The Complementarity of Rules and Frecedents for Classification
ثبت نشده
چکیده
This paper describes a model of the complementarity of rules and precedents in the classification task. Under this model, precedents assist rule-based reasoning by operationalizing abstract rule antecedents. Conversely, rules assist case-based reasoning through case elaboration, the process of inferring case facts in order to increase the similarity between cases, and term reformulation, the process of replacing a term whose precedents only weakly match a case with terms whose precedents strongly match the case. Fully exploiting this complementarity requires a control strategy characterized by impartiality, the absence of arbitrary ordering restrictions on the use of rules and precedents. An impartial control strategy was implemented in GREBE in the domain of Texas worker’s compensation law. In a preliminary evaluation, GREBE’s performance was found to be as good or slightly better than the performance of law students on the same task. The Complementarity of Rules and Frecedents for Classification In a variety of domains, such as law, both general rules and specific precedents are useful for performing classification the task of assigning a given input, or case, to a category and explaining the assignment. This section explains the complementarity of rules and precedents for performing classification and the disadvantages of arbitrarily restricting the order in which they can be combined. A case is classified as belonging to a particular category by relating its description to the criteria for category membership. The justifications, or warrants (Toulmin, 1958), that can relate a case to a category, can vary widely in the generality of their antecedents. For example, consider warrants for classifying a case into the legal category “negligence.” A rule, such as “An action is negligent if the actor fails to use reasonable care and the failure is the proximate cause of an injury,” has very general antecedent terms (e.g., “breach of reasonable care”). Conversely, a precedent, such as “Dr. Jones was negligent because he failed to University of Texas at Austin Austin, Texas 78712 [email protected] count sponges during surgery and as a result left a sponge in Smith,” has very specific antecedent terms (e.g., “failure to count sponges”). Both types of warrants have been used by classification systems to relate cases to categories.
منابع مشابه
GENERATING FUZZY RULES FOR PROTEIN CLASSIFICATION
This paper considers the generation of some interpretable fuzzy rules for assigning an amino acid sequence into the appropriate protein superfamily. Since the main objective of this classifier is the interpretability of rules, we have used the distribution of amino acids in the sequences of proteins as features. These features are the occurrence probabilities of six exchange groups in the seque...
متن کاملA QUADRATIC MARGIN-BASED MODEL FOR WEIGHTING FUZZY CLASSIFICATION RULES INSPIRED BY SUPPORT VECTOR MACHINES
Recently, tuning the weights of the rules in Fuzzy Rule-Base Classification Systems is researched in order to improve the accuracy of classification. In this paper, a margin-based optimization model, inspired by Support Vector Machine classifiers, is proposed to compute these fuzzy rule weights. This approach not only considers both accuracy and generalization criteria in a single objective fu...
متن کاملOn Mining Fuzzy Classification Rules for Imbalanced Data
Fuzzy rule-based classification system (FRBCS) is a popular machine learning technique for classification purposes. One of the major issues when applying it on imbalanced data sets is its biased to the majority class, such that, it performs poorly in respect to the minority class. However many cases the minority classes are more important than the majority ones. In this paper, we have extended ...
متن کاملA hybridization of evolutionary fuzzy systems and ant Colony optimization for intrusion detection
A hybrid approach for intrusion detection in computer networks is presented in this paper. The proposed approach combines an evolutionary-based fuzzy system with an Ant Colony Optimization procedure to generate high-quality fuzzy-classification rules. We applied our hybrid learning approach to network security and validated it using the DARPA KDD-Cup99 benchmark data set. The results indicate t...
متن کاملOn Mining Fuzzy Classification Rules for Imbalanced Data
Fuzzy rule-based classification system (FRBCS) is a popular machine learning technique for classification purposes. One of the major issues when applying it on imbalanced data sets is its biased to the majority class, such that, it performs poorly in respect to the minority class. However many cases the minority classes are more important than the majority ones. In this paper, we have extended ...
متن کامل