Dynamic user modelling for personalised report generation of time-series data

The proposed work focuses on demonstrating how personal interests and preferences can be used to improve report generation from time-series data. We apply these ideas in the weather and health informatics domains. In this extended abstract, we propose a dynamic User Model approach to data-driven generation that adds personalisation to the generated reports. Our approach proposes that a User Model should be updated and influenced from both user’s traits and socially similar users taking into account the fact that these attributes may change over time. This is currently work in progress as part of a PhD by Dimitra Gkatzia and in connection with the Help4Mood project.

[1]  Anja Belz,et al.  Extracting Parallel Fragments from Comparable Corpora for Data-to-text Generation , 2010, INLG.

[2]  Richard Cox,et al.  A Bayesian Approach to Modelling Users' Information Display Preferences , 2005, User Modeling.

[3]  Ehud Reiter,et al.  Lessons from Deploying NLG Technology for Marine Weather Forecast Text Generation , 2004, ECAI.

[4]  Jim Hunter,et al.  Segmenting Time Series for Weather Forecasting , 2003 .

[5]  Dan Klein,et al.  A Simple Domain-Independent Probabilistic Approach to Generation , 2010, EMNLP.

[6]  Albert Gatt,et al.  BT-Nurse: computer generation of natural language shift summaries from complex heterogeneous medical data , 2011, J. Am. Medical Informatics Assoc..

[7]  Pablo Gervás,et al.  User-model based personalized summarization , 2007, Inf. Process. Manag..

[8]  Ehud Reiter,et al.  Types of Knowledge Required to Personalise Smoking Cessation Letters , 1999, AIMDM.

[9]  David M. Lane,et al.  Narrative monologue as a first step towards advanced mission debrief for AUV operator situational awareness , 2011, 2011 15th International Conference on Advanced Robotics (ICAR).

[10]  Ehud Reiter,et al.  Acquiring and Using Limited User Models in NLG , 2003, ENLG@EACL.

[11]  Nava Tintarev,et al.  Using NLG and Sensors to Support Personal Narrative for Children with Complex Communication Needs , 2010, SLPAT@NAACL.

[12]  Dan Klein,et al.  Learning Semantic Correspondences with Less Supervision , 2009, ACL.

[13]  Oliver Lemon,et al.  Adaptive Referring Expression Generation in Spoken Dialogue Systems: Evaluation with Real Users , 2010, SIGDIAL Conference.

[14]  Maria Klara Wolters,et al.  Managing Data in Help4Mood , 2013, EAI Endorsed Trans. Ambient Syst..

[15]  Nina Dethlefs,et al.  Combining Hierarchical Reinforcement Learning and Bayesian Networks for Natural Language Generation in Situated Dialogue , 2011, ENLG.

[16]  Albert Gatt,et al.  From data to text in the Neonatal Intensive Care Unit: Using NLG technology for decision support and information management , 2009, AI Commun..

[17]  Ehud Reiter,et al.  Generating Readable Texts for Readers with Low Basic Skills , 2005, ENLG.

[18]  Oliver Lemon,et al.  Optimising Information Presentation for Spoken Dialogue Systems , 2010, ACL.