Improving Performance of Case-Based Classification Using Context-Based Relevance

Classification involves associating instances with particular classes by maximizing intra-class similarities and minimizing inter-class similarities. Thus, the way similarity among instances is measured is crucial for the success of the system. In case-based reasoning, it is assumed that similar problems have similar solutions. The case-based approach to classification is founded on retrieving cases from the case base that are similar to a given problem, and associating the problem with the class containing the most similar cases. Similarity-based retrieval tools can advantageously be used in building flexible retrieval and classification systems. Case-based classification uses previously classified instances to label unknown instances with proper classes. Classification accuracy is affected by the retrieval process – the more relevant the instances used for classification, the greater the accuracy. The paper presents a novel approach to case-based classification. The algorithm is based on a notion of similarity assessment and was developed for supporting flexible retrieval of relevant information. Case similarity is assessed with respect to a given context that defines constraints for matching. Context relaxation and restriction is used for controlling the classification accuracy. The validity of the proposed approach is tested on real-world domains, and the system's performance, in terms of accuracy and scalability, is compared to that of other machine learning algorithms.

[1]  J. R. Quinlan.  Learning With Continuous Classes , 1992 .

[2]  Dale Schuurmans,et al.  Learning to classify incomplete examples , 1997, COLT 1997.

[3]  Derek G. Bridge,et al.  On Concept Space and Hypothesis Space in Case-Based Learning Algorithms , 1995, ECML.

[4]  Igor Jurisica,et al.  Case-based classification using similarity-based retrieval , 1996, Proceedings Eighth IEEE International Conference on Tools with Artificial Intelligence.

[5]  Tibor Kökény,et al.  Constraint Satisfaction Problems with Order-Sorted Domains , 1995, Int. J. Artif. Intell. Tools.

[6]  David W. Aha,et al.  Feature Selection for Case-Based Classification of Cloud Types: An Empirical Comparison , 1994 .

[7]  Terry Gaasterland,et al.  Restricting query relaxation through user constraints , 1993, [1993] Proceedings International Conference on Intelligent and Cooperative Information Systems.

[8]  Igor Jurisica,et al.  Supporting Flexibility. a Case-based Reasoning Approach , 1996 .

[9]  Igor Jurisica,et al.  An Efficient Approach to Iterative Browsing and Retrieval for Case-Based Reasoning , 1998, IEA/AIE.

[10]  Ray Bareiss,et al.  Concept Learning and Heuristic Classification in WeakTtheory Domains , 1990, Artif. Intell..

[11]  Igor Jurisica.  Inductive Learning and Case-Based Reasoning , 1996 .

[12]  Thomas G. Dietterich,et al.  An experimental comparison of the nearest-neighbor and nearest-hyperrectangle algorithms , 1995, Machine Learning.

[13]  David W. Aha,et al.  Learning Representative Exemplars of Concepts: An Initial Case Study , 1987 .

[14]  Igor Jurisica,et al.  How to Retrieve Relevant Information , 1994 .

[15]  James Kelly,et al.  AutoClass: A Bayesian Classification System , 1993, ML.

[16]  David W. Aha,et al.  An Implementation and Experiment with the Nested Generalized Exemplars Algorithm , 1995 .

[17]  Igor Jurisica,et al.  Applying Case-Based Reasoning to Control in Robotics , 1995 .

[18]  Peter D. Turney.  Cost-Sensitive Classification: Empirical Evaluation of a Hybrid Genetic Decision Tree Induction Algorithm , 1994, J. Artif. Intell. Res..

[19]  Thomas Rose,et al.  Task-oriented and similarity-based retrieval , 1994, Proceedings KBSE '94. Ninth Knowledge-Based Software Engineering Conference.

[20]  Igor Jurisica,et al.  A Similarity-Based Retrieval Tool for Software Repositories , 1995 .

[21]  J. Ross Quinlan,et al.  Combining Instance-Based and Model-Based Learning , 1993, ICML.

[22]  Julio Ortega,et al.  On the Informativeness of the DNA Promoter Sequences Domain Theory , 1994, J. Artif. Intell. Res..

[23]  John Mylopoulos,et al.  Case-based reasoning in IVF: prediction and knowledge mining , 1998, Artif. Intell. Medicine.

[24]  Matthias Jarke,et al.  Telos: representing knowledge about information systems , 1990, TOIS.

[25]  Igor Juri ica.  How to Retrieve Relevant Information , 1994 .

[26]  Ray Bareiss,et al.  Protos: An Exemplar-Based Learning Apprentice , 1988, Int. J. Man Mach. Stud..

[27]  I. Bratko,et al.  Information-based evaluation criterion for classifier's performance , 2004, Machine Learning.

[28]  Ryszard S. Michalski,et al.  Conceptual Clustering: Inventing Goal-Oriented Classifications of Structured Objects , 1986 .

[29]  Andrea Bonzano,et al.  An Incremental Case Retrieval Mechanism for Diagnosis , 1995 .

[30]  William Frawley,et al.  Knowledge Discovery in Databases , 1991 .

[31]  David W. Aha,et al.  Generalizing from Case studies: A Case Study , 1992, ML.

[32]  David W. Aha,et al.  Learning to Catch: Applying Nearest Neighbor Algorithms to Dynamic Control Tasks , 1994 .

[33]  Edwina L. Rissland,et al.  Case-Based Diagnostic Analysis in a Blackboard Architecture , 1993, AAAI.