Learning Solution Similarity in Preference-Based CBR
Learning Solution Similarity in Preference-Based CBR
复制标题
DOI：
10.1007/978-3-319-11209-1_3
复制
发表时间：
2014-09
期刊：
影响因子：
--
通讯作者：
Amira Abdel-Aziz;M. Strickert;Eyke Hüllermeier
中科院分区：
文献类型：
--
作者：
Amira Abdel-Aziz;M. Strickert;Eyke Hüllermeier
本文是我们最近的工作基于偏好的CBR，或Pref-CBR简称的延续。后者被认为是一种基于案例的推理方法，其中解决问题的经验表示在上下文的偏好，即在要解决的目标问题的上下文中的候选解决方案的偏好的形式。在我们的Pref-CBR框架中，基于案例的问题解决被形式化为候选解决方案空间中的偏好引导搜索过程，该过程配备了相似性（或等效的距离）度量。由于Pref-CBR的有效性受到这种措施的充分性的影响，我们提出了一种学习方法，用于根据CBR系统在时间过程中收集的经验来适应解决方案的相似性。更具体地说，我们的方法利用一个潜在的概率模型，并实现适应贝叶斯推理。该方法的有效性说明了一个案例研究，处理基于案例的推荐的红葡萄酒。
This paper is a continuation of our recent work on preference-based CBR, or Pref-CBR for short. The latter is conceived as a case-based reasoning methodology in which problem solving experience is represented in the form of contextualized preferences, namely preferences for candidate solutions in the context of a target problem to be solved. In our Pref-CBR framework, case-based problem solving is formalized as a preference-guided search process in the space of candidate solutions, which is equipped with a similarity (or, equivalently, a distance) measure. Since the efficacy of Pref-CBR is influenced by the adequacy of this measure, we propose a learning method for adapting solution similarity on the basis of experience gathered by the CBR system in the course of time. More specifically, our method makes use of an underlying probabilistic model and realizes adaptation as Bayesian inference. The effectiveness of this method is illustrated in a case study that deals with the case-based recommendation of red wines.