Paper: Finding middle ground? Multi-objective Natural Language Generation from time-series data

ACL ID E14-4041
Title Finding middle ground? Multi-objective Natural Language Generation from time-series data
Venue Annual Meeting of The European Chapter of The Association of Computational Linguistics
Session Main Conference
Year 2014
Authors

A Natural Language Generation (NLG) system is able to generate text from non- linguistic data, ideally personalising the content to a user?s specific needs. In some cases, however, there are multiple stake- holders with their own individual goals, needs and preferences. In this paper, we explore the feasibility of combining the preferences of two different user groups, lecturers and students, when generating summaries in the context of student feed- back generation. The preferences of each user group are modelled as a multivariate optimisation function, therefore the task of generation is seen as a multi-objective (MO) optimisation task, where the two functions are combined into one. This ini- tial study shows that treating the prefer- ences of each user group equally smooths the weights o...