Learning to verify design solutions from failure knowledge

1998 ◽  
Vol 12 (2) ◽  
pp. 107-115 ◽  
Author(s):  
PAULO GOMES ◽  
CARLOS BENTO ◽  
PEDRO GAGO
Keyword(s):  
Machine Learning ◽  
Experimental Results ◽  
Case Based Reasoning ◽  
Design Problems ◽  
Design Solutions ◽  
Case Based

Learning is an intrinsic product of case-based reasoning. Acquiring new cases is one possible way of learning in a case-based system. These cases comprise mainly success knowledge. The successful cases are essentially used to generate new design solutions. But a case-based system also can make use of failure knowledge. In this paper we present how a case-based system can acquire failure cases for verification of the solution created by success cases. We describe IM-RECIDE, a system that uses case-based reasoning for solving design problems that are imperfectly described and explained. The learning aspect is focused and some of the machine learning dimensions in design are criticized. Experimental results in the domain of room configuration also are presented.

scholarly journals Learning and Applying Case Adaptation Rules for Classification: An Ensemble Approach

2017 ◽  
Author(s):  
Vahid Jalali ◽  
David Leake ◽  
Najmeh Forouzandehmehr
Keyword(s):  
Machine Learning ◽  
Rule Learning ◽  
Experimental Results ◽  
Case Based Reasoning ◽  
Popular Approach ◽  
Machine Learning Methods ◽  
Reasoning Systems ◽  
Adaptation Rules ◽  
Classification Tasks ◽  
Case Based

The ability of case-based reasoning systems to solve novel problems depends on their capability to adapt past solutions to new circumstances. However, acquiring the knowledge required for case adaptation is a classic challenge for CBR. This motivates the use of machine learning methods to generate adaptation knowledge. A popular approach uses the case difference heuristic (CDH) to generate adaptation rules from pairs of cases in the case base, based on the premise that the observed differences in case solutions result from the differences in the problems they solve, so can form the basic of rules to adapt cases with similar problem differences. Extensive research has successfully applied the CDH approach to adaptation rule learning for case-based regression (numerical prediction) tasks. However, classification tasks have been outside of its scope. The work presented in this paper addresses that gap by extending CDH-based learning of adaptation rules to apply to cases with categorical features and solutions. It presents the generalized case value heuristic to assess case and solution differences and applies it in an ensemble-based case-based classification method, ensembles of adaptations for classification (EAC), built on the authors' previous work on ensembles of adaptations for regression (EAR). Experimental results support the effectiveness of EAC.

Retrieving Assembly Part Design Using Case-Based Reasoning and Genetic Algorithms

2005 ◽  
Author(s):  
Guanghsu A. Chang ◽  
Cheng-Chung Su ◽  
John W. Priest
Keyword(s):  
Artificial Intelligence ◽  
Genetic Algorithms ◽  
Nearest Neighbor ◽  
Similarity Measures ◽  
Experimental Results ◽  
Case Based Reasoning ◽  
Feature Weight ◽  
Similar Part ◽  
Case Base ◽  
Case Based

Artificial intelligence (AI) approaches have been successfully applied to many fields. Among the numerous AI approaches, Case-Based Reasoning (CBR) is an approach that mainly focuses on the reuse of knowledge and experience. However, little work is done on applications of CBR to improve assembly part design. Similarity measures and the weight of different features are crucial in determining the accuracy of retrieving cases from the case base. To develop the weight of part features and retrieve a similar part design, the research proposes using Genetic Algorithms (GAs) to learn the optimum feature weight and employing nearest-neighbor technique to measure the similarity of assembly part design. Early experimental results indicate that the similar part design is effectively retrieved by these similarity measures.

A Blackboard Architecture to Support Case-Based Reasoning in Mechanical Design

1992 ◽  
Author(s):  
Theodore Bardsz ◽  
Ibrahim Zeid
Keyword(s):  
Problem Solving ◽  
Engineering Design ◽  
Mechanical Design ◽  
Maximum Amount ◽  
Design Solution ◽  
Case Based Reasoning ◽  
Design Problems ◽  
Engineering Design Problems ◽  
Design Plan ◽  
Case Based

Abstract One of the most significant issues in applying case-based reasoning (CBR) to mechanical design is to integrate previously unrelated design plans towards the solution of a new design problem. The total design solution (the design plan structure) can be composed of both retrieved and dynamically generated design plans. The retrieved design plans must be mapped to fit the new design context, and the entire design plan structure must be evaluated. An architecture utilizing opportunistic problem solving in a blackboard environment is used to map and evaluate the design plan structure effectively and successfuly. The architecture has several assets when integrated into a CBR environment. First, the maximum amount of information related to the design is generated before any of the mapping problems are addressed. Second, mapping is preformed as just another action toward the evaluation of the design plan. Lastly, the architecture supports the inclusion of memory elements from the knowledge base in the design plan structure. The architecture is implemented using the GBB system. The architecture is part of a newly developed CBR System called DEJAVU. The paper describes DEJAVU and the architecture. An example is also included to illustrate the use of DEJAVU to solve engineering design problems.

