HBCP Corpus: A New Resource for the Analysis of Behavioural Change Intervention Reports

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 1967–1975 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 1967 HBCP Corpus: A New Resource for the Analysis of Behaviour Change Intervention Reports Francesca Bonin∗, Ailbhe N. Finnerty†, Candice Moore†, Charles Jochim∗, Emma Norris†, Yufang Hou∗, Martin Gleize∗, Debasis Ganguly∗, Alison J. Wright†, Emily Hayes†, Silje Zink†, Alessandra Pascale∗, Pol Mac Aonghusa∗, Susan Michie† ∗IBM Research, † UCL ∗ Dublin, Ireland, † London, UK {fbonin, charlesj, yhou, martin.gleize,debasis.ganguly1,apascale,aonghusa}@ibm.ie {a.finnerty, candice.moore.11, emma.norris, alison.j.wright, emily.hayes.16, s.zink, s.michie}@ucl.ac.uk Abstract Due to the fast pace at which research reports in behaviour change are published, researchers, consultants and policymakers would benefit from more automatic ways to process these reports. Automatic extraction of the reports’ intervention content, population, settings and their results etc. are essential in synthesising and summarising the literature. However, to the best of our knowledge, no unique resource exists at the moment to facilitate this synthesis. In this paper, we describe the construction of a corpus of published behaviour change intervention evaluation reports aimed at smoking cessation. We also describe and release the annotation of 57 entities, that can be used as an off-the-shelf data resource for tasks such as entity recognition, etc. Both the corpus and the annotation dataset are being made available to the community. 1. Introduction There has been significant growth in the number of sci- entific publications in many disciplines in recent times (Larsen and Ins, 2010) and researchers can have difficulty keeping track of the state of the art in many fields. This is also true for behaviour change researchers, health pro- fessionals and consultants that explore the literature of be- haviour change intervention reports, in order to understand the most effective methodology (or intervention) to help a certain population improve a specific target behaviour (for example, stopping smoking). The volume and rate at which research is produced about behaviour change is beyond the capability of human researchers to compare and understand which interventions are most effective and to be able to gen- eralise the results to varying populations in different con- texts (Michie and Johnston, 2017). More evidence is pro- duced and published than it is possible for researchers to be able to use, synthesise and analyse effectively with cur- rent conventional research methods and the current waste in research is being increasingly recognised (Glasziou et al., 2014; Elliott et al., 2014; Macleod et al., 2014). Systematic reviews seek to collate evidence that fits pre- specified eligibility criteria in order to answer a specific re- search question (Higgins and Thomas, 2019). An ongoing challenge is the time taken to complete them, estimated at ∼ 1,000 hours of highly skilled manual work (Allen and Olkin, 1999) or 67 weeks (Borah et al., 2017), from pre- registration stage to publication. It is clear from these stud- ies that the current method of conducting systematic re- views is not sustainable. For this reason both domain ex- perts in behaviour change and policymakers …

Content truncated to 500 words