Evaluation of a chatbot for ontology learning

Dear all,
during my bachelor thesis at Bielefeld University I have built a chatbot
(named RLBot) that is intended to enable a diverse crowd of annotators to
contribute to a domain-specific ontology.
No prior knowledge of any kind is required except speaking English. The
only Software needed is a somewhat recent web browser.
Together with RLBot you will label single sentences. You will be asked
simple questions aimed at finding objects and classes in these sentences.
(Thus, not everything that could be modeled in an ontology is learned via
the chatbot — this is a feasibility study with limited scope.)
From prior testing, labeling one sentence takes about 10 minutes, you can
label as many sentences as you like and quit at any point.

No personal data will be collected by the bot itself, and the collected
annotations will be deleted shortly after I have handed in my thesis on
September 1st. Hosting of all components and processing of the chatbot
interactions is done with Google Firebase. The chatbot is realized with
Google Dialogflow.

If you have any questions or feedback feel free to shoot me an email.

The prototype is available here:
https://bachelorarbeit-308416.web.app/

Best Regards and thanks for your help in advance

-- 

Jean Finas

openpgp4fpr:e4ebcb6e3026e0fdb3a671ddd21bc9d2f69596a2