%0 Conference Paper %1 Kunzer.2001 %A Künzer, A. %A Schlick, C. %A Ohmann, F. %A Schmidt, L. %A Luczak, H. %B Proceedings of the UM2001 Workshop on Machine Learning for User Modeling (Sonthofen 2001) %C Sonthofen %D 2001 %E Schäfer, R. %E Müller, M. E. %E Macskassy, S. A. %K Benutzerverhalten Interaktion Mathematisches_Modell Mensch-Maschine-Modell mmspub SFB_368 Software-Ergonomie %P 1–10 %T An Empirical Study of Dynamic Bayesian Networks for User Modeling %U http://www.iaw.rwth-aachen.de/download/publikationen/3592_kuenzer.pdf %X Six topologies of dynamic Bayesian Networks are evaluated for predicting the future user events: (1) Markov Chain of order 1, (2) Hidden Markov Model, (3) autoregressive Hidden Markov Model, (4) factorial Hidden Markov Model, (5) simple hierarchical Hidden Markov Model and (6) tree structured Hidden Markov Model. Goal of the investigation is to evaluate, which of these models has the best fit for modeling the prediction of rule-based interaction behaviour for a real domain. Case study of the experiments is a multimodal user interface for supervisory control of advanced manufacturing cells. A group of experienced users were observed while executing a typical task, to build a data analysis basis for the evaluation. The results show that the number of user cases has high influence on the prediction quality and that there are no significant differences in using Markov Chain of order 1, factorial or tree structured Hidden Markov Models.

@inproceedings{Kunzer.2001, abstract = {Six topologies of dynamic Bayesian Networks are evaluated for predicting the future user events: (1) Markov Chain of order 1, (2) Hidden Markov Model, (3) autoregressive Hidden Markov Model, (4) factorial Hidden Markov Model, (5) simple hierarchical Hidden Markov Model and (6) tree structured Hidden Markov Model. Goal of the investigation is to evaluate, which of these models has the best fit for modeling the prediction of rule-based interaction behaviour for a real domain. Case study of the experiments is a multimodal user interface for supervisory control of advanced manufacturing cells. A group of experienced users were observed while executing a typical task, to build a data analysis basis for the evaluation. The results show that the number of user cases has high influence on the prediction quality and that there are no significant differences in using Markov Chain of order 1, factorial or tree structured Hidden Markov Models.}, added-at = {2024-09-17T10:38:40.000+0200}, address = {Sonthofen}, author = {Künzer, A. and Schlick, C. and Ohmann, F. and Schmidt, L. and Luczak, H.}, biburl = {https://www.bibsonomy.org/bibtex/2a002b69d04749abe4442f1de3207a740/sdt}, booktitle = {Proceedings of the UM2001 Workshop on Machine Learning for User Modeling (Sonthofen 2001)}, editor = {Schäfer, R. and Müller, M. E. and Macskassy, S. A.}, interhash = {d5128d72ad3135542cd55f6f16f70af0}, intrahash = {a002b69d04749abe4442f1de3207a740}, keywords = {Benutzerverhalten Interaktion Mathematisches_Modell Mensch-Maschine-Modell mmspub SFB_368 Software-Ergonomie}, pages = {1–10}, timestamp = {2024-09-17T10:38:40.000+0200}, title = {An Empirical Study of Dynamic Bayesian Networks for User Modeling}, url = {http://www.iaw.rwth-aachen.de/download/publikationen/3592_kuenzer.pdf}, year = 2001 }