Case-Base Injection Schemes to Case Adaptation Using Genetic Algorithms

Abstract

Abstract. Case adaptation has always been a difficult process to engineer within the case-based reasoning (CBR) cycle. To combat the difficulties of CBR adaptation, such as its domain dependency, computational cost and the inability to produce novel cases to solve new problems, genetic algorithms (GAs) have been applied to CBR adaptation. As the quality of cases stored in a case library has a significant effect on the solutions produced by a case-based reasoner, it is important to investigate the impact of the quality and quantity of cases injected into a GA initial population for adapting fitter solutions to new problems. This work explores a method applying a GA to CBR adaptation, where a learning mechanism is applied to feed knowledge back from the CBR revision stage into the reuse stage, allowing the GA to learn which mutations result in invalid solutions. In collaboration with this learning mechanism, the number of cases to be injected, and the fitness of cases to be injected from retrieval into reuse is explored. The fitness of adapted cases and their response to our developed learning feedback is also trialled through varying the size and quality of the GA initial population.

Information Reuse and Stochastic Search

“…Some of this work demonstrates the benefits of plan reuse, such as by reusing parts of existing plans that target particular goals in new situations that share those goals [54]; concurrently executing and dynamically switching between plans designed to handle contingencies [4]; modifying plans produced under assumed optimal conditions to handle common problems found in simulation [34]; or iteratively transforming simple plans to produce complex plans [1]. Case-based plan adaptation [42] explicitly reuses past plans in new contexts, in which context GAs have been explored directly [19,35], e.g., by injecting solutions to previous problems into a GA population to speed the solution of new problems. Although the mechanism is similar, our approach is importantly novel in that it addresses a broader class of uncertainty.…”

“…One potential way to respond to unanticipated adaptation needs is to automatically reuse or adapt prior knowledge to new situations. Indeed, research in artificial intelligence [1,54] and casebased reasoning [19,35,42] has explored the potential of plan reuse, using knowledge contained in previously-created plans to speed the synthesis of new plans in response to unanticipated changes. However, the self-* context poses unresolved domain-specific challenges, since these systems must autonomously respond to uncertainty from a number of sources throughout the adaptation cycle.…”

Integrating multiple adaptation results by utilization of support vector regression in case-based mechanical product design

“…However, the parametric design is a complex problem when there are massive parameters in the process of design, and the utilization of classical rule-based adaptation method in this situation demands a significant knowledge engineering effort to capture abundant adaptation rules. This prompted some studies to research machine learning-based adaptation under k-NN principle, and several learning methods have been employed in this area, for example, neural networks, [19][20][21][22]36,37 SVR, 38 genetic algorithm, 12,[14][15][16] and partial-order planning. 39 But insufficient knowledge badly affects the selection of an appropriate machine learning algorithm and its performance in featureoriented adaptation.…”

“…One is intelligent feature-oriented adaptation (IFA) based on various machine learning models; the other is hybridization of intelligent techniques with statistical adaptation methods. Typical approaches for the first case are gene adaptation [12][13][14][15][16][17] and neuroadaptation. [18][19][20][21][22] For the second case, Huang et al 13 and Qi et al 23 introduced the adjusting parameters in statistical adaptation models and used genetic algorithm and decision tree, respectively, to obtain the optimized adjusting parameters.…”

Case-Based Plan Adaptation: An Analysis and Review

“…CHEF constructs cooking recipes that are viewed as plans, because recipes constitute sequences of cooking steps such as adding ingredients. Since then, CBP has been a frequent research topic that includes a wide range of application areas such as manufacturing (Costas & Kashyap, 1993), military planning (Veloso, Mulvehill & Cox, 1997), route planning (Haigh, Shewchuk & Veloso, 1997), academic course scheduling (Grech & Main, 2004), and medicine (Salem, Nagaty, & El Bagoury, 2003;Schmidt, Vorobieva, & Gierl , 2003).…”