Steffen F. Bocklisch's research while affiliated with Technische Universität Chemnitz and other places

Publications (27)

Article
Full-text available
Modeling forms the basis for optimal control of complex technical processes in the context of industry 4.0 development and, hence, for high product quality as well as efficient production. For the mechanical joining process of self-pierce riveting with 11 input and 5 output variables, two modeling approaches based on (1) experimental data and (2) F...
Article
Full-text available
In this paper we introduce a new fuzzy system using adaptive fuzzy pattern classification (AFPC) for data-based online evolvement. The fuzzy pattern concept represents an efficient tool for handling uncertainty in multi-dimensional data streams and combines powerful performance, flexibility and meaningful interpretability within one consistent fram...
Article
Full-text available
Zusammenfassung Modelle dienen der Speicherung von Wissen und als Basis für Entscheidungen. Sie finden Anwendung in unterschiedlichsten Bereichen wie Technik, Medizin, Wirtschaft, Psychologie, Umwelt oder Verkehr. Gerade für komplexe Zusammenhänge sind Verfahren, die auf interpretierbaren Mustern beruhen, hoch flexibel und adaptiv. Die Theorie der...
Article
Since the foundation of the Department of Automatic Control at Chemnitz University of Technology in 1964, nonlinear systems have been a thematic priority in academic training and research. This line of research traces back to Alfred Pfeiffer (1900- 1985), who was a PhD student in Berlin at the research laboratory of the famous physicist and Nobel P...
Conference Paper
This paper examines the role of imprecision in the interpretation of verbal symptom intensities (e.g., high fever) depending on the level of medical expertise. In a contrastive study we compare low, medium and high level experts (medical students vs. physicians with M = 5.3 vs. M = 24.9 years of experience) concerning their interpretation of sympto...
Conference Paper
In this paper we used a two-step procedure for the numerical translation of verbal frequency expressions of the response scale of a questionnaire (STAI-T). In an empirical study, 70 participants estimated numerical equivalents for verbal frequency expressions, data was modeled, and fuzzy membership functions were calculated. Results show that the s...
Article
Full-text available
This article proposes an individual fuzzy modelling and treatment of data allowing for the specific uncertainty of each datum. A modelling approach based on parametric fuzzy sets is being introduced which can be employed to model both data with individual uncertainties as well as abstract phenomena in a feature space (classes). An aggregation proce...
Article
Full-text available
This article contributes to clustering and fuzzy modelling of data such that specific characteristics of each datum can be incorporated. Particularly, each object may exhibit an individual area of in-fluence in its feature space, for which it is repre-sentative. For such objects, a similarity measure is introduced, which is used to modify common cl...
Article
Full-text available
The article describes a general two-step procedure for the numerical translation of vague linguistic terms (LTs). The suggested procedure consists of empirical and model components, including (1) participants' estimates of numerical values corresponding to verbal terms and (2) modeling of the empirical data using fuzzy membership functions (MFs), r...
Poster
Full-text available
The context is known to influence words’ interpretation (Pepper & Prytulak, 1974; Teigen & Brun, 2003) such that the same verbal expression can have a different meaning in different contexts. This is an interesting problem concerning decision support systems (Boegl et al., 2004) and man-machine interactions where technical systems receive spoken la...
Conference Paper
The context is known to influence words’ interpretation (Pepper & Prytulak, 1974; Teigen & Brun, 2003) such that the same verbal expression can have a different meaning in different contexts. This is an interesting problem concerning decision support systems (Boegl et al., 2004) and man-machine interactions where technical systems receive spoken la...
Article
In some nonstationary time series, where a global model is neither available nor applicable, we may observe recurring patterns that can be extracted to create several local models instead. This article proposes knowledge-based short-time prediction methods for multivariate streaming time series that rely on the early recognition of such local patte...
Conference Paper
Full-text available
The paper describes a general two-step procedure for the numerical translation of linguistic terms using parametric fuzzy potential membership functions. In an empirical study 121 participants estimated numerical values that correspond to 13 verbal probability expressions. Among the estimates are the most typical numerical equivalent and the minima...
