A novel approach to ranking fuzzy numbers based on Fuzzy Acceptability Analysis (FAA) is considered. The use of FAA represents a consistent approach to fuzzy decision analysis in accordance with the concept that "...
详细信息
ISBN:
(纸本)9789813146969
A novel approach to ranking fuzzy numbers based on Fuzzy Acceptability Analysis (FAA) is considered. The use of FAA represents a consistent approach to fuzzy decision analysis in accordance with the concept that "decision taken in a fuzzy environment should be inherently fuzzy". The two preference relations for ranking fuzzy numbers with the use of FAA are presented. Implementation of FAA within fuzzy Multicriteria Decision Analysis is considered based on ranking fuzzy numbers/alternatives with indicating a fuzzy measure (degree of confidence) for each rank.
In view of the semi-supervised classification problem for imbalanced data, a new semi-supervised learning algorithm based on SVM is proposed. In this method, the classification method based on SVM for unbalanced data ...
详细信息
ISBN:
(纸本)9789813146969
In view of the semi-supervised classification problem for imbalanced data, a new semi-supervised learning algorithm based on SVM is proposed. In this method, the classification method based on SVM for unbalanced data is used to tag unlabeled documents in order to deal with the imbalance of data. The experimental results on several benchmark data sets show the validity of this method.
Among different road user types, drivers represent the largest share of road fatalities. As a result, more attention should be paid to the behavior of drivers, especially their behavior over time. By using driving sim...
详细信息
ISBN:
(纸本)9789813146969
Among different road user types, drivers represent the largest share of road fatalities. As a result, more attention should be paid to the behavior of drivers, especially their behavior over time. By using driving simulator data, this study aims to investigate the relative performance of individual drivers over time. To this end, 20 participants (14 in the end) completed a particular simulator scenario over five days, and their driving performance at various points along the driving scenario was recorded. By taking all this information into account, the technique of data envelopment analysis was applied to assess the relative performance of each driver, and the window analysis was used to measure the variations in performance over time.
Most of the web content today is generated on the fly using dynamic server side scripts. This web is known as hidden web or deep web. Extracting data from deep web is a non-trivial task as the layout and structure of ...
详细信息
ISBN:
(纸本)9789813146969
Most of the web content today is generated on the fly using dynamic server side scripts. This web is known as hidden web or deep web. Extracting data from deep web is a non-trivial task as the layout and structure of deep web is highly irregular. Deep web data extraction is important as it is useful for meta-search engine applications and comparative shopping lists. Before such data can be used for further processing, it must first be aligned so that the processing task could be made easier. This process is called data alignment. In the early days, data are aligned based on the conventional DOM Tree structure, and its underlying visual cue and more recently, ontologies have been used to align deep web data. However, this approach makes little use of the full semantics provided by WordNet libraries. In this paper, we propose a full-fledged multilingual WordNet to align data records. Unlike existing approaches, we make full use of the semantic properties provided by WordNet libraries, with multi-language support. Experimental results show that our approach is highly efficient in data alignment.
We propose a new theoretical background of autoregressive fuzzy associative memories (AFAM). It stems from the theory of fuzzy relation equations and eigen sets of their solutions. We introduce a couple of related AFA...
详细信息
ISBN:
(纸本)9789813146969
We propose a new theoretical background of autoregressive fuzzy associative memories (AFAM). It stems from the theory of fuzzy relation equations and eigen sets of their solutions. We introduce a couple of related AFAM models that share the same fuzzy relation. For one particular couple, we propose a fast algorithm of data retrieval. We characterize the types of noise that can be removed/reduced by the related models of a couple.
The problem of the accumulation of experience and the use of decision-making in the previously observed situations is researched. The main objective of the research is development of the data model that provides the u...
详细信息
ISBN:
(纸本)9789813146969
The problem of the accumulation of experience and the use of decision-making in the previously observed situations is researched. The main objective of the research is development of the data model that provides the upgrade of reliability of decision-making on the basis of experience. The concept of the image of the situation, which has not clearly defined center and a neighborhood, is introduced. The main thing is not trajectories in feature space but the admissible transformations of situations and solutions.
Traditional pattern making models are mainly linear models. This kind of model has many shortcomings. As Back Propagation (BP) neural network using simple nonlinear transfer functions can approximate any nonlinear fun...
详细信息
ISBN:
(纸本)9789813146969
Traditional pattern making models are mainly linear models. This kind of model has many shortcomings. As Back Propagation (BP) neural network using simple nonlinear transfer functions can approximate any nonlinear functions with any precision, we proposed a BP neural network model to predict all pattern making- related body dimensions by inputting few key human body dimensions. Sixty students in the northeast of China were measured for collecting a learning data to train the proposed model, and eleven of the sixty subjects' body dimensions data were applied to test the accuracy of the model. The results show that the prediction accuracies of linear regression model and BP neural network model have little difference. As the traditional linear model can be well applied in pattern making, the BP neural network model also can be well used for pattern making. Moreover, if increasing the number of learning samples, the precision of the proposed model is further improved.
In this paper, a rule based representation with interval certitude structure is presented as well as its inference method based on evidential reasoning, operational research and fuzzy set theory. A rule base is design...
详细信息
ISBN:
(纸本)9789813146969
In this paper, a rule based representation with interval certitude structure is presented as well as its inference method based on evidential reasoning, operational research and fuzzy set theory. A rule base is designed with interval certitude degrees embedded in the antecedent terms and consequent terms, which is shown to be capable of capturing uncertainties of human knowledge and human judgment. And the evidential reasoning approach is applied to the rule combination.
作者:
Madrid, N.Hodakova, P.Rusnok, P.Univ Malaga
Dept Appl Math Doctor Pedro Ortiz Ramos S-N E-29071 Malaga Spain Univ Ostrava
Ctr Excellence IT4Innovat Inst Res & Applicat Fuzzy Modeling 30 Dubna 22 Ostrava 70103 Czech Republic
In general, the F-transform is a powerful technique to approximate functions or compress data. In this article we explore the residuated versions of F-transform and extend their results for data that can not be repres...
详细信息
ISBN:
(纸本)9789813146969
In general, the F-transform is a powerful technique to approximate functions or compress data. In this article we explore the residuated versions of F-transform and extend their results for data that can not be represented by a function. We present some properties and demonstrate their applicability on an illustrative example.
暂无评论