Journal of Intelligent & Fuzzy Systems - Volume 21, issue 5

Authors: Baykasoglu, Adil | Dereli, Turkay | Burhan Turksen, I.

Article Type: Other

DOI: 10.3233/IFS-2010-0460

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 275-276, 2010

Authors: Eminli, Mübariz | Guler, Nevin

Article Type: Research Article

Abstract: In this study, we propose fuzzy modeling algorithm to improve Takagi-Sugeno fuzzy model. This algorithm initially finds desirable number of rules at once, in advance, and then identifies the premise and consequent parameters separately by fixing number determined. The proposed algorithm consists of three stages: determination of the optimal number of fuzzy rules, coarse tuning of parameters and fine tuning of these parameters. To find the optimal number of rules, the new cluster validity algorithm that is based on the validity criterion Vsv adapted to the usage of FCRM-like clustering, is proposed. In coarse tuning, by using the mentioned …clustering algorithm for input-output data and the projection scheme, the consequent and premise parameters are coarsely defined. In fine tuning, the gradient descent (GD) method is used to precisely adjust parameters of fuzzy model but unlike other similar modeling algorithms, the premise parameters are adjusted with respect to multidimensional membership function in premise part of rule. Finally, two examples are given to demonstrate the validity of suggested modeling algorithm and show its excellent predictive performance. Show more

Keywords: Takagi-Sugeno fuzzy model, multidimensional fuzzy sets, course tuning, fine tuning, fuzzy C-regression model, gradient descent method

DOI: 10.3233/IFS-2010-0461

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 277-287, 2010

Authors: Egemen Yilmaz, A. | Telatar, Ziya

Article Type: Research Article

Abstract: Recently, there has been a great tendency to computer aided music generation especially due to the rapid developments in technology. In this area, many researchers handled a variety of methods in order to perform this fact by taking into account some kinds of rules defined by music theory. Unfortunately, handling all these complicated rules by means of advanced methods would not be sufficient. For that reason, most of the methods and applications presented in literature were criticized to be “lacking of sprit” like contributing a human sense (for harmonization) even by the developer of that particular method. In this study, …we consider the particular problem of four-part choral harmonization to overcome this issue, and try to discuss at which points fuzzy logic is applicable throughout the solution. Departing from the fact that the definition of consonance can be related to a fuzzy set, we try to handle the harmonization problem by defining some mathematical operators on the fuzzy sets encountered. Show more

Keywords: Music theory, choral harmonization, fuzzy logic, fuzzy sets

DOI: 10.3233/IFS-2010-0462

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 289-301, 2010

Authors: Özdemir, R.G. | Çiçek, B. | Özenç, D.

Article Type: Research Article

Abstract: This paper deals with the problem of multi objective new product development and its solution with an intelligent approach. The proposed approach uses fuzzy parameters for economic analysis and combines a multi-criteria decision making method to goal programming. In economic analysis, fuzzy life cycle costs and fuzzy life cycle pricing are considered with respect to demand changes during life cycle of new product alternatives. Fuzzy equivalent worth of each new product alternative is determined using fuzzy life cycle monetary input, and results are input to the multi-criteria analysis. The selected multi-criteria analysis tool is TOPSIS in the paper and some …additional judgmental criteria are also considered to rank new product alternatives. The preference weights used to rank alternatives are then fed to a goal programming model which makes an ultimate selection of new product(s) to be produced in the manufacturing system under capacity, sales potential, and workforce balance constraints. The goal programming model has an objective of minimizing the weighted sum of positive deviation from target total cost and negative deviation from target total preference weights. The proposed approach is then implemented using real data of a continuous production system for the need of introducing new products into market. Show more

Keywords: Fuzzy life-cycle-cost parameters, multi objective decision making, goal programming, new product development

