FCTA 2016 Abstracts


Full Papers
Paper Nr: 3
Title:

Classification Confusion within Nefclass Caused by Feature Value Skewness in Multi-dimensional Datasets

Authors:

Jamileh Yousefi and Andrew Hamilton-Wright

Abstract: This paper presents a model for treatment of skewness effect on accuracy of the Nefclass classifier by changing embedded discretization method within the classifier. Nefclass is a common example of the popular construction of a Neurofuzzy system. The popular Nefclass classifier exhibits surprising behaviour when the feature values of the training and testing data sets exhibit significant skew. As skewed feature values are commonly observed in biological data sets, this is a topic that is of interest in terms of the applicability of such a classifier to these types of problems. From this study it is clear that the effect of skewness on classification accuracy is significant and this must be considered in work dealing with skewed data distributions. We compared accuracy of Nefclass classifier with two modified versions of Nefclass embedded with MME and CAIM discretization methods. From this study it is found the CAIM and MME discretization methods results in greater improvements in the classification accuracy of Nefclass classifier as compared to using the original EqualWidth technique.

Paper Nr: 5
Title:

Hyperresolution for Propositional Product Logic

Authors:

Dušan Guller

Abstract: We provide the foundations of automated deduction in the propositional product logic. Particularly, we generalise the hyperresolution principle for the propositional product logic. We propose translation of a formula to an equivalent satisfiable finite order clausal theory, which consists of order clauses - finite sets of order literals of the augmented form: e1 @ e2 where e1 is either a truth constant, 0, 1, or a conjunction of powers of propositional atoms, and @ is a connective from =, <. = and < are interpreted by the equality and strict linear order on [0,1], respectively. We devise a hyperresolution calculus over order clausal theories, which is refutation sound and complete for the finite case. By means of the translation and calculus, we solve the deduction problem T |= phi for a finite theory T and a formula phi.

Paper Nr: 11
Title:

Bijective Fuzzy Relations - A Graded Approach

Authors:

Martina Daňková

Abstract: The bijectivity is one of the crucial mathematical notions. In this paper, we will present a fuzzy bijective mapping as a fuzzy relation that has several special properties. These properties come with degrees and so the bijectivity is also graded property. We will focuse on properties of this type of relations and we show graded versions of theorems on fuzzy bijections that are known from the classical Fuzzy Set Theory.

Paper Nr: 15
Title:

Robust Fuzzy Modeling and Symbolic Regression for Establishing Accurate and Interpretable Prediction Models in Supervising Tribological Systems

Authors:

Edwin Lughofer, Gabriel Kronberger, Michael Kommenda, Susanne Saminger-Platz, Andreas Promberger, Falk Nickel, Stephan Winkler and Michael Affenzeller

Abstract: In this contribution, we discuss data-based methods for building regression models for predicting important characteristics of tribological systems (such as the friction coefficient), with the overall goal of improving and partially automatizing the design and dimensioning of tribological systems. In particular, we focus on two methods for synthesis of interpretable and potentially non-linear regression models: (i) robust fuzzy modeling and (ii) enhanced symbolic regression using genetic programming, both embedding new methodological extensions. The robust fuzzy modeling technique employs generalized Takagi-Sugeno fuzzy systems. Its learning engine is based on the Gen-Smart-EFS approach, which in this paper is (i) adopted to the batch learning case and (ii) equipped with a new enhanced regularized learning scheme for the rule consequent parameters. Our enhanced symbolic regression method addresses (i) direct gradient-based optimization of numeric constants (in a kind of memetic approach) and (ii) multi-objectivity by adding complexity as a second optimization criterion to avoid over-fitting and to increase transparency of the resulting models. The comparison of the new extensions with state-of-the-art non-linear modeling techniques based on nine different learning problems (including targets wear, friction coefficients, temperatures and NVH) shows indeed similar errors on separate validation data, but while (i) achieving much less complex models and (ii) allowing some insights into model structures and components, such that they could be confirmed as very reliable by the experts working with the concrete tribological system.

Short Papers
Paper Nr: 2
Title:

Fuzzy Modeling and Control for Intention Recognition in Human-robot Systems

Authors:

Rainer Palm, Ravi Chadalavada and Achim J. Lilienthal

Abstract: The recognition of human intentions from trajectories in the framework of human-robot interaction is a challenging field of research. In this paper some control problems of the human-robot interaction and their intentions to compete or cooperate in shared work spaces are addressed and the time schedule of the information flow is discussed. The expected human movements relative to the robot are summarized in a so-called ”compass dial” from which fuzzy control rules for the robot’s reactions are derived. To avoid collisions between robot and human very early the computation of collision times at predicted human-robot intersections is discussed and a switching controller for collision avoidance is proposed. In the context of the recognition of human intentions to move to certain goals, pedestrian tracks are modeled by fuzzy clustering, lanes preferred by human agents are identified, and the identification of degrees of membership of a pedestrian track to specific lanes are discussed. Computations based on simulated and experimental data show the applicability of the methods presented.

