New Approaches for Data Reduction in

Generalized Multi-valued Decision Information System:

Case study of Rheumatic Fever Patients 

 

A.M. Kozae, M.M.E. Abd El-Monsef, and S. Abd El-Badie

Department of Mathematics, Faculty of Science,

Tanta University, Egypt

 Abstract: A multi-valued information system (MIS) is a generalization of the idea of a single valued information system (SIS). In a multi-valued information system, attribute functions are allowed to map elements to sets of attribute values. In this paper, we initiated a new approach for data reduction in Generalized Multi–Valued Decision Information System (GMDIS). In the beginning we converted the Single-Valued Decision Information System (SDIS) by collecting the attributes to a GMDIS. Two general relations are defined on condition attributes and decision attribute. We constructed new classes using the general relations which are used for data reduction. The measure of decision dependency on the condition attributes is studied in our approach. To evaluate the performance of the approach, an application of, rheumatic fever datasets has been chosen and the reduct approach have been applied to see their ability and accuracy.

 Keywords:  Multi-Valued Information System, Rough Sets, Reduction

 [ Full paper in pdf format]  [Guideline to produce your PowerPoint presentation] [Power point Presentation]

A Multi Criteria Decision Analysis Using Fuzzy Logic

                              F.F. Farahat

     Dept. of Computer Science and Information Systems, Sadat Academy, Al-Maadi- Cairo (Egypt)

            E-mail: farahat123@yahoo.com

Abstract: The fuzzy sets (FSs) and rough sets (RSs) are very important research areas, especially in Hybrid and Integrated Intelligent Systems, and Multi-Criteria Decision Analysis Problem.  In the real economic life, the optimum moment to replace equipment with a new one plays an important role. One can find that in the classical financial mathematics almost all models have as a goal to find the optimum moment to replace the equipment under the condition of the minimum expenses. These classical mathematical models do not keep some quantitative and qualitative parameters together on the one hand, and ignore some uncertainty, on the other hand. In this paper, a trial is presented to eliminate these deficiencies by applying fuzzy models. Three fuzzy models will be proposed, here. The first one is developed to find the best moment of the equipment replacement, while the second fuzzy model is given to select new equipment. The third one is presented to determine multiple Internal Revenue Rate (IRR), based on J.T.C. Mao's algorithm.  In all cases, the quantitative and the qualitative criteria will be taken into consideration. These models are based on the following principles: (1) the rigorous methods to transform all criteria into some fuzzy sets on the same universe; (2) the use of the appropriate aggregation operators (AOs) of the type of generalized means; and (3) the decision – making in the multifactor framework. Fuzzy models are accessible ones, easy to simulate and do not get a computation complexity.  Some examples will be presented for persons who are working in this area.

 

Keywords: Rough Sets, Fuzzy Sets, Economical Models, Fuzzy Models, Aggregation Operators, Decision – Marking, Efficient Solutions,  IRR, Financial Analysis.

 [Full paper in pdf format]   [Guideline to produce your PowerPoint presentation] [PowerPoint Presentation]

A Novel Rough Net Approach for Rules Representation and Verification  

   Hala S. Own                                        

 National Research Institute of Astronomy and Geophysics, Helwan, Cairo, Egypt

Email: hala@sci.kuniv.edu.kw

Abstract: Rule based systems store knowledge as a set of rules and reason over them to solve problems. A new hype rough net approach for rule generation, representation and reasoning of knowledge-based system using rough sets and Petri nets is presented in this paper. A Rough Petri Net (RPN) is proposed for representing knowledge and formalism for the verification of rule-based systems.  The main stages of the proposed approach are: Rules will be first generated and normalized, then transform the normalized rules into a Rough Petri net, and finally we verify these normalized rules. A rough Petri net model (RPN) is presented to represent the rule of a rule-based system in which a rough production rule describes the rough relation between two propositions. An algorithm is presented for checking the consistency of a rough knowledge based via a set of reduction rules that preserve the properties of the RPN.

 Keyword:   Reasoning, Petri net, rough sets, Rule generation, Knowledge-based system

   [Full paper in pdf format]   [Guideline to produce your PowerPoint presentation]

Decision Analysis via Granulation Based on

General Binary Relation

    N. M. Kilany¹  and  M.M.E. Abd El-Monsef²

 Commercial Technical for Computer Sciences Institute in Suez, Egypt¹ 

Mathematics Department, Faculty of Science, Tanta University, Egypt²

 Abstract: Decision theory considers how best to make decisions in the light of uncertainty about data. There are several methodologies that may be used to determine the best decision. In rough set theory, the classification of objects according to approximation operators can be fitted into the Bayesian decision–theoretic, with respect to three regions (Positive, Negative and Boundary region). Granulation using equivalence classes is a restriction that limits the decision makers. In this paper, we introduced a generalization and modification of decision–theoretic rough set model by using granular computing on general binary relations. We obtain two new types of approximation that enable us to classify the objects into five regions instead three regions. The classification of decision region into five areas will enlarge the range of choice for decision make

Keywords: Decision Analysis, Granulation, Binary Relations.                           

