FinnSentiment

FinnSentiment estimates a sentiment (positive, negative, or neutral) for each sentence in the input text, and also for the input text as a whole.

The sentiment analysis relies on three resources:

1. Word embeddings calculated from a corpus of Finnish text.
2. Product reviews harvested from the Internet.
3. A word-based convolutional neural network with 100 kernels each of sizes 2, 3, 4 and 5 words. The neural network is trained to predict the rating associated with product reviews, and the prediction it gives to the input text is converted to a sentiment.

At the moment this tool is available as a demo version.

This resource group page has a Persistent Identifier: http://urn.fi/urn:nbn:fi:lb-2021110405

Last modified on 2026-02-18