Narrow Results

Case retrieval is one of the key steps of case-based reasoning. The quality of case retrieval determines the effectiveness of the system. The common similarity calculation methods based on attributes include distance and inner product. Different similarity calculations have different influences on the effect of case retrieval. How to combine different similarity calculation results to get a more widely used and better retrieval algorithm is a hot issue in the current case-based reasoning research. In this paper, the granularity of quotient space is introduced into the similarity calculation based on attribute, and a case retrieval algorithm based on granularity synthesis theory is proposed. This method first uses similarity calculation of different attributes to get different results of case retrieval, and considers that these classification results constitute different quotient spaces, and then organizes these quotient spaces according to granularity synthesis theory to get the classification results of case retrieval. The experimental results verify the validity and correctness of this method and the application potential of granularity calculation of quotient space in case-based reasoning.

In Case-Based Reasoning (CBR), case retrieval is generally guided by similarity. However, the most similar case may not be the easiest one to reuse (hard to adapt). As recommended by Smyth and Keane, it might be more efficient to use an adaptability criterion to guide the retrieval process (adaptation-guided retrieval or AGR). In the same trend but with the goal of optimizing case reuse, our approach is to consider what is similar to copy and what is different to adapt during the retrieval stage. We introduce a more general framework for retrieval, namely the reuse-guided retrieval (RGR). The goal of this paper is twofold: first, it proposes a case retrieval approach that relies on reuse cost; then, it illustrates its use by integrating adaptation cost into the case retrieval net (CRN) memory model, a similarity-based case retrieval system. The described retrieval framework optimizes case reuse early in the inference cycle, without incurring the full cost of an adaptation step. Our results on two case studies reveal that the proposed approach yields better recall quality than CRN's similarity only-based retrieval while having similar computational complexity.

In this paper we evaluate the usefulness of seeding genetic algorithms (GAs) from a case-base. This is motivated by the expectation that the seeding will speed up the GA by starting the search in promising regions of the search space. We evaluate this case-based seeding on popular GA solutions to the Travelling Salesman Problem (TSP) and the Job-Shop Scheduling Problem (JSSP). We find that seeding the initial population gives the GA a head start and that this advantage is maintained to the end of the GA in both problem domains. In addition, we have also found that the scale of the initial advantage can be greatly improved through the introduction of problem specific information in the seeding process. Finally, the improvements of the seeded GA over the random GA can be used to reduce the number of generations required to find suitably fit solutions.

As existing similarity algorithms are hampered by low accuracy and recall rates, a comprehensive ontology-based weighted similarity algorithm was proposed for case retrieval. Taking into account the various attributes of the case, different attribute calculation methods were used and each attribute was weighted in accordance and the similarity of the case was calculated. The similarity algorithm was validated by three psychological counseling cases. By comparing the traditional similarity algorithm based on the nearest distance and the weighted similarity algorithm, this paper demonstrates that the weighted similarity algorithm exhibit higher accuracy.