Sentiment / intent analysis

Tone and intent are scored across correspondence, feedback and tickets, at document level and at the level of individual passages. Because tone is scored separately from intent, an angry message asking a simple question is not confused with a serious complaint. Trends aggregate across channels and time, which is usually more useful than any single score.

You can use the models we operate or bring your own model and we run it on your cluster. Work arrives as an API call, a batch over a folder or bucket, or a resident on-premise service, and results return as structured output with a per-job audit record.