A Research Project on Inductive Situation Evolution Modeling
Cognitive Situation Management | Situation Awareness | High-level Information Fusion | Artificial Intelligence | Intelligent Information Systems | Software Engineering
Project Title: Inductive Situation Evolution Modeling (inSiTUEVO)
Investigator: Dr. Andrea Salfinger
Institution: Johannes Kepler University Linz, Austria
Duration: Oct. 2018 - Jan. 2022
Funding: Austrian Science Fund (FWF), Hertha-Firnberg Program
Type: single person project (post-doc career development program)
inSiTUEVO is a research project in the area of Cognitive Situation Management. Broadly speaking, Cognitive Situation Management involves the problem of how machines can not only perceive the individual elements of their observed environment, but understand the whole "big picture", i.e., encountered situations, by interrelating these individual perceptions. To do so, artifical agents require suitable situation models representing the event patterns of interest, against which they can compare their sensed observations.
inSiTUEVO aims at supporting the acquisition of these situation models, by developing novel Knowledge Discovery & Data Mining approaches to automatically induce symbolic situation evolution models from data.
Oct. 13, 2020: Chaired the session "Knowledge Acquisition in Intelligent Systems" at the virtual SMC 2020.
Sep. 12, 2020: Presented my work on "Deep Learning for Cognitive Load Monitoring: A Comparative Evaluation" at UbiTtention 2020 (virtual workshop in conjunction with UbiComp 2020).
Aug. 25, 2020: Gave my talk on "Reinforcement Learning Meets Cognitive Situation Management: A Review of Recent Learning Approaches from the Cognitive Situation Management Perspective" at CogSIMA 2020 (virtual conference).
Aug. 2020: News article about this project in the Austrian daily newspaper "Die Presse"!
July 2020: Paper on deep learning for cognitive load monitoring accepted at UbiTtention 2020!
July 6-9, 2020: Presenting our paper "Towards Neural Situation Evolution Modeling: Learning a Distributed Representation for Predicting Complex Event Sequences" at the virtual FUSION 2020. Recorded talk here (registered attendees only) and summary here.
May 2020: Paper "Towards Neural Situation Evolution Modeling [...]" accepted at FUSION 2020, joint work with Prof. Lauro Snidaro (University of Udine, IT)!
Feb. 2020: Review paper on Reinforcement Learning (from the Cognitive Situation Management perspective) accepted at CogSIMA 2020!
Oct. 2019: Very excited to serve as Co-Chair on the IEEE SMC Society Technical Committee for Cognitive Situation Management, joining Prof. Leo Motus and Prof. Christian Lebiere!
Aug. 2019: Co-organizing a special session on "Computational Models of Cognition and Situated Behavior in Cyber-Physical-Human Systems" with Dr. Gabe Jakobson, Dr. Kellyn Rein and Dr. Giuseppe D'Aniello at SMC'19!
July 3, 2019: Gave my talk on "Framing Situation Prediction as a Sequence Prediction Problem: A Situation Evolution Model Based on Continuous-Time Markov Chains" at FUSION 2019 in Ottawa, Canada.
May 2019: Publication on situation prediction accepted at FUSION 2019!
April 9, 2019: Gave my talk on "Situation Mining: Event Pattern Mining for Situation Model Induction" at CogSIMA 2019 in Las Vegas, USA.
Feb. 2019: First project publication on Situation Mining accepted at CogSIMA 2019!
Oct. 2018: Project started.