How to Talk to Strangers: generating medical reports for first time users

Gkatzia, Dimitra; Rieser, Verena; Lemon, Oliver

Abstract

We propose a novel approach for handling first-time
users in the context of automatic report generation from timeseries
data in the health domain. Handling first-time users is
a common problem for Natural Language Generation (NLG)
and interactive systems in general - the system cannot adapt
to users without prior interaction or user knowledge. In this
paper, we propose a novel framework for generating medical
reports for first-time users, using multi-objective optimisation
(MOO) to account for the preferences of multiple possible
user types, where the content preferences of potential users
are modelled as objective functions. Our proposed approach
outperforms two meaningful baselines in an evaluation with
prospective users, yielding large (= :79) and medium (= :46)
effect sizes respectively.

Citation

Gkatzia, D., Rieser, V., & Lemon, O. (2016). How to Talk to Strangers: generating medical reports for first time users. In 2016 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). https://doi.org/10.1109/FUZZ-IEEE.2016.7737739

• Conference Name: FUZZ-IEEE 2016
• Start Date: Jul 24, 2016
• End Date: Jul 26, 2016
• Acceptance Date: Mar 14, 2016
• Online Publication Date: Nov 10, 2016
• Publication Date: Nov 10, 2016
• Deposit Date: Mar 16, 2016
• Publicly Available Date: Nov 10, 2016
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Book Title: 2016 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)
• ISBN: 978-1-5090-0626-7, 978-1-5090-0625-0,
• DOI: https://doi.org/10.1109/FUZZ-IEEE.2016.7737739
• Keywords: Multi-objective evolutionary algorithm; automatically generated natural language medical reports
• Public URL: http://researchrepository.napier.ac.uk/id/eprint/9673