Conference Paper
The paper describes a general two-step procedure for the numerical translation of linguistic terms using parametric fuzzy potential membership functions. In an empirical study 121 participants estimated numerical values that correspond to 13 verbal probability expressions. Among the estimates are the most typical numerical equivalent and the minima...
Conference Paper
This article proposes knowledge-based short-time prediction methods for multivariate streaming time series, relying on the early recognition of local patterns. A parametric, well-interpretable model for such patterns is presented, along with an online, classification-based recognition procedure. Subsequently, two options are discussed to predict ti...
Conference Paper
The present work dedicates itself to the aggregation of nonconvex data-inherent structures into fuzzy classes. A key feature of this aggregation is its conduction within a closed fuzzy classification framework, being built around a single, generic type of a convex membership function. After a short elaboration concerning this essential building blo...
Conference Paper
This article deals with the recognition of recurring multivariate time series patterns modelled sample-point-wise by parametric fuzzy sets. An efficient classification-based approach for the online recognition of incompleted developing patterns in streaming time series is being presented. Furthermore, means are introduced to enable users of the rec...
Article
Full-text available
The present article dedicates itself to fuzzy modelling of data-inherent structures. In particular two main points are dealt with: the introduction of a fuzzy modelling framework and the elaboration of an automated, data-driven design strategy to model complex data-inherent structures within this framework. The innovation concerning the modelling f...
Conference Paper
This work is dedicated to a network oriented modelling approach based on the interconnection of multivariate Fuzzy Pattern Classifier nodes. The main point is the elaboration of a data driven and hierarchical design strategy for such fuzzy classifier network models. In detail two essential issues will be dealt with: the automatic layout of the netw...
Article
This article aims at extending fuzzy classification methods to the recognition of local patterns in time series. Firstly, a classifier model for patterns will be introduced, which allows for the prevalent uncertainty in measured data. It is able of classifying subsequences, and, due to its white box character, is easily comprehensible as well as mo...
Article
Kurzfassung In diesem Beitrag wird ein methodischer Ansatz für den Fuzzy Abgleich von Anforderungsvektoren mit Beschreibungsvektoren der Kompetenzzellen in hierarchielosen Produktionsnetzen vorgestellt. Als Entscheidungssystem dient dabei ein so genannter Fuzzy Pattern-Klassifikator, der als Knoten innerhalb eines Klassifikatornetzwerks eine Teilbe...
Article
Selection of competence cells in non-hierarchical production networks by fuzzy comparison of requirement vectors with description vectors. The paper introduces a methodical approach for the fuzzy comparison of requirement vectors with description vectors of competence cells in non-hierarchical production networks. A so-called fuzzy pattern classifi...
Article
Kurzfassung In diesem Beitrag wird eine Methode zur Implementierung unscharfer Informationen in Kompetenzzellenmodelle vorgestellt. Dabei wird die bisher existierende und auf scharfen Beschreibungsparametern basierende Modellierung von Kompetenzzellen in hierarchielosen Netzen so erweitert, dass die beschreibenden Merkmale durch unscharfe (Fuzzy) G...
Article
A method for implementing of non-sharp information's for competence cells is described. Existing modelling is extended from sharp descriptions to non-sharp (fuzzy) descriptions. In this way, the characteristics to describe the competence cells, are replaced and/or supplemented by non sharp data. This creates a prerequisite for modelling groups of s...
Article
Full-text available
In this article the issue of data based modeling is dealt with the help a network of uniform multivariate fuzzy classifiers. Within this framework the innovation consists in the specification of a hierarchical design strategy for such a network. Concretely, the two network specifying factors, namely the layout of the network structure and the class...

Citations

... A promising approach for the numerical translation of such vague linguistic terms is provided by Bocklisch (2012;cf. Bocklisch et al. 2010) based on a fuzzy approach, which has already been applied to the driving context before (Bocklisch, 2011). This approach will be used in future studies to derive more specific numerical recommendations for the design of comfortable ADS. ...
... It should also be noted that the classifier type is caused by the source data type. In particular, the crisp and fuzzy data result to use of fuzzy and crisp accordingly [4,9,12,13]. Binary Decision Diagram (BDD) can be useful for the binary attributes [48] or Multi-Valued Decision Diagram (MDD) is acceptable for the categorical data [41]. ...
