Case-Based Reasoning relies on the underlying hypothesis that similar problems have similar solutions. The extent to which this hypothesis holds good in the case base has been used by CBR designers as a measure of case base complexity, which in turn gives insights on the generalization ability of the reasoner. Several local and global complexity measures have been proposed in the literature. However, the existing measures rely only on the similarity knowledge to compute complexity. We propose a new complexity measure called Reachability-Based Complexity Measure (RBCM) that goes beyond the similarity knowledge to include the effects of all knowledge containers in the reasoner. The proposed measure is evaluated on several real-world datasets and results suggest that RBCM corroborates well with the generalization accuracy of the reasoner.

We study the problem of incorporating prior knowledge into a deep Transformer-based model,i.e.,Bidirectional Encoder Representations from Transformers (BERT), to enhance its performance on semantic textual matching tasks. By probing and analyzing what BERT has already known when solving this task, we obtain better understanding of what task-specific knowledge BERT needs the most and where it is most needed. The analysis further motivates us to take a different approach than most existing works. Instead of using prior knowledge to create a new training task for fine-tuning BERT, we directly inject knowledge into BERT's multi-head attention mechanism. This leads us to a simple yet effective approach that enjoys fast training stage as it saves the model from training on additional data or tasks other than the main task. Extensive experiments demonstrate that the proposed knowledge-enhanced BERT is able to consistently improve semantic textual matching performance over the original BERT model, and the performance benefit is most salient when training data is scarce.