Chang Rui, Brauer Wilfried, Stetter Martin
Department of Computer Science, Technical University of Munich, Germany.
Neural Netw. 2008 Mar-Apr;21(2-3):182-92. doi: 10.1016/j.neunet.2007.12.042. Epub 2007 Dec 31.
We propose a novel framework for performing quantitative Bayesian inference based on qualitative knowledge. Here, we focus on the treatment in the case of inconsistent qualitative knowledge. A hierarchical Bayesian model is proposed for integrating inconsistent qualitative knowledge by calculating a prior belief distribution based on a vector of knowledge features. Each inconsistent knowledge component uniquely defines a model class in the hyperspace. A set of constraints within each class is generated to describe the uncertainty in ground Bayesian model space. Quantitative Bayesian inference is approximated by model averaging with Monte Carlo methods. Our method is firstly benchmarked on ASIA network and is applied to a realistic biomolecular interaction modeling problem for breast cancer bone metastasis. Results suggest that our method enables consistently modeling and quantitative Bayesian inference by reconciling a set of inconsistent qualitative knowledge.
我们提出了一种基于定性知识进行定量贝叶斯推理的新颖框架。在此，我们关注定性知识不一致情况下的处理方法。提出了一种分层贝叶斯模型，通过基于知识特征向量计算先验信念分布来整合不一致的定性知识。每个不一致的知识组件在超空间中唯一地定义一个模型类。在每个类中生成一组约束来描述基础贝叶斯模型空间中的不确定性。通过蒙特卡罗方法进行模型平均来近似定量贝叶斯推理。我们的方法首先在ASIA网络上进行基准测试，并应用于乳腺癌骨转移的实际生物分子相互作用建模问题。结果表明，我们的方法能够通过协调一组不一致的定性知识来实现一致的建模和定量贝叶斯推理。
