DisBot: A Portuguese Disaster Support Dynamic Knowledge Chatbot
Chatbot
Knowledge graph
DOI:
10.3390/app10249082
Publication Date:
2020-12-18T16:42:09Z
AUTHORS (4)
ABSTRACT
This paper presents DisBot, the first Portuguese speaking chatbot that uses social media retrieved knowledge to support citizens and first-responders in disaster scenarios, order improve community resilience decision-making. It was developed tested using Design Science Research Methodology (DSRM), being progressively matured with field specialists through several design development iterations. DisBot a state-of-the-art Dual Intent Entity Transformer (DIET) architecture classify user intents, makes use of dialogue policies for managing conversations, as well storing relevant information be used further turns. To generate responses, it real-world safety knowledge, infers dynamic graph is dynamically updated real-time by disaster-related extraction tool, presented previous works. Through its iterations, has been validated specialists, who have considered valuable asset management.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (33)
CITATIONS (13)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....