India Multilingual News & Sentiment API

Collect recent Indian news from named publications in English, Hindi, Kannada, and Tamil. Filter by source, keyword, and date; remove duplicate stories; optionally extract article text; and add a multilingual sentiment label to every result.

Use it for brand monitoring, regional media research, sentiment dashboards, competitive intelligence, and AI or data pipelines.

Why use this Actor?

• Regional-language coverage — collect news beyond English-only aggregators.
• Named publishers — results include the original publication and article URL.
• Multilingual sentiment — classify each story as positive, negative, or neutral in its own language.
• Useful filters — select languages, publishers, keywords, and a date range.
• Clean output — one structured dataset row per article, with duplicates removed.
• Optional full text — request article body extraction when your workflow needs it.
• Fail-soft processing — one unavailable feed or article does not stop the whole run.

Sources covered

Language	Publications
English	Times of India, The Hindu, NDTV, Hindustan Times, The Indian Express, Livemint
Hindi	NDTV India, Amar Ujala, Dainik Bhaskar
Kannada	Vijaya Karnataka, Prajavani, OneIndia Kannada
Tamil	OneIndia Tamil

Need another Indian language or publication? Contact the developer with the source you need.

Quick start

1. Choose one or more languages.
2. Optionally select publishers, keywords, or dates.
3. Set a result limit with maxItems.
4. Keep sentiment enabled, or turn it off for article-only results.
5. Click Start and open the Dataset when the run finishes.

Example: regional brand monitoring

{
  "languages": ["Kannada", "Hindi"],
  "keywords": ["ಬೆಂಗಳೂರು", "मुंबई"],
  "maxItems": 50,
  "moodScore": true
}

Save a tested configuration as an Apify Task and attach it to an Apify Schedule for recurring monitoring. Send the resulting dataset to your database, spreadsheet, webhook, or automation platform.

Input

Field	Description
languages	Languages to include. Default: English, Hindi, and Kannada.
sources	Optional list of supported publications.
keywords	Keep articles matching any of these terms.
dateFrom / dateTo	Optional publication date range.
maxItems	Maximum articles returned; also caps the run cost. Default: 200.
includeBody	Attempt to extract the full article body. Default: off.
moodScore	Add multilingual sentiment and a short reason. Default: on.
aiApiKey	Optional advanced setting for using your own Gemini API key.

Output

Each dataset item contains structured article data similar to:

{
  "title": "ಬೆಂಗಳೂರಿನಲ್ಲಿ ಹೊಸ ಮೆಟ್ರೋ ಮಾರ್ಗ ಉದ್ಘಾಟನೆ",
  "link": "https://example.com/article",
  "source": "OneIndia Kannada",
  "language": "Kannada",
  "category": "top",
  "publishedAt": "2026-07-01T09:15:00.000Z",
  "summary": "...",
  "mood": "positive",
  "moodReason": "The story describes improved transport access."
}

Available fields can also include the extracted article body when includeBody is enabled.

Common workflows

Brand and reputation tracking

Run keyword searches for a company, product, executive, or campaign across multiple Indian languages. Use the mood field to prioritize stories for review.

Regional media research

Compare coverage by language or publisher and export the dataset as JSON, CSV, or Excel.

AI and analytics pipelines

Feed normalized article records into an LLM, search index, RAG system, BI dashboard, or downstream classifier.

Pricing

You pay per article returned. Article-only results cost less; results with sentiment are priced slightly higher to cover AI processing. Set maxItems to control the maximum result count and cost.

Notes

• Source feeds can occasionally be unavailable or change without notice. The Actor continues processing other sources when this happens.
• Full-text extraction depends on each publisher's page structure and may not succeed for every article.
• Headlines, summaries, dates, and links come from public feeds. Respect publisher terms and copyright when using or storing full article text.

Support

Questions, missing sources, or feature requests are welcome through the Actor's Issues tab or the developer profile.