Can custom sentiment rules be created?

It is possible to tailor the general sentiment rules to fit a specific industry or use case. For example, if the analyzed texts are insurance-related, it might be misleading if the word “accident” was negative, while in other cases it is the expected sentiment.

We can apply custom sentiment rules in general or for certain projects individually. We can change the sentiment value of certain phrases in a specified language. It is also possible to apply stemming on the phrase (meaning the inflected forms will have the same sentiment value) or set the sentiment value of only a specific form of a phrase (e.g. have the word “cool” as positive, but “cooling” and “cooled” are neutral).