No abstract

Case Base Maintenance (CBM) presents one of the key factors success for Case Based Reasoning (CBR) systems. Thence, several CBM policies are proposed to improve their problem-solving performance and competence. However, to the best of our knowledge, all of them are not able to make use of prior knowledge which can be offered by domain experts, especially that CBR is widely applied in real-life domains. For instance, given symptoms of two different cases in medicine area, the doctor can affirm that these two cases should never follow the same treatment, or conversely. This kind of prior knowledge is presented in form of Cannot-Link and Must-link constraints. In addition, most of them cannot manage uncertainty in cases during CBM. To overcome this shortcoming, we propose, in this paper, a CBM policy that handles constraints to exploit experts' knowledge during case base learning along with managing uncertainty using the belief function theory. This new CBM approach consists mainly in noisy and redundant cases deletion.

Maintaining the vocabulary of case knowledge within Case Based Reasoning (CBR) presents a crucial task to ensure a high-quality problem-solving and to improve retrieval performance for large-scale CBR systems. To do, we propose, in this paper, a method that manages uncertainty while selecting the best attributes characterizing case knowledge by using belief function theory. Actually, this method is based on a new evidential attribute clustering technique to eliminate redundant and noisy attributes describing cases.

The high influence of case bases quality on Case-Based Reasoning success gives birth to an important study on cases competence for problems resolution. The competence of a case base (CB), which presents the range of problems that it can successfully solve, depends on various factors such as the CB size and density. Besides, it is not obvious to specify the exactly relationship between the individual and the overall cases competence. Hence, numerous Competence Models have been proposed to evaluate CBs and predict their actual coverage and competence on problem-solving. However, to the best of our knowledge, all of them are totally neglecting the uncertain aspect of information which is widely presented in cases since they involve real world situations. Therefore, this paper presents a new competence model called CEC-Model (Coverage & Evidential Clustering based Model) which manages uncertainty during both of cases clustering and similarity measurement using a powerful tool called the belief function theory.