Text Analysis

Analyze text content for spam, toxicity, profanity, threats, insults, and language detection. Our AI models provide accurate classification with confidence scores.

Sentiment Analysis

Detect toxic content, profanity, threats, and insults with confidence scores.

Spam Detection

Identify spam, phishing attempts, and promotional content.

Language Detection

Automatically detect the language of input text with high accuracy.

Content Flags

Get clear true/false flags for quick decision making.

Endpoint

POST https://api.discuse.com/api/v2/check

Request Format

Send your text content in the request body. The text field is required.

{
  "content": {
    "text": "Your text content to analyze goes here"
  }
}

Request Parameters

Parameter	Type	Required	Description
`content.text`	string	Yes	The text content to analyze (max 10,000 characters)

Response Format

The API returns detailed analysis results including sentiment scores, spam detection, and language.

{
  "has_violations": false,
  "cached": false,
  "message": "Content appears safe",
  "results": {
    "hits": false,
    "sentiment": {
      "hit": false,
      "toxic": 0.02,
      "profanity": 0.01,
      "threat": 0.00,
      "insult": 0.03
    },
    "spamfinder": {
      "is_spam": false,
      "confidence": 0.05
    },
    "language": {
      "detected": "en",
      "confidence": 0.98
    }
  },
  "processing_time_ms": 45,
  "usage": {
    "api_requests_used": 150,
    "api_requests_limit": 10000,
    "api_requests_remaining": 9850
  }
}

Response Fields

Top-Level Fields

Field	Type	Description
`has_violations`	boolean	True if any violations were detected
`cached`	boolean	True if result was served from cache
`message`	string	Human-readable summary of the analysis
`processing_time_ms`	number	Processing time in milliseconds

Sentiment Analysis Fields

All sentiment scores are values between 0.0 and 1.0, where higher values indicate stronger presence of that sentiment.

Field	Type	Description
`hit`	boolean	True if any sentiment threshold was exceeded
`toxic`	number	Toxicity score (0.0 - 1.0)
`profanity`	number	Profanity score (0.0 - 1.0)
`threat`	number	Threat score (0.0 - 1.0)
`insult`	number	Insult score (0.0 - 1.0)

Spam Detection Fields

Field	Type	Description
`is_spam`	boolean	True if content is classified as spam
`confidence`	number	Classification confidence (0.0 - 1.0)

Language Detection Fields

Field	Type	Description
`detected`	string	ISO 639-1 language code (e.g., "en", "es", "fr")
`confidence`	number	Detection confidence (0.0 - 1.0)

Example Responses

Toxic Content Detected

{
  "has_violations": true,
  "cached": false,
  "message": "Potential violations detected",
  "results": {
    "hits": true,
    "sentiment": {
      "hit": true,
      "toxic": 0.92,
      "profanity": 0.15,
      "threat": 0.78,
      "insult": 0.85
    },
    "spamfinder": {
      "is_spam": false,
      "confidence": 0.12
    },
    "language": {
      "detected": "en",
      "confidence": 0.99
    }
  },
  "processing_time_ms": 52,
  "usage": {
    "api_requests_used": 151,
    "api_requests_limit": 10000,
    "api_requests_remaining": 9849
  }
}

Spam Content Detected

{
  "has_violations": true,
  "cached": false,
  "message": "Spam detected",
  "results": {
    "hits": true,
    "sentiment": {
      "hit": false,
      "toxic": 0.05,
      "profanity": 0.02,
      "threat": 0.01,
      "insult": 0.03
    },
    "spamfinder": {
      "is_spam": true,
      "confidence": 0.94
    },
    "language": {
      "detected": "en",
      "confidence": 0.97
    }
  },
  "processing_time_ms": 38,
  "usage": {
    "api_requests_used": 152,
    "api_requests_limit": 10000,
    "api_requests_remaining": 9848
  }
}

Best Practices

1
Use appropriate thresholds
Sentiment scores are between 0.0 and 1.0. A threshold of 0.7 is typically recommended for high confidence detection.
2
Handle edge cases
Very short text may have lower confidence scores. Consider requiring minimum text length for reliable analysis.
3
Leverage caching
The API caches results for identical content. Cached responses do not count against your usage quota - you only pay for actual checks. The cached field indicates if a cached result was returned.
4
Monitor usage
Check the usage field to monitor your API quota and avoid unexpected rate limits.

Try Text Analysis

Test text analysis with our interactive demo. No API key required.

Open Demo