DOI: 10.3233/IFS-2010-0463

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 303-316, 2010

Authors: Özbakır, Lale | Tapkan, Pınar

Article Type: Research Article

Abstract: Many real life problems contain imprecise variables, constraints and objectives. Fuzzy set theory gives an opportunity to handle imprecise terms in such situations. Two-sided assembly line balancing (2sALB) problem which is a generalization of the well known simple assembly line balancing problem can also be modeled more realistically by employing fuzzy approaches. Such an approach is presented in this study to model and solve 2sALB problem by employing fuzzy mathematical programming and Bees Algorithm (BA). 2sALB problem is a combinatorial complex problem. For this reason BA is employed as a search mechanism for obtaining good solutions to it. BA is …a relatively new member of swarm intelligence based meta-heuristics that tries to mimic natural behavior of real honey bees in food foraging in solving complex optimization problems. BA is generally applied to continuous optimization in the literature. Its application to combinatorial problems is rare. This study also presents one of the first application of BA to an assembly line balancing problem which is member of combinatorial optimization. Show more

Keywords: Fuzzy multi-objective programming, two-sided assembly line balancing problem, Bees Algorithm, soft computing

DOI: 10.3233/IFS-2010-0464

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 317-329, 2010

Authors: Niskanen, Vesa A.

Article Type: Research Article

Abstract: Lotfi Zadeh’s recent ideas on his extended logic, the FLe approach, is applied to hypothesis verification from the standpoint of the philosophy of science. First, such typical conventional hypothesis verification methods as disjunctive and modus tollens syllogisms are presented. Second, Zadeh’s extended logic, which stems from approximate reasoning, is sketched. Third, application of the FLe to hypothesis verification is considered at metalevel. In particular, the hypothetico-deductive model and statistical reasoning are examined.

Keywords: Hypothesis verification, approximate reasoning

DOI: 10.3233/IFS-2010-0465

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 331-339, 2010

Authors: Şekkeli, Gizem | Köksal, Gülser | Batmaz, İnci | Türker Bayrak, Özlem

Article Type: Research Article

Abstract: Fuzzy linear regression (FLR) approaches are widely used for modeling relations between variables that involve human judgments, qualitative and imprecise data. Tanaka’s FLR analysis is the first one developed and widely used for this purpose. However, this method is not appropriate for classification problems, because it can only handle continuous type dependent variables rather than categorical. In this study, we propose three alternative approaches for building classification models, for a customer satisfaction survey data, based on Tanaka’s FLR approach. In these models, we aim to reflect both random and fuzzy types of uncertainties in the data in different ways, and …compare their performances using several classification performance measures. Thus, this study contributes to the field of fuzzy classification by developing Tanaka based classification models. Show more

Keywords: Fuzziness, fuzzy classification, fuzzy linear regression (FLR), customer satisfaction

DOI: 10.3233/IFS-2010-0466

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 341-351, 2010

Authors: Dereli, Türkay | Durmuşoğlu, Alptekin

Article Type: Research Article

Abstract: Identification and assessment of technological advances have been vital for companies to keep their competitive position or to gain new capabilities for the competition. In this content, Technology Watch Systems (TWS) have been tools of systematic analysis of technology developments that outputs regarding the technological opportunities and threats could be easily interpreted by an analyst. Among several TWS’s, Patent Alert System (PAS) has been a recently developed one which enables users to set or configure alert(s) for the trend changes in a certain technology area of requested sector. Data of associated patent counts is retrieved by extended markup language (XML) …mechanism located in PAS, and then an internal alert triggering mechanism is used to search for trend changes on the associated data. This internal alert triggering mechanism is a kind of modified linear regression which sets a newer trend line once a certain amount of deviation (threshold) has risen. Although alerts, indicating the direction of technological changes, provide supportive information to the analysts, extracted trend lines have been narrowed by a strict line where possible deviations have not been reflected. However, deviations in techno-systems are known to occur as the consequence of the vagueness coming from the nature of the system. Therefore, in this work, alert triggering mechanism of PAS is reconsidered using “possibilistic linear fuzzy regression”. Results yielded better and promising outcomes for the reconsidered algorithm. Show more

Keywords: Technology watch, trend extraction, fuzzy linear regression

DOI: 10.3233/IFS-2010-0467

Citation: Journal of Intelligent & Fuzzy Systems, vol. 21, no. 5, pp. 353-363, 2010