Methods of Software Quality Prediction With Similarity Measures

2017 ◽  
pp. 34-56
Keyword(s):  
Machine Learning ◽  
Software Quality ◽  
Similarity Measures ◽  
Fault Prediction ◽  
Estimation Accuracy ◽  
Case Based Reasoning ◽  
Quality Prediction ◽  
Software Fault Prediction ◽  
Software Quality Prediction ◽  
Case Based

To improve the software quality the number of errors or faults must be removed from the software. This chapter presents a study towards machine learning and software quality prediction as an expert system. The purpose of this chapter is to apply the machine learning approaches such as case-based reasoning to predict software quality. Five different similarity measures, namely, Euclidean, Canberra, Exponential, Clark and Manhattan are used for retrieving the matching cases from the knowledgebase. The use of different similarity measures to find the best method significantly increases the estimation accuracy and reliability. Based on the research findings in this book it can be concluded that applying similarity measures in case-based reasoning may be a viable technique for software fault prediction

scholarly journals Sistem Cerdas Permainan Papan The Battle Of Honor dengan Decision Making dan Machine Learning

2021 ◽  
Vol 12 (2) ◽  
pp. 136
Author(s):  
Arnan Dwika Diasmara ◽  
Aditya Wikan Mahastama ◽  
Antonius Rachmat Chrismanto
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Decision Making ◽  
Intelligent System ◽  
Third Party ◽  
Case Based Reasoning ◽  
Board Game ◽  
Rule Based ◽  
Case Based ◽  
Rule Based Systems

Abstract. Intelligent System of the Battle of Honor Board Game with Decision Making and Machine Learning. The Battle of Honor is a board game where 2 players face each other to bring down their opponent's flag. This game requires a third party to act as the referee because the players cannot see each other's pawns during the game. The solution to this is to implement Rule-Based Systems (RBS) on a system developed with Unity to support the referee's role in making decisions based on the rules of the game. Researchers also develop Artificial Intelligence (AI) as opposed to applying Case-Based reasoning (CBR). The application of CBR is supported by the nearest neighbor algorithm to find cases that have a high degree of similarity. In the basic test, the results of the CBR test were obtained with the highest formulated accuracy of the 3 examiners, namely 97.101%. In testing the AI scenario as a referee, it is analyzed through colliding pieces and gives the right decision in determining victoryKeywords: The Battle of Honor, CBR, RBS, unity, AIAbstrak. The Battle of Honor merupakan permainan papan dimana 2 pemain saling berhadapan untuk menjatuhkan bendera lawannya. Permainan ini membutuhkan pihak ketiga yang berperan sebagai wasit karena pemain yang saling berhadapan tidak dapat saling melihat bidak lawannya. Solusi dari hal tersebut yaitu mengimplementasikan Rule-Based Systems (RBS) pada sistem yang dikembangkan dengan Unity untuk mendukung peran wasit dalam memberikan keputusan berdasarkan aturan permainan. Peneliti juga mengembangkan Artificial Intelligence (AI) sebagai lawan dengan menerapkan Case-Based reasoning (CBR). Penerapan CBR didukung dengan algoritma nearest neighbour untuk mencari kasus yang memiliki tingkat kemiripan yang tinggi. Pada pengujian dasar didapatkan hasil uji CBR dengan accuracy yang dirumuskan tertinggi dari 3 penguji yaitu 97,101%. Pada pengujian skenario AI sebagai wasit dianalisis lewat bidak yang bertabrakan dan memberikan keputusan yang tepat dalam menentukan kemenangan.Kata Kunci: The Battle of Honor, CBR, RBS, unity, AI

scholarly journals Economic models as analogies

2015 ◽  
pp. 106-130
Author(s):  
I. Gilboa ◽  
A. Postlewaite ◽  
L. Samuelson ◽  
D. Schmeidler
Keyword(s):  
Economic Models ◽  
Experimental Results ◽  
Case Based Reasoning ◽  
Rule Based ◽  
Equal Footing ◽  
Economic Problems ◽  
Sources Of Knowledge ◽  
General Rules ◽  
Case Based

People often wonder why economists analyse models whose assumptions are known to be false, while economists feel that they learn a lot from such exercises. We suggest that part of the knowledge generated by academic economists is case-based rather than rule-based. That is, instead of offering general rules or theories that should be contrasted with data, economists often analyse modelsthat are ‘theoretical cases′, which help understand economic problems by drawinganalogies between the model and the problem. Thus, economic models, empiricaldata, experimental results and other sources of knowledge are all on equal footing, that is, they all provide cases to which a given problem can be compared. We offer complexity arguments that explain why case-based reasoning may sometimes be the method of choice and why economists prefer simple cases.

Sign in / Sign up

Export Citation Format

Share Document