Reviews & Marketplaces Moderation

The moderation problem

Reviews are emotional by design. A one-star rant is a legitimate review even when it's angry and a little crude. Over-moderate and you silence honest feedback (and invite accusations of censorship); under-moderate and you let abuse of sellers or staff through.

Why a word list isn't enough

A word list can't tell 'this product is absolute garbage and a waste of money' (a valid review) from 'the owner is a [slur] who should be run out of town' (abuse). It flags both or neither.

How The Profanity API handles it

Use context `review` with `balanced` mode so situational venting about a product or service is allowed through while targeted abuse of an individual is escalated and flagged.

const res = await fetch("https://api.theprofanityapi.com/v1/check", {
  method: "POST",
  headers: {
    "Authorization": "Bearer " + process.env.PROFANITY_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ text, context: "review", mode: "balanced" }),
});

const { flagged, intent, score } = await res.json();