Rather than identifying a method by merely using its name,
our test users encouraged us to provide concrete examples for the individual methods.
Some of the testers noted that there is no clear statement w.r.t. to the reliability of the suggested methods.
Furthermore, our users suggested the integration of a feature that supports the anonymous evaluation of previous suggestions made by the recommendation application.
Most users noticed that our novel recommendation model is able to discriminate between similar methods depending on concrete project scenarios and thus confirmed the adequacy of the model.
However,
two users complained that the system’s decision procedure is not precise enough to be employed for everyday use.