Paper Nr: 4
Title:

Individual and Global Assessments with Signed Distance Defuzzification, and Characteristics of the Output Distributions based on an Empirical Analysis

Authors:

Rédina Berkachy and Laurent Donzé

Abstract: Considering a particular kind of questionnaires called linguistic questionnaires, we apply fuzzy logic to provide individual and global weighted evaluations in the case of the signed distance defuzzification method. We test our method on real data coming from a survey of the financial place of Zurich (Switzerland). Furthermore, we have been enable to give a look at the output distributions, and put into evidence their statistical properties. In particular, normality of distributions draws our attention. One of our main findings is that the individual evaluations calculated with the signed distance defuzzification method tend to be normally distributed.

Paper Nr: 6
Title:

A New Distance on a Specific Subset of Fuzzy Sets

Authors:

Majid Amirfakhrian

Abstract: In this paper, first we propose a definition for fuzzy LR sets and then we present a method to assigning distance between these form of fuzzy sets. We show that this distance is a metric on the set of all trapezoidal fuzzy sets with the same height and all trapezoidal fuzzy numbers and is a pseudo-metric on the set of all fuzzy sets.

Paper Nr: 16
Title:

The Effect of Noise and Outliers on Fuzzy Clustering of High Dimensional Data

Authors:

Ludmila Himmelspach and Stefan Conrad

Abstract: Clustering high dimensional data is still a challenging problem for fuzzy clustering algorithms because distances between each pair of data items get similar with the increasing number of dimensions. The presence of noise and outliers in data is an additional problem for clustering algorithms because they might affect the computation of cluster centers. In this work, we analyze the effect of different kinds of noise and outliers on fuzzy clustering algorithms that can handle high dimensional data: FCM with attribute weighting, the multivariate fuzzy c-means (MFCM), and the possibilistic multivariate fuzzy c-means (PMFCM). Additionally, we propose a new version of PMFCM to enhance its ability handling noise and outliers in high dimensional data. The experimental results on different high dimensional data sets show that the possibilistic versions of MFCM produce accurate cluster centers independently of the kind of noise and outliers.

Paper Nr: 17
Title:

On Bipartite Fuzzy Stochastic Differential Equations

Authors:

Marek T. Malinowski

Abstract: The paper contains a discussion on solutions to new type of fuzzy stochastic differential equations. The equations under study possess drift and diffusion terms at both sides of equations. We claim that such the equations have unique solutions in the case that equations’ coefficients satisfy a certain generalized Lipschitz condition. We use approximation sequences to reach solutions.

Paper Nr: 19
Title:

Labeled Fuzzy Rough Sets Versus Fuzzy Flow Graphs

Authors:

Leszek Rolka and Alicja Mieszkowicz-Rolka

Abstract: This paper presents the idea of labeled fuzzy rough sets which constitutes a novel approach to rough approximation of fuzzy information systems. The labeled fuzzy rough sets approach is compared with the fuzzy flow graph approach. The standard definition of fuzzy rough sets is based on comparing the elements of a universe by using a fuzzy similarity relation. This is a complex task, especially in the case of large universes. The idea of labeled fuzzy rough sets consists in comparison of elements of the universe to some ideals represented by linguistic values of attributes. Every element of the universe can be bound up with a linguistic label. Fuzzy rough approximations of any fuzzy set are obtained by describing its elements with the help of characteristic elements of linguistic labels. In this paper, new parameterized notions of the positive, boundary, and negative linguistic values are introduced.

Posters
Paper Nr: 7
Title:

Instruction Structure Analysis Appling Fuzzy Number

Authors:

Seiji Saito and Takenobu Takizawa

Abstract: Applying fuzzy clustering method to the instruction structure analysis, we can investigate whether the order of teaching item is suitable or not. However, when the teacher gives learners partial points, it is difficult to judge whether the leaner solve the problem correctly or not. In this paper, the authors regard the score of the test as the fuzzy number, and present a new analysis method using fuzzy number. We show some graphs required for analysis based on the results of examination for high school students and represent the effectivity of the method.

Paper Nr: 8
Title:

Contingency Table Analysis Applying Fuzzy Number and Its Application - Needs Analysis for Media Lectures

Authors:

Hiroaki Uesu

Abstract: Generally, we could efficiently analyse the inexact information by applying fuzzy theory. We would extend contingency table, and propose type-2 fuzzy contingency table. In this paper, we would discuss about type-2 fuzzy contingency table and a needs analysis method applying type-2 fuzzy contingency table.