Summarising News Stories for Children

Abstract

This paper proposes a system to automatically summarise news articles in a manner suitable for children by deriving and combining statistical ratings for how important, positively oriented and easy to read each sentence is. Our results demonstrate that this approach succeeds in generating summaries that are suitable for children, and that there is further scope for combining this extractive approach with abstractive methods used in text simplification.

Content Selection for Real-time Sports News Construction from Commentary Texts

“…The approach of selecting sentences to construct news reports can be treated as a special kind of document summarization (Nenkova et al, 2011). Among the large number of papers in summarization literature, some of them are based on simple definitions of sentence scoring with different components addressing different requirements in specific task settings (Yih et al, 2007;Christensen et al, 2013;Macdonald and Siddharthan, 2016, for instance), which is similar to this paper. The main difference between document summarization and the task in this study is the way to characterize importance.…”

HTSS: A novel hybrid text summarisation and simplification architecture

“…Simplification has previously been incorporated as part of the summarisation pipeline, where it is necessary to generate texts for children (Macdonald & Siddharthan, 2016), or for clinical summaries (Acharya et al, 2019). Our methods vary from these as we are using the pointer generator model with an improved loss function to model simplification.…”

Incorporating Personalization Features in a Hospital-Stay Summary Generation System

“…Research on producing personalized content has been pursued for quite some time now. Some systems are able to produce different versions of the same content, e.g., generating descriptions of devices for people of different age groups [8], while others produce a single version that is aimed at some specific audience, e.g., summarizing news stories for children [9]. Mairesse and Walker [10] developed a parameterizable language generator that takes the user's linguistic style into account and generates restaurant recommendations.…”