Preface The Bayesian Modeling Applications Workshop (BMAW) has been held in con- junction with the annual Conference on Uncertainty in Artificial Intelligence twelve times since 2003. The workshop brings together researchers and practi- tioners who apply the technologies pioneered by the UAI community to address important real-world problems in a diverse set of fields. The workshop fosters discussion on the challenges of building applications, such as understanding and addressing stakeholder needs; integrating Bayesian models and tools into larger applications; validating models; interacting with users; construction of models through knowledge elicitation and learning; agile model and system development strategies; and deploying and managing webbased Bayesian applications. The theme of the Workshop has adapted from year to year, as real-world problems change and technologies evolve to meet them. The frenzy to apply con- ventional machine learning methods for commercial applications has the danger of overwhelming Bayesian methods where they might be best applied. Bayesian methods face a similar challenge to the one they faced a decade ago by this community: To demonstrate their timeliness in the current environment of in- telligent systems and a long tail of related decision and prediction tasks. This Workshop demonstrates that through several tools and current applications of Bayesian methods. A call for papers encouraged submissions in a variety of domains, but not limited to any specific vertical market or discipline. Submissions were expected to foster discussion of critical issues within the community of practice. There were 12 submissions. Each submission was reviewed by at least 1, and on the average 3.1, program committee members. Eight papers were accepted and presented at the Workshop. Six of these appear full length in these proceedings. Two appear as extended abstract to facilitate future publication. Besides that, Dr. Brandon Malone has blessed the Workshop with an interesting invited talk on ”Empirical Investigations into Bayesian Network Structure Learning Algorithms”. The Twelfth Annual BMAW was held on July 16, 2015, in Amsterdam, The Netherlands. About 20 people attended the Workshop, which consisted of eleven paper presentations, questions, and the accompanying discussions. Papers and presentations addressed Bayesian learning algorithms, tools, and several appli- cations involving cost, benefit, risk, trading, forecasting, education, vehicle, and aircraft domains. We are grateful to the paper authors and presenters for their contributions, and to the program committee members for their careful efforts reviewing and commenting on submissions. We also appreciate the help Easy- Chair has always provided us and the organizational support provided by the UAI conference organizers, without whom the workshop would not be possi- ble. Finally, we also thank Dr. Brandon Malone for accepting our invitation to present an invited talk. i July 2015 John Mark Agosta Amsterdam, The Netherlands Rommel Novaes Carvalho Workshop Co-Chairs ii Table of Contents Risk, Ivestment, and Tool Project Cost, Benefit and Risk Analysis using Bayesian Networks . . . . . . . 1 Barbaros Yet, Anthony Constantinou, Norman Fenton, Martin Neil, Eike Luedeling and Keith Shepherd Bayesian Optimisation of Gated Bayesian Networks for Algorithmic Trading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Marcus Bendtsen A Tool for Visualising the Output of a DBN for Fog Forecasting . . . . . . . . 12 Tal Boneh, Xuhui Zhang, Ann Nicholson and Kevin Korb Education An IRT-based Parameterization for Conditional Probability Tables . . . . . . 14 Russell Almond Bayesian Network Models for Adaptive Testing . . . . . . . . . . . . . . . . . . . . . . . 24 Martin Plajner and Jirka Vomlel Computer Adaptive Testing Using the Same-Decision Probability . . . . . . . 34 Suming Chen, Arthur Choi and Adnan Darwiche Vehicle and Aircraft Applications Influence Diagrams for the Optimization of a Vehicle Speed Profile . . . . . . 44 Václav Kratochvı́l and Jirka Vomlel Bayesian Predictive Modelling: Application to Aircraft Short-Term Conflict Alert System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Vitaly Schetinin, Livija Jakaite and Wojtek Krzanowski iii Program Committee John Mark Agosta Microsoft Azure Russell Almond Florida State University Rommel Novaes Carvalho University of Brası́lia / Brazil’s Office of the Comptroller General Feng Chen SUNY Albany Fabio Cozman Universidade de Sao Paulo Marek Druzdzel University of Pittsburgh, Bialystok University of Technology Julia Flores Universitat de València Pablo González Instituto de Investigaciones Electricas Jim Jones George Mason University Oscar Kipersztok Boeing Branislav Kveton Adobe Research Kathryn Laskey George Mason University Manuel Luque UNED Ole Mengshoel Carnegie Mellon University Ann Nicholson Monash University Tomas Singliar Amazon.com Enrique Sucar INAOE Charles Twardy George Mason University Tom Walsh MIT iv