Case-Based Reasoning adaptation of numerical representations of human organs by interpolation

Abstract

Case-Based Reasoning (CBR) and interpolation tools are capable of providing solutions to unknown problems through the adaptation of other problems already solved. This paper proposes a generic approach using interpolation tool during the CBR-adaptation phase. Modelling was applied to EquiVox which attempts to design three dimensional representations of human organs according to external measurements. EquiVox follows the CBR-cycle and its adaptation tool is based on Artificial Neural Networks. The performances of this adaptation tool are evaluated and discussed. The results show that EquiVox fulfills the requirements of radiation protection experts who use such prototypes while also showing the limits of use of interpolation tools in CBR-adaptation. When adaptations are guided by experience and by the trial and error of experts, interpolation tools are well-fitted methods to automatically and quickly provide adaptation strategies and knowledge through training phases.

Incorporating adaptability-related knowledge into support vector machine for case-based design adaptation

“…The design of these ML-based knowledge-light algorithms is ideally independent of the domain knowledge, or very little domain knowledge is required for making the adaptation methods. Among them, genetic algorithm (GA) based adaptation (Huang et al, 2009;Saridakis and Dentsoras, 2007;Renner and Ekárt, 2003) and neuro-adaptation (Henriet et al, 2014;Butdee, 2012;Jung et al, 2009;Craw et al, 2006;Lofty and Mohamed, 2003) are two typical knowledge-light methods, where the definition of GA and NN can be guided by the domain knowledge, but the evolution of GA or modelization of NN are not necessarily guided by domain knowledge. Overall, these investigations explore the utilization of inductive learning to acquire adaptation knowledge from examples and apply the acquired knowledge to implement automatic case adaptation.…”

Trustworthiness evaluation and retrieval-based revision method for case-based reasoning classifiers

“…Qi, Hu, and Peng (2012) proposed a new adaptation method for the solution feature values of the retrieved cases by introducing the adaptability value to improve the adaptation performance. In addition, other revised techniques based on classical methods have been developed, such as expert experience (Fan et al, 2015;Petrovic, Mishra, & Sundar, 2011;Yan and Wang et al, 2014;Yan et al, 2012), genetic algorithm (GA) (Liao et al, 2012), multiple regression analysis (Jin, Cho, Hyun, & Son, 2012), interpolation tool (Henriet, Leni, Laurent, & Salomon, 2014), regression revision model using support vector machine (Han & Cao, 2015), grey relational analysis (Hu, Qi, & Peng, 2015) and so on. In the problem solving process using CBR, to some extent, these methods have made significant contributions to the evaluation and revision of the suggested solutions for the target cases from different perspectives.…”

A novel case adaptation method based on an improved integrated genetic algorithm for power grid wind disaster emergencies

“…In addition to the practical applications in case adaptation, GRAMOGA can be 37 used as a novel method for expanding the case base. (Henrieta, Lenia, Laurenta, & Salomonb, 2014). The former obtains 65 results through adjusted models or formulas 66 ; the latter makes case adaptation a reality 67 with genetic algorithms (Liao, Hannam, Xia, & Zhao, 2012a) neural 68 networks (Callow, Lee, Blumenstein, Guan, & Loo, 2013) and k-NN 69 (Qi, Hu, & Peng, 2012).…”