languages with language-specific training. Each understands cultural nuances and language-specific expressions that affect sentiment interpretation.

Use the time-series endpoints to aggregate sentiment data by hour, day, week, or month. Track how sentiment around topics, brands, or events changes, and correlate changes with specific news events or campaigns.