Integrating kNN Retrieval with Inference on Graphical Models in Case-Based Reasoning

In Case-Based Reasoning, when the similarity assumption does not hold, knowledge about the adaptability of solutions has to be exploited, in order to retrieve cases with adaptable solutions.We propose a novel approach to address this issue, where kNN retrieval is integrated with inference on a metric Markov Random Field (MRF). Nodes of the MRF represent cases and edges connect nodes whose solutions are close in the solution space. States of the nodes represent different adaptation levels with respect to the potential query. Metric-based potentials enforce connected nodes to share the same state, since cases having similar solutions should share the same adaptability effort with respect to the query. The goal is to enlarge the set of potentially adaptable cases that are retrieved, by controlling precision and accuracy of retrieval. We experiment on a retrieval architecture where a simple kNN retrieval (on the problem description) is followed by a further retrieval step based on MRF inference, and we discuss the promising results we have obtained in two different setting: using manually-engineered adaptation rules and adopting an automatic learning strategies for such rules.