Handling Default Data under a Case-based Reasoning Approach Bruno Fernandes 1 , Mauro Freitas 1 , Cesar Analide 1 , Henrique Vicente 2 and José Neves 1 1 Department of Informatics, University of Minho, Braga, Portugal 2 Department of Chemistry, University of Évora, Évora, Portugal Keywords: Case-based Reasoning, Intelligent Systems, Similarity Analysis. Abstract: The knowledge acquired through past experiences is of the most importance when humans or machines try to find solutions for new problems based on past ones, which makes the core of any Case-based Reasoning approach to problem solving. On the other hand, existent CBR systems are neither complete nor adaptable to specific domains. Indeed, the effort to adapt either the reasoning process or the knowledge representation mechanism to a new problem is too high, i.e., it is extremely difficult to adapt the input to the computational framework in order to get a solution to a particular problem. This is the drawback that is addressed in this work. 1 INTRODUCTION Case-Based Reasoning (CBR) provides the ability of solving new problems by reusing knowledge acquired from past experiences. In fact, it can be described as the process of solving new problems based on solutions of similar past ones. One well- known example is the mechanic that listening to the strange sound that the engine is making and the symptoms of the problem, uses his/her past experiences to make a first analysis of the problem, and uses a solution that has already been castoff on similar situations. This technique can be used in domains where there is a large amount of information that has been acquired through experience (Aamodt and Plaza, 1994). CBR is used, especially when similar cases have similar terms and solutions, even when they have different backgrounds, i.e., the knowledge acquired when solving some situation can be used as a first approach to solve new ones (Balke T. et al, 2009). Nowadays, CBR is used in several areas and has a tremendous potential. There are examples of its use in The Law with respect to dispute resolution (Carneiro D. et al, 2009; Carneiro D. et al, 2010), in medicine (Kolodner J., 1992; Khan A. and Hoffman A., 2002), among others. A typical, and maybe the greatest, problem in the use of CBR is related to the availability of the data and the cost of obtaining it (Stahl A. and Gabel T., 2006). The typical CBR cycle (Figure 1) clearly defines the vocabulary of the domain and the steps that should be followed to have a consistent model. The initial description of the problem becomes a new case that is used to Retrieve one or more cases from the repository. At this stage it is important to identify the characteristics of the new problem and retrieve cases with a higher degree of similarity to it. Then, a solution for the problem emerges when combining the new case with the retrieved case, on the Reuse phase. Here, the solution of the retrieved case is reused, tested and adapted, if possible, to the new case, thus creating the solved case that is suggested as a solution (Aamodt and Plaza, 1994). However, when adapting the solution it is essential to have feedback from the user since automatic adaptation in existing systems is almost impossible. Currently, the main adaptation techniques are the null adaptation (no adaptation is needed); replace the attributes of the solution and not the structure of the solution; use rules; and by analogy transfer the knowledge of the past problems to the new ones and with that generate the solution (Kolodner J., 1992). It is on the Revise stage that the suggested solution is tested and it is here where exists an interaction with the user letting him/her correct/adapt/change the suggested solution by creating the Test Repaired Case that sets the solution of the new problem. The test repaired case must be correctly tested to ensure that the solution is indeed correct. This is iterative since the solution must be tested and adapted while the result of applying that solution is unsatisfying. 294 Fernandes B., Freitas M., Analide C., Vicente H. and Neves J.. Handling Default Data under a Case-based Reasoning Approach. DOI: 10.5220/0005184602940304 In Proceedings of the International Conference on Agents and Artificial Intelligence (ICAART-2015), pages 294-304 ISBN: 978-989-758-074-1 Copyright c  2015 SCITEPRESS (Science and Technology Publications, Lda.)