Minkowski’s Inequality Based Sensitivity Analysis of Fuzzy Signatures

@article{Harmati2016MinkowskisIB,
  title={Minkowski’s Inequality Based Sensitivity Analysis of Fuzzy Signatures},
  author={Istv{\'a}n {\'A}. Harmati and {\'A}d{\'a}m Bukovics and L{\'a}szl{\'o} T. K{\'o}czy},
  journal={Journal of Artificial Intelligence and Soft Computing Research},
  year={2016},
  volume={6},
  pages={219 - 229}
}
Abstract Fuzzy signatures were introduced as special tools to describe and handle complex systems without their detailed mathematical models. The input parameters of these systems naturally have uncertainties, due to human activities or lack of precise data. These uncertainties influence the final conclusion or decision about the system. In this paper we discuss the sensitivity of the weigthed general mean aggregation operator to the uncertainty of the input values, then we analyse the… 

Figures and Tables from this paper

On the Sensitivity of Weighted General Mean Based Type-2 Fuzzy Signatures
TLDR
The input sensitivity of type-2 weighted general mean aggregation operator and fuzzy signatures which are equipped with general means as aggregation operators are discussed.
On the Sensitivity of the Weighted Relevance Aggregation Operator and Its Application to Fuzzy Signatures
TLDR
The sensitivity of the weighted relevance aggregation operator is discussed, namely it is given bounds on the change of the output in terms of vector norms of thechange of the input variables.
Improving the Quality of Transport Management Services with Fuzzy Signatures
TLDR
The goal is to create a decision support model based on fuzzy signatures that can assist the work of operational management automatically and if the model sets parameters properly, the management of transport could be more economical and efficient.
A Method for Genetic Selection of the Most Characteristic Descriptors of the Dynamic Signature
TLDR
A new method for genetic selection of the most characteristic descriptors of the dynamic signature, which increases precision of the biometric system by eliminating the descriptors which do not increase efficiency of verification procedure.
Stability Evaluation of the Dynamic Signature Partitions Over Time
TLDR
The proposed analysis is based on a set of coefficients defined in the context of the dynamic signature partitioning, which consists in division of the signature into parts which can be related to e.g. high and low velocity of pen in the initial and final phase of signing.
Distributed image retrieval with colour and keypoint features
TLDR
A system for retrieving images in relational databases in a distributed environment using the Apache Hadoop software framework with HDFS to solve content-based image retrieval problems.
Handwriting Recognition with Extraction of Letter Fragments
TLDR
Intelligent character recognition of handwritten texts relies on the processing of uncertainty in the form of fuzzy quality values starting from the identification of strokes, through the construction of words and phrases, up to future application of language filters and possible contextual recognition.
Distributed image retrieval with color and keypoint features
TLDR
This paper presents a system for retrieving images in relational databases in a distributed environment using the Apache Hadoop software framework with HDFS to distribute the process to several machines.

References

SHOWING 1-10 OF 49 REFERENCES
Sensitivity analysis of the weighted generalized mean aggregation operator and its application to fuzzy signatures
TLDR
Bounds are given on the changing of the weighted generalized mean in terms of vector norms of the changes of the variables and the sensitivity of fuzzy signatures which equipped with weighted generalizedmean operators in their nodes are characterized.
A sensitivity-based analysis of weighted fuzzy aggregation
  • U. Kaymak, H. van Nauta Lemke, T. den Boer
  • Computer Science
    1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36228)
  • 1998
TLDR
A general framework based on sensitivity analysis for extending fuzzy set aggregation operators to their weighted counterparts is introduced and it is shown that some of the methods suggested in literature for weighted aggregation with t-norms apply only in particular situations.
Fuzzy signatures in data mining
  • T. Vámos, L. Kóczy, G. Biro
  • Computer Science
    Proceedings Joint 9th IFSA World Congress and 20th NAFIPS International Conference (Cat. No. 01TH8569)
  • 2001
TLDR
A software system was developed to implement fuzzy signature ideas and works on a rather extensive database of a two-decade neuro-developmental project, and another project initiated in the subjects of transition economy and society is ongoing.
Construction of fuzzy signature from data: an example of SARS pre-clinical diagnosis system
TLDR
This work presents the concept of a fuzzy signature and how its flexibility can be used to quickly construct a medical pre-clinical diagnosis system using severe acute respiratory syndrome (SARS) as an example to show many advantages of the fuzzy signature.
Decision Support System for Evaluating Existing Apartment Buildings Based on Fuzzy Signatures
TLDR
This paper proposes a fuzzy signature model including relevance weights and weighted aggregations for each node and parent node, respectively, so that as a result a single membership value may be calculated for each building in question.
L-fuzzy sets
A Multiobjective Genetic Algorithm for Feature Selection in Data Mining
TLDR
This work presents an application of Multi-objective Genetic Algorithms to the Feature Selection problem, combining different criteria measuring the importance of the subsets of features.
Sensitivity analysis for WOWA, OWA and WM operators
  • V. Torra
  • Computer Science
    ISIE 2001. 2001 IEEE International Symposium on Industrial Electronics Proceedings (Cat. No.01TH8570)
  • 2001
TLDR
It is shown that the arithmetic mean is the one with less sensitivity and that the behavior of OWA, WM and WOWA for small errors is different than the case of large errors.
...
...