How to identify fake news and misinformation#

The APITube News API has no single “is_fake” flag — misinformation is not a field, it is a judgement you build from signals. The reliable approach is to combine several verifiable credibility filters in one request: weight toward high-rank and verified sources, drop clickbait and manipulative framing, remove duplicate re-posts, and check a source’s political lean. This query keeps coverage from strong sources, excludes clickbait, and drops duplicate re-posts:

curl "https://api.apitube.io/v1/news/everything?title=election%20fraud&is_verified_source=1&is_clickbait=0&is_duplicate=0&api_key=YOUR_API_KEY"

You can pass the key as the api_key query parameter or as the X-API-Key header. None of these filters proves a story true or false; together they raise or lower the odds that what you are reading is credible, which is exactly how a human vets a source.

Is there a single “fake news” flag in the API?#

No. There is no boolean that labels an article as fake or true, because credibility is not something the crawler can measure directly. What the API does expose is a set of measurable proxies for trustworthiness — the source’s authority rank, whether it is a verified source, its political bias, whether the headline is clickbait, and whether the piece is a duplicate. You assemble these into a credibility check. Start strict and loosen if you get too few results:

curl "https://api.apitube.io/v1/news/everything?title=miracle%20cure&is_verified_source=1&is_clickbait=0&is_duplicate=0&source.bias=center&api_key=YOUR_API_KEY"

Each parameter below is one signal in that check, and each has a dedicated guide if you want the full behaviour.

How to weight results toward trustworthy sources#

The strongest single signal is source authority, measured by Open Page Rank (OPR). Two shortcut filters bundle this for you:

• is_verified_source=1 keeps only sources with an OPR of 5 or higher and automatically drops duplicates — the tightest built-in quality gate.
• is_premium_source=1 is stricter still, keeping only sources with an OPR of 6 or higher.

For manual control, source.rank.opr.min sets your own floor — for example source.rank.opr.min=7 for only the most authoritative outlets. Higher OPR means a more widely-linked, more established site, which correlates with editorial accountability:

curl "https://api.apitube.io/v1/news/everything?title=vaccine&source.rank.opr.min=6&is_duplicate=0&api_key=YOUR_API_KEY"

See how to filter for high-quality, verified sources for the difference between the three, and what is the OPR score for how the rank itself is built.

How to spot clickbait and manipulative framing#

Misinformation often rides on an emotionally loaded headline that the body does not back up. Two sentiment signals catch that pattern. is_clickbait=1 returns articles whose headline tone sharply disagrees with the body tone and whose headline sentiment is strong — the classic bait-and-switch — while is_clickbait=0 filters those out of a clean feed. For a finer knob, sentiment_gap.min measures the raw distance between headline sentiment and body sentiment on a 0-to-2 scale; a large gap flags a headline written to provoke rather than inform:

curl "https://api.apitube.io/v1/news/everything?title=climate&sentiment_gap.min=0.8&api_key=YOUR_API_KEY"

Use is_clickbait=1 or a high sentiment_gap.min when you are hunting for suspect stories to review, and is_clickbait=0 when you want them kept out of a trustworthy feed. The detect and filter clickbait guide covers both directions.

How to check a source’s political lean#

A source’s slant is not misinformation on its own, but knowing it helps you read a story in context and balance a set of results. source.bias filters by the outlet’s political lean and accepts left, center and right (up to three, comma-separated). To read across the spectrum on one topic, pull each lean and compare how the same event is framed:

curl "https://api.apitube.io/v1/news/everything?title=immigration&source.bias=left,center,right&api_key=YOUR_API_KEY"

Balancing bias is a defence against a one-sided narrative: if only one side is reporting a claim, that itself is a signal to dig deeper. There is a matching ignore.source.bias to exclude a lean you do not want.

How to verify a specific claim and trace its spread#

Filters narrow the field; verification confirms a single claim. Two tools finish the job. To trace how far a story spread and reach its origin, follow the duplicates back to the first publisher — how to find the original source of a story walks through using the story cluster to do this. To check a concrete statement against the underlying reporting, the /v1/fact-check endpoint extracts the claims in your text and returns a verdict for each one, plus a summary. Fact-check is a paid feature: on the free plan it returns 403 with error code ER0706, so it needs a paid subscription. The fact-check a claim or article guide shows the request and response in full.

For the complete list of every parameter used here, the News API parameters reference is the canonical source.

Common Questions#

• Can the API tell me an article is definitely fake?
• What is the difference between is_verified_source and is_premium_source?
• Does source.bias support an “unbiased” value?

Can the API tell me an article is definitely fake?#

No. There is no field that classifies an article as fake or true. The API gives you measurable credibility signals — source rank (OPR), verified-source status, political bias, clickbait detection and duplicate detection — plus a /v1/fact-check endpoint for verifying a specific claim. You combine those into a judgement; the API supplies the evidence, not the verdict.

What is the difference between is_verified_source and is_premium_source?#

Both are OPR shortcuts. is_verified_source=1 keeps sources with an OPR of 5 or higher and also removes duplicate articles in the same pass. is_premium_source=1 is stricter, requiring an OPR of 6 or higher. Use is_verified_source for a broad trustworthy feed and is_premium_source when you want only top-tier outlets; for a custom threshold, set source.rank.opr.min yourself.

Does source.bias support an “unbiased” value?#

No. source.bias accepts left, center and right only. If you want centrist coverage, use source.bias=center; to exclude a particular lean, use ignore.source.bias. Any value outside left, center and right is ignored rather than applied, so check the spelling if a bias filter seems to have no effect.