Forms of Selective Attention in Intelligent Transportation Systems
Description
Selective attention is the capability exhibited by humans for selecting the relevant portions of information from the environment. It is thought to be necessary because there are too many things in the environment to perceive and respond to at once. This is the case of urban scenarios, in which the increase of the number of ubiquitous information devices such as smartphones might lead to the problem of charging humans with the superabundance of the information those devices collect about the agents - mostly human beings - and their surrounding elements - transportation systems, buildings, weather, etc., that populate those environments. Although humans exhibit natural selective mechanisms, this does not prevent them from being interrupted from whatever they are doing to deal with the information provided by those devices. This is critical when those interruptions are dangerous as happens when humans are driving and are interrupted continuously by those devices many times with irrelevant information for the task they are carrying out.
Providing those devices with an artificial selective attention mechanism that selects and communicates to the human user only the relevant information might be a solution of primary importance. The goal of this project is to build such artificial attention mechanism for those personal devices so that they select and communicate to their human holders only traveller information that is relevant for them, preventing those human agents from a superabundance of information and unnecessary interruptions. Our approach relies on the results of psychological and neuroscience studies about selective attention. We propose a computational model of selective attention that relies on the assumption that uncertain, surprising and motive congruent/incongruent information demands attention from an intelligent agent, i.e., that only cognitively and affectively, interesting/relevant information is selected and forwarded to reasoning/decision-making units.
Researchers
Funded by
FCT
Total budget
65 252,00 €
Keywords
Selective Attention, Information Triage, Alerting, Emotion, Transportation Systems
Start Date
2010-05-01
End Date
2013-10-31
Conference Articles
2013
(3 publications) - Costa, Hernani and Furtado, B. and Pires, D. and Macedo, L. and Amilcar Cardoso , "Recommending POIs Based on the User's Context and Intentions", in 11th International Conference on Practical Applications of Agents and Multi-Agent Systems, Salamanca, Spain, 2013
- Macedo, L. and Costa, Hernani and Amilcar Cardoso , "A Selective Attention-based, Multi-Agent, Travel Information System", in 16th Portuguese Conference on Artificial Intelligence, 2013
- Macedo, L. , "Arguments for a Computational Model for Forms of Selective Attention based on Cognitive and Affective Feelings", in 5th International Conference on Affective Computing and Intelligent Interaction (ACII 2013), 2013
2012
(2 publications) - Costa, Hernani and Furtado, B. and Pires, D. and Macedo, L. and Amilcar Cardoso , "Context and Intention-Awareness in POIs Recommender Systems", in 6th ACM Conference on Recommender Systems (RecSys 2012), 4th Workshop on Context-Aware Recommender Systems, (CARS 2012) Dublin, Ireland. ACM, 2012
- Macedo, L. , "A computational model for forms of selective attention based on cognitive and affective feelings", in International Conference on Cognitive Modelling, 2012
2010
(3 publications) - Macedo, L. , "Selecting Information based on Artificial Forms of Selective Attention", in 19th European Conference on Artificial Intelligence (ECAI 2010), 2010
- Macedo, L. , "The Practical Advantage of Surprise-based Agents", in 9th Int. Conf. on Autonomous Agents and Multiagent Systems (AAMAS 2010), 2010
- Macedo, L. , "A Surprise-based Selective Attention Agent for Travel Information", in 6th Workshop on Agents in Traffic and Transportation - 9th International Conference on Autonomous Agents and Multiagent Systems 2010, 2010
Tech Report