![23/3/18 3
MOTIVATIONAL MESSAGES FOR E-COACHING
MAIN CHALLENGES
Represent the principal, and perhaps more natural,
means for translating behavioral findings into easy-to-
follow and realizable recommendations (actions)
KEY challenges:
Generation of relevant messages tailored to the
performance, needs and characteristics of each
specific user [Noar2007]
Fostering the diversity of the messages to
increase adherence and make the coaching system
more realistic and trustworthy [opdenAkker2015]
Noar et al. Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological bulletin 133, 4 (2007),
673.
op den Akker et al.Tailored motivational message generation: A model and practical framework for real-time physical activity coaching. Journal of
Biomedical Informatics 55 (2015), 104-115.](https://image.slidesharecdn.com/villalongaucami2017slides-180323134135/75/Automatic-mapping-of-motivational-text-messages-into-ontological-entities-for-smart-coaching-applications-3-2048.jpg)
Uploaded byOresti Banos
Automatic mapping of motivational text messages into ontological entities for smart coaching applications
The document discusses a framework for the automatic mapping of motivational text messages into ontological entities for smart coaching applications, addressing the challenges of generating relevant and tailored messages. It outlines a linguistic analysis approach for breaking down and inferring the semantics of motivational messages, ultimately contributing to a new method for creating personalized coaching recommendations. Future work includes evaluating and improving the mapping method, as well as expanding the ontology to enhance message generation capabilities.
Automatic mapping of motivational text messages into ontological entities for smart coaching applications
- 1. AUTOMATIC MAPPING OF MOTIVATIONALTEXT MESSAGES INTO ONTOLOGICAL ENTITIES FOR SMART COACHING APPLICATIONS C. VILLALONGA(*), H. OP DEN AKKER, H. HERMENS, L.J. HERRERA, H. POMARES, I. ROJAS, O. VALENZUELA, O. BANOS (*) claudia.villalonga@unir.net
- 2. 23/3/18 2 E-COACHING WHY ISIT THAT CHALLENGING? Major efforts have been mainly targeted at improving the quantification part (oversimplification vs exaggeration) “Generic” motivational messages (one- size-does-not-fit-all) Shallow recommendations (explanation of goals, how to reach them, implications, etc.) We are still learning…
- 3. 23/3/18 3 MOTIVATIONAL MESSAGESFOR E-COACHING MAIN CHALLENGES Represent the principal, and perhaps more natural, means for translating behavioral findings into easy-to- follow and realizable recommendations (actions) KEY challenges: Generation of relevant messages tailored to the performance, needs and characteristics of each specific user [Noar2007] Fostering the diversity of the messages to increase adherence and make the coaching system more realistic and trustworthy [opdenAkker2015] Noar et al. Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological bulletin 133, 4 (2007), 673. op den Akker et al.Tailored motivational message generation: A model and practical framework for real-time physical activity coaching. Journal of Biomedical Informatics 55 (2015), 104-115.
- 4. 23/3/18 4 FRAMEWORK OFMOTIVATIONAL MESSAGES PROPOSED SOLUTION
- 5. 23/3/18 5 FRAMEWORK OFMOTIVATIONAL MESSAGES MOTIVATIONAL MESSAGE ONTOLOGY “You should walk the dog to the park early in the morning”
- 6. 23/3/18 6 FRAMEWORK OFMOTIVATIONAL MESSAGES AUTOMATIC MAPPING OF MOTIVATIONAL TEXT MESSAGES “You should walk the dog to the park early in the morning” Automatic Mapping
- 7. 23/3/18 7 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES LINGUISTIC ANALYSIS “You should walk the dog to the park early in the morning” predicatesubject Part of Speech pronoun verb verb article article article noun noun noun adverb preposition preposition Grammatical Structure object modifiermodifier
- 8. 23/3/18 8 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES LINGUISTIC ANALYSIS “You should walk the dog to the park early in the morning” predicatesubject pronoun verb verb article article article noun noun noun adverb preposition preposition object modifiermodifier ACTION LOCATION TIMEELEMENT
- 9. 23/3/18 9 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES ONTOLOGICAL REPRESENTATION ACTION PLACE TIMEELEMENT “You should walk the dog to the park early in the morning”
- 10. 23/3/18 10 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES ONTOLOGICAL REPRESENTATION “You should walk the dog to the park early in the morning” TIME ACTION ELEMENT PLACE
- 11. 23/3/18 11 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES MESSAGE SPLITTING “Walk or run to the park!” “Walk to the park!” “Run to the park!”
- 12. 23/3/18 12 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES MESSAGE INFERENCE “Why don’t you go to the gym and practice some exercise?” “Why don’t you go to the park and practice some exercise?” Ontology reasoning
- 13. 23/3/18 13 AUTOMATIC MAPPINGOF MOTIVATIONAL TEXT MESSAGES IMPLEMENTATION Java Implementation Stanford CoreNLP Apache Jena (v2.11.2)
- 14. 23/3/18 14 CONCLUSIONS This workcontributes with: A new approach for automatically extracting the semantics of motivational messages and creating the ontological representation of these messages Future work: Evaluation of the message mapping method Improvement of the method to infer new messages based on the knowledge modeled in the ontology Extension of the motivational message ontology to include more concepts by linking available ontologies and thesaurus Implementation of the message retrieval method to have a fully functional framework for the automatic generation of tailored coaching messages
- 15. 23/3/18 15 Thank you! Questions? Thisproject has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 769553.
Editor's Notes
- #8 temporal modifier