... Unfortunately, knowledge-based fuzzy rule models face serious difficulties modelling complex issues and a complete and sufficiently consistent model specification is demanding; the number of rules increases almost exponentially in multidimensional input spaces and the system's "tuning" is difficult, as well [45,46]. In FPC models, the multidimensional connection is principally pattern-not rulebased, but, special patterns can be generated from fuzzy rules [47] (e.g., see context classifier in Section 3.). ...
... The respondents' estimates were then modelled using fuzzy MFs of different types (Padma and Balasubramanie developed -MFs [25], Bocklisch and colleagues potential type-MFs [32]). Results show that MFs are typically non-equidistantly distributed along a numerical scale and not necessarily symmetric in shape (see [32,34]). Furthermore, MFs for verbal symptom intensities (e.g., immense pain or high fever) of medical students are very vague in contrast to seasoned experts (i.e., physicians with >16 years practical experience), whose MFs are characterized by almost crisp categories [33]. ...
... Therefore the aggregation of repeated measurements (which may, of course, be afflicted with uncertainties as described above) provides one possible access to obtain an object's area of influence or its fuzzy representation, respectively. For this case, [6] presents an aggregation procedure that directly results in membership functions of the type used in this article. In the spirit of fuzzy set theory, another approach to the determination of an object's area lies in the inclusion of domain-specific expert knowledge for each object. ...
... There are many papers of this kind (e.g. Angelov and Yager 2013;Angelov and Kasabov 2005;Angelov et al. 2004;Boulmakoul et al. 2017;Bronevich and Rozenberg 2014;Angelov 2014, 2012;Cintra et al. 2010;Cross 2018;Dey et al. 2011;Henzgen et al. 2014;Herbst and Bocklisch 2010;Huang and Wang 1995;Hulianytskyi and Riasna 2016;Jung et al. 2011;Lan et al. 2016;Leng et al. 2012;Lukka 2011;Mansoori and Shafiee 2016;Palanisamy and Selvan 2009;Prasad et al. 2011;Scozzafava and Vantaggi 2009;Sussner and Valle 2008;Zhang and Yang 2016;Zhou et al. 2015) and it would be very pleasant and useful if part of our work could be connected with some of these studies. Something like this could give us further significant information and help us improve our research. ...
... While these two studies do not suggest that verbal quantifiers are effective ways to communicate social norms, other studies advocate that their vague meaning and subjective interpretation could make them more or less motivating than exact numbers [20,21]. In Bocklisch and colleagues' study, participants believed that a 'possible' event had an average likelihood rating of 51.4 out of 100 with a standard deviation of 21.6 [20]. ...
... A variety of methods based on the fuzzy membership function (MF) concept can be flexibly employed depending on the research question. Fuzzy logic rules, fuzzy cognitive maps, neuro-fuzzy systems, fuzzy reinforcement learning [64] as well as (adaptive) fuzzy pattern classification (FPC) [17,65] are only a few such examples. For the issues addressed here, we chose the latter method -FPCbecause the pattern-based classifier approach (1) allows modeling complex interdependencies of human and technical variables (e.g., fixation duration or voltage characteristics) in a n-dimensional feature space. ...
... Among the many developments involving the fuzzy conversion approach, one can mention the so-called fuzzy SERVQUAL (see, for instance, Aydin and Pakdil [1], Chou et al. [6], and Hu et al. [14]), or the fuzzification of verbal/linguistic inputs/expressions (see, for instance, Bocklisch et al. [4,5], Herrera [9], Lalla et al. [15], and Turksen and Willson [19]). I all these papers one can find several proposals to convert the most usual labels or verbal ratings into fuzzy numbers. ...
... In 1999, Singh introduced a Pattern Modelling and Recognition System (PMRS) in the field of time series forecasting in economics (Singh 1999a, b;Fieldsend 2000, 2001;Singh 2001), by assuming that what happened in the past will be reproduced in the future, thus finding the closest match to historic data. Since then, the PMRS method has been gradually used in time series prediction (Fan et al. 2008;Herbst and Bocklisch 2010;Cheng 2014), to more accurate effect. ...