作者: Frank Wittig
DOI: 10.1007/978-3-7091-2490-1_45
关键词: Bayesian programming 、 Variable-order Bayesian network 、 Hidden variable theory 、 Basis (linear algebra) 、 User modeling 、 Artificial intelligence 、 Machine learning 、 Context (language use) 、 Computer science 、 Bayesian network 、 Graphical model
摘要: The goal of the research summarized here is to develop methods for learning Bayesian networks on basis empirical data, focusing issues that are especially important in context user modeling. These include treatment theoretically interpretable hidden variables, ways partial and combining them into one single compound network, taking account special properties datasets acquired through psychological experiments.