  [Full paper in pdf format]   [Guideline to produce your PowerPoint presentation]

  Evolution Rough Sets

Osama Badawy¹, Yasser Foaud ², Moustafa Fahmy¹  Wedad Sagar ^2,

 ¹College of Computing and Information Technology, Arab Academy for Science &Technology & Maritime Transport, Alexandria, Egypt

 ²Department of Mathematics (Computer, Science), Alexandria University, Egypt

Abstract: Rough set theory, which emerged about 20 years ago, is nowadays a rapidly developing branch of artificial intelligence and soft computing. In this paper, we present a new approach (RS_DT) for construction of decision tree based on rough set theory, which will be induced a simplified tree. This tree is transformed into an initial population of genetic probabilistic rough induction (GA+GDT-RS), which is a combination between probabilistic rough induction and genetic algorithm. We will use probabilistic rough induction for modeling a classification system and applying genetic operators to a population of chromosomes. However, it is interesting to try to incorporate these approaches into the hybrid system. The challenge is to get as much as possible from this association.

                  Keywords: Generalization Distribution Table, Rough Sets, Genetic Algorithm, Machine learning, Data mining.

                  [Full paper in pdf format]   [Guideline to produce your PowerPoint presentation]

Reducing the Response Time for Data Warehouse Queries Using Rough Set Theory

Mahmoud Mohamed Al-Bouraie,  Yasser Fouad Hassan, Wesam Fathy Jasser

  Department of Math. (Computer Science), Faculty of Science

 Email: y_fouad@hotmail.com & wesamjasser@yahoo.com

 

Abstract: One of the most important problems in relational databases applications like data warehouses is a size of real-world databases. In practical problems, data may contain millions of records in many data tables bounded by relations. One approach to reduce the size of these applications is rough set theory. The Rough set theory is deeply investigated, and an approach for data filtering based on rough set theory is proposed. In addition, to improve processing of star queries on data and processing of aggregation star queries a new optimization technique is the called pre-grouping transformation. Although this transformation is expected to reduce the time needed to answer large aggregation queries to less than 50%, there are several cases where it is not beneficial. In this paper we retry reach to the optimization case by applying the two above theories.

 

        Keywords: Rough sets, data warehouse, Transformation

        [Full paper in pdf format]   [Guideline to produce your PowerPoint presentation] [Power point presentation]

Rough Neural Intelligent Approach for Classification

Sanaa Rashed Abdallah and Yasser Fouad Hassan

Department of Mathematics (computer science),Faculty of Science, Alex University

Abstract: This paper describes rough neural network where neural network systems and rough sets theory are completely integrated into a hybrid system and are used cooperatively for decision and classification support. Also, rough sets and neural network are chosen for the combined method because they can discover patterns in ambiguous and imperfect data,and provide tools for data and pattern analysis. The common characteristic of rough sets and neural networks is that both approaches have the ability to learn decision models by examples.

Keywords: Rough sets, neural network, hybrid systems

[Full paper in pdf format]   [Guideline to produce your PowerPoint presentation]

Rough Sets in Hybrid Intelligent System For Breast Cancer Detection

Aboul Ella Hassanien

Cairo university, faculty of computer and information, IT dept.

email: abo@cba.edu.kw  & itcairo@hotmail.com

   Abstract: Hybridization of intelligent systems is a promising research field of modern Artificial Intelligence concerned with the development of the next generation of intelligent systems. The objective of this paper  is to introduce a hybrid intelligent system that combines the advantages of different intelligent systems; fuzzy sets, rough sets and rough neural networks in conjunction with statistical feature extraction techniques. An application of, breast cancer imaging has been chosen and hybridization soft computing techniques have been applied to see their ability and accuracy to classify the breast cancer images into two outcomes: cancer or non-cancer. The system starts with fuzzy image processing as a pre-processing stage to enhance the contrast of the whole image; to extracts the region of interest and then to enhance the edges surrounding the region of interest. Next, subsequently extract features from the extracted regions characterizing the underlying texture of the interested regions using the gray-level co-occurrence matrix. A rough set approach to attribute reduction and rule generation is presented. Finally, rough neural networks are designed for discrimination for different regions of interest to test whether they are cancer or nun-cancer. A rough neuron can be viewed as a pair of neurons. One neuron corresponds to the upper bound and the other corresponds to the lower bound. Upper and lower neuron exchange information with each other during the calculation of their outputs. To evaluate the performance of the system, different images from the Mammographic Image Analysis Society (MIAS) database were selected. The experimental results show that the hybrid intelligent systems applied in this study perform well reaching over 98% in overall accuracy with 154 minimal number of generated rules.

Keywords: rough sets, neural network, fuzzy image processing

[Full paper in pdf format]  [power point presentation]