Matches in ScholarlyData for { ?s ?p ?o. }
- 55 sameAs 55.
- 55 abstract "We study classification over a slow stream of complex objects like customers or students. The learning task must take into account that an object's label is influenced by incoming data from adjoint, fast streams of transactions, e.g. customer purchases or student exams, and that this label may even change over time. This task involves combining the streams, and exploiting associations between the target label and attribute values in the fast streams. We propose a method for the discovery of classification rules over such a confederation of streams, and we use it to enhance a decision tree classifier. We show that the new approach has competitive predictive power while building much smaller decision trees than the original classifier.".
- 55 hasAuthorList authorList.
- 55 isPartOf proceedings.
- 55 keyword "classification rules".
- 55 keyword "propositionalisation".
- 55 keyword "relational mining".
- 55 keyword "stream mining".
- 55 title "Classification Rule Mining for A Stream of Perennial Objects".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasLastItem item-2.
- item-1 type ListItem.
- item-1 hasContent zaigham-faraz-siddiqui.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent myra-spiliopoulou.
- item-2 previous item-1.
- 56 creator adam-ghandar.
- 56 creator ralf-zurbruegg.
- 56 creator zbigniew-michalewicz.
- 56 type InProceedings.
- 56 label "A Case for Learning Simpler Rule Sets with Multiob jective Evolutionary Algorithms".
- 56 sameAs 56.
- 56 abstract "Fuzzy rules can be understood by people because of their specification in structured natural language. In a wide range of decision support applications in business, the interpretability of (fuzzy) rule based systems is a distinguishing feature, and advantage, over possible alternate approaches that are perceived as a “black box”, for example in facilitating accountability. The motivation of this paper is to consider the relationships between rule simplicity (the key component of interpretability) and out-of-sample performance. Forecasting has been described both art and science to emphasize intuition and experience aspects of the process: both are aspects of intelligence notoriously difficult to reproduce artificially. We explore, computationally, the widely appreciated forecasting “rule-of-thumb” expressed in Ockham’s principle that “simpler explanations are more likely to be correct”.".
- 56 hasAuthorList authorList.
- 56 isPartOf proceedings.
- 56 keyword "Evolutionary Computation".
- 56 keyword "Fuzzy Systems".
- 56 keyword "Multiobjective optimization".
- 56 title "A Case for Learning Simpler Rule Sets with Multiob jective Evolutionary Algorithms".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasLastItem item-3.
- item-1 type ListItem.
- item-1 hasContent adam-ghandar.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent zbigniew-michalewicz.
- item-2 next item-3.
- item-2 previous item-1.
- item-3 type ListItem.
- item-3 hasContent ralf-zurbruegg.
- item-3 previous item-2.
- 57 creator antoni-ligeza.
- 57 creator grzegorz-j-nalepa.
- 57 creator krzysztof-kaczor.
- 57 type InProceedings.
- 57 label "Overview of Knowledge Formalization with XTT2 Rules".
- 57 sameAs 57.
- 57 abstract "The paper discusses a new knowledge representation for rule-based systems called XTT2. This hybrid knowledge representation combines decision diagrams with extended decision tables. A single decision table contains a set of rules of similar structure operating within a common context. The decision diagram takes the form of a graph linking the tables in a kind of decision network. Hence, the overall structure of XTT2 constitutes a kind of hierarchical knowledge representation. At the lower level knowledge units specification is provided by a set of rules working in the same context and represented in a single table, and, at the~higher level, where the decision diagram defines the overall structure of the knowledge base. This model has a concise formalization which opens up possibility of rigorous design and verification. The focus of the paper is on the detailed presentation of the formal aspects of the approach starting at initial logical specification and ending with practical tools. ".
- 57 hasAuthorList authorList.
- 57 isPartOf proceedings.
- 57 keyword "formalization".
- 57 keyword "rules".
- 57 keyword "visual design".
- 57 title "Overview of Knowledge Formalization with XTT2 Rules".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasLastItem item-3.
- item-1 type ListItem.
- item-1 hasContent grzegorz-j-nalepa.
- item-1 next item-2.
- item-2 type ListItem.
- item-2 hasContent antoni-ligeza.
- item-2 next item-3.
- item-2 previous item-1.
- item-3 type ListItem.
- item-3 hasContent krzysztof-kaczor.
- item-3 previous item-2.
- 60 creator andreas-hamfelt.
- 60 creator giacomo-aceto.
- 60 creator jenny-eriksson-lundstroem.
- 60 type InProceedings.
- 60 label "A dynamic metalogic argumentation framework implementation".
- 60 sameAs 60.
- 60 abstract "One of the main challenges that faces the AI-community is to express close approximations of human reasoning as computational formalizations of argument. In this paper we present a full implementation and accompanying software for defeasible adversarial argumentation. The work is based on the metalogic framework of defeasible adversarial argumentation games of [9]. The software we developed consists of: a meta-interpreter, a declarative implementation of the argumentation game model and a graphical interface developed in Java that shows the results of the game execution and the construction of the argumentation derivation tree.".
- 60 hasAuthorList authorList.
- 60 isPartOf proceedings.
- 60 keyword "Argumentation games".
- 60 keyword "Defeasible argumentation software".
- 60 keyword "Defeasible rules".
- 60 keyword "Game trees".
- 60 keyword "Metalogic programming".
- 60 title "A dynamic metalogic argumentation framework implementation".
- authorList type List.
- authorList hasFirstItem item-1.
- authorList hasItem item-1.
- authorList hasItem item-2.
- authorList hasItem item-3.
- authorList hasLastItem item-3.
- item-1 type ListItem.