Comparison of Lazy Classification Algorithms Based on Deterministic and Inhibitory Decision Rules
9
Citation
13
Reference
10
Related Paper
Citation Trend
Abstract:
In the paper, two lazy classification algorithms of polynomial time complexity are considered. These algorithms are based on deterministic and inhibitory decision rules, but the direct generation of rules is not required. Instead of this, for any new object the considered algorithms extract from a given decision table efficiently some information about the set of rules. Next, this information is used by a decision-making procedure. The reported results of experiments show that the algorithms based on inhibitory decision rules are often better than those based on deterministic decision rules.Keywords:
Decision table
Decision rule
Decision system
Decision table
Decision rule
Table (database)
Decision system
Granular Computing
Cite
Citations (21)
Decision table
Decision rule
Decision system
Table (database)
Cite
Citations (17)
Rough set-based data analysis starts from a data table, called an information system. The attribute of information system is usually divided into two parts, condition attributes and decision. Such information system is called decision table. In every decision table a set of decision rules, called a decision algorithm, can be associated. It is shown that every decision algorithm reveals some well-known probabilistic properties, and we compare it with Bayes’theorem.
Decision table
Decision rule
Decision system
Table (database)
Cite
Citations (0)
In this paper, we assume that a dispersed data is represented by a finite set S of decision tables with equal sets of attributes. We discuss one of the possible ways to the study decision trees common to all tables from the set S: building a decision table for which the set of decision trees coincides with the set of decision trees common to all tables from S. We show when we can build such a decision table and how to build it in a polynomial time. If we have such a table, we can apply to it various decision tree learning algorithms. We extend the considered approach to the study of decision rules and test (reducts) common to all tables from S.
Decision table
Decision rule
Table (database)
Cite
Citations (1)
Recently a new type of decision table has been proposed, OR-decision table. The idea of the OR-decision table is one in which any of the actions in the set may be performed in order to satisfy the corresponding condition, while the classical decision table requires the execution of all the actions in the set of actions (AND-decision table). Due to the characteristic of the OR-decision table, this paper proposes the fast OR-decision table conversion to a decision tree algorithm.
Decision table
Table (database)
Decision rule
Decision problem
Decision model
Cite
Citations (6)
Decision system
Decision rule
Depth-first search
Tree (set theory)
Decision table
Cite
Citations (9)
To select a proper decision model in intelligent decision,a method to acquire the knowledge of model selection and select a decision model by utilizing the knowledge was proposed based on the rough set theory.With this method,the decision table of models with continuous attribute values is acquired by setting parameters in the models randomly.Then,the relation between objects in the decision table is obtained by setting errors.From the relation,the reduction of attributes for the decision table is conducted by applying the rough set theory to gain decision rules.Based on these decision rules,a proper model can be selected. The feasibility of the proposed method has been illustrated using an example.
Decision table
Decision rule
Decision model
Table (database)
Decision theory
Decision system
Cite
Citations (0)
Decision table
Admissible decision rule
Rule induction
Decision rule
Table (database)
Cite
Citations (65)
Analyzes the traditional methods of extracting decision rules in Rough Sets, defines the concept of the decision dependability and proposes a novel algorithm of extracting short decision rules. Only the length of decision rules is extended when the current decision rules can’t classify all the samples in the decision table. At the same time, three methods are proposed to reduce the computational complexity: 1) defines the concept of bound coefficient, 2) only classify the samples with the same decision values at a time thus averting the time-consuming classification of the equivalence classes with different decision values, 3) defines the Remain set and only classify the samples in the Remain set, so the computational complexity will decrease proportional with the reduction of the samples in the Remain set. Above-mentioned methods can be used directly for incomplete information systems and have great practicability.
Decision table
Decision rule
Decision system
Cite
Citations (0)
The rough set theory is an effective tool in dealing with uncertain knowledge and incomplete data. This paper proposes a simplified table decision approach base on rough set theory and tourism knowledge, using rough set theory handling large amounts of business information, extract useful rules and generate minimal decision rules through analysis and reasoning. Finally, By analyzing real examples, we proves the feasibility on the combination of rough set theory and the tourism decision support system. Our method effectively solves problems such as the acquisition and understanding of smart marketing decision rules in decision support system.
Decision table
Decision rule
Decision system
Table (database)
Cite
Citations (0)