Alexander Nikov

A Fuzzy Bat Clustering Method for Ergonomic Screening of Office Workplaces

A method for screening of company workplaces with high ergonomic risk is developed. For clustering of company workplaces a fuzzy modification of bat algorithm is proposed. Using data gathered by a checklist from workplaces, information for ergonomic related health risks is extracted. Three clusters of workplaces with low, moderate and high ergonomic risk are determined. Using these clusters, workplaces with moderate and high ergonomic risk levels are screened and relevant solutions are proposed. By a case study this method is illustrated and validated. Important advantages of the method are reduction of computational effort and fast screening of workplaces with major ergonomic problems within a company.

UWIS: An assessment methodology for usability of web-based information systems

A methodology for usability assessment and design of web-based information systems (UWIS) is proposed. It combines web-based service quality and usability dimensions of information systems. Checklist items with the highest and the lowest contribution to the usability performance of a web-based information system can be specified by UWIS. A case study by a student information system at Fatih University is carried out to validate the methodology. UWIS reveals a strong relationship between quality and usability which is assumed to exist by many researchers but not experimentally analyzed yet. This study depicts a strong relevance between web-based service quality and usability of web-based information systems. UWIS methodology can be used for designing more usable and higher quality web-based information systems.

A fuzzy backpropagation algorithm

This paper presents an extension of the standard backpropagation algorithm (SBP). The proposed learning algorithm is based on the fuzzy integral of Sugeno and thus called fuzzy backpropagation (FBP) algorithm. Necessary and sufficient conditions for convergence of FBP algorithm for single-output networks in case of single- and multiple-training patterns are proved. A computer simulation illustrates and confirms the theoretical results. FBP algorithm shows considerably greater convergence rate compared to SBP algorithm. Other advantages of FBP algorithm are that it reaches forward to the target value without oscillations, requires no assumptions about probability distribution and independence of input data. The convergence conditions enable training by automation of weights tuning process (quasi-unsupervised learning) pointing out the interval where the target value belongs to. This supports acquisition of implicit knowledge and ensures wide application, e.g. for creation of adaptable user interfaces, assessment of products, intelligent data analysis, etc.

NN-AirPol: a neural-networks-based method for air pollution evaluation and control

A method for air pollution evaluation and control, based on one of the most popular neural networks - the backpropagation algorithm, is proposed. After the backpropagation training, the neural network, based on weather forecasting data, determines the future concentration of critical air pollution indicators. Depending on these concentrations, relevant episode warnings and actions are activated. A case study is carried out to illustrate and validate the method proposed, based on Istanbul air pollution data. Sulphur dioxide and inhalable particulate matter are selected as air pollution indicators (neural network outputs). Relevant episode measures are proposed. Among ten backpropagation algorithms, the BFGS algorithm (Quasi-Newton algorithms) is adopted since it showed the lowest training error. A comparison of NN-AirPol method against regression and perceptron models showed significantly better performance.

Quick fuzzy backpropagation algorithm

A modification of the fuzzy backpropagation (FBP) algorithm called QuickFBP algorithm is proposed, where the computation of the net function is significantly quicker. It is proved that the FBP algorithm is of exponential time complexity, while the QuickFBP algorithm is of polynomial time complexity. Convergence conditions of the QuickFBP, resp. the FBP algorithm are defined and proved for: (1) single output neural networks in case of training patterns with different targets; and (2) multiple output neural networks in case of training patterns with equivalued target vector. They support the automation of the weights training process (quasi-unsupervised learning) establishing the target value(s) depending on the networks input values. In these cases the simulation results confirm the convergence of both algorithms. An example with a large-sized neural network illustrates the significantly greater training speed of the QuickFBP rather than the FBP algorithm. The adaptation of an interactive web system to users on the basis of the QuickFBP algorithm is presented. Since the QuickFBP algorithm ensures quasi-unsupervised learning, this implies its broad applicability in areas of adaptive and adaptable interactive systems, data mining, etc. applications.

A Taguchi-based Kansei engineering study of mobile phones at product design stage

This article is aimed at applying Taguchi method in Kansei engineering and explores a way to integrate it into an industrial product design stage. Emotional customer needs are derived using Kansei image word pairs. The Taguchi-based approach is validated by a case study with mobile phones. Experimental work in implementing the proposed approach was able to suggest design attributes of a mobile phone that are essential to be considered at the product design stage to satisfy the customers’ expectations and hence to increase the companys sales.