Evally
Research & data

Sentiment & emotion

Sentiment, the dominant emotion, its intensity, and whether it's sarcastic.

The classic, done with calibrated probabilities: five-point sentiment, the dominant emotion, how intense it is, and a sarcasm check so 'great, another outage' doesn't count as praise.

Try it live

Real calls to Jev. Pick an example or paste your own.

Examples
100
No verdict yet

Pick an example or paste your own input, then run it.

Under the hood

Every question and rule, in the open. Fork it to change any of them.

What it asks

Reads text, then asks 4 questions in one call.

  1. Q1SentimentPick one

    What is the sentiment of the text? (If it's sarcastic, judge what the writer really means.)

    Very negativeNegativeNeutralPositiveVery positive
  2. Q2Dominant emotionPick one

    Which emotion dominates the text?

    JoyTrustFearSurpriseSadnessDisgustAngerAnticipationNone
  3. Q3IntensityScore

    How intense is the emotion in the text?

    1. 0Mild
    2. 1Moderate
    3. 2Strong
  4. Q4SarcasticYes / no

    Is the text sarcastic or ironic?

How it decides

Rules run in code, top to bottom. The first one that matches gives the verdict.

  1. 1
    Otherwisethe answer to “Sentiment”
  2. ·
    DefaultNeutral

Call it from your code

Every template runs as-is with an API key, no setup: POST /api/v1/judges/sentiment-emotion/run.

curl -X POST https://tryevally.com/api/v1/judges/sentiment-emotion/run \  -H "Authorization: Bearer $EVALLY_API_KEY" \  -H "Content-Type: application/json" \  -d '{  "input": {    "text": "Oh great, the app logged me out AGAIN in the middle of checkout. Truly a masterpiece of engineering."  }}'

Seen in the wild

Community builds on Jev that use this pattern. Not affiliated with Evally.

More in Research & data