Graph clustering is a very popular research field with numerous practical applications. Here we focus on finding fuzzy clusters of nodes in unweighted, undirected, and irreflexive graphs. We introduce three new algorithms for fuzzy graph clustering (Newman–Girvan NERFCM, Small World NERFCM, Signal NERFCM)....

To face the big data challenge, sampling can be used as a preprocessing step for clustering. In this paper, an hybrid algorithm is proposed. It is density-based while managing distance concepts. The algorithm behavior is investigated using synthetic and realworld data sets. The first experiments proved...

Uncertainty treatment in self-optimising systems touches two design-issues. Firstly, a valid estimation of uncertainties within the system is impossible beforehand as the uncertainties as well as the systems behaviour changes during run-time due to self-optimisation. Secondly, the design of a selfoptimising...

This paper presents our clustering technique based on generalized SOMs with evolving splittingmerging tree-like structures and its application to complex clustering problems including some benchmark data sets and, first of all, WWW-document clustering. Our approach that works in a fully unsupervised...

Association rules is a useful tool to extract new information from raw data expressed in a comprehensive way for decision makers. However, in some applications raw data might not be available for several reasons. First, stream data are only temporarily available for their processing or if it is stored,...

Likelihood functions, as well as the more recent concept of evidential likelihood, are essential statistical tools to perform estimation. Beyond the maximal likelihood value, the shape of the likelihood can also give interesting information about the data used to get the estimation. Indeed, it is generally...

The K-Nearest Neighbors (k-NN) classifier has become a well-known, successful method for pattern classification tasks. In recent years, many enhancements to the original algorithm have been proposed. Fuzzy sets theory has been the basis of several proposed models towards the enhancement of the nearest...

Medoid clustering frequently gives better results than those of the K-means clustering in the sense that a unique object is the representative element of a cluster. Moreover the method of medoids can be applied to nonmetric cases such as weighted graphs that arise in analyzing SNS(Social Networking Service)...

Evolutionary Fuzzy Systems (EFSs) are fuzzy systems augmented by a learning process based on evolutionary computation such as evolutionary algorithms (EAs). These systems contribute with several advantages in the development of algorithms, and specifically in the development of subgroup discovery (SD)...

We present a technique for size reduction of a base of fuzzy association rules which is created using an automated approach and which is intended for inference. Our approach is based on controlling the coverage of training data by the rule base and removing only such rules that do not increase that coverage....

Soft computing techniques may provide various forms of human-consistent summaries about large time series databases, e.g., linguistic summaries, frequent patterns, fuzzy IF-THEN rules. Within this research, we focus on linguistic summaries constructed as linguistically quantified propositions, that may...

Rule base generation from numerical data has been a dynamic research topic within the fuzzy community in the last decades, and several well-established methods have been proposed. While some authors presented simple, empirical approaches, but which generally show high error rates, others turned to complex...

The K-means algorithm is one of the most often used clustering techniques. However, when it comes to discovering clusters in informetric data sets that consist of non-increasingly ordered vectors of not necessarily conforming lengths, such a method cannot be applied directly. Hence, in this paper, we...

In a previous work we proposed to enhance the performance of FARC-HD fuzzy classifier in multi-class classification problems using decomposition strategies. This synergy was further improved by introducing n-dimensional overlap functions in the learning algorithm and the inference of FARC-HD instead...

This work aims to develop a practical study of the uni-norms as rule antecedent aggregation operator family in Linguistic Fuzzy Modeling. Although they are well known from a theoretical point of view, they have only recently been introduced in a few specific and recent applications. Uninorms are parameterized...

We consider the task of finding frequent parallel episodes in parallel point processes, allowing for imprecise synchrony of the events constituting occurrences (temporal imprecision) as well as incomplete occurrences (selective participation). We tackle this problem with frequent pattern mining based...

This work presents a novel Genetic Fuzzy System (GFS), called Genetic Programming Fuzzy Inference System for Regression problems (GPFISRegress). It makes use of Multi-Gene Genetic Programming to build the premises of fuzzy rules, including t-norms, negation and linguistic hedge operators. GPFIS-Regress...

In this paper, the theoretical foundations of generalised fuzzy Bayesian Network based on Vectorial Centroid defuzzification is introduced. The extension of Bayesi-an Network takes a broad view by examples labelled by a fuzzy set of attributes, instead of a classical set. Com-bining fuzzy set theory...

The study of energy efficiency in buildings is an active field of research. Modelling and predicting energy related magnitudes leads to analyse electric power consumption and can achieve economical benefits. In this study, machine learning techniques are applied to predict active power in buildings....

The median of a fuzzy random variable has been extended either by applying Zadeh’s extension principle or by minimizing its mean distance w.r.t. a fuzzy number when a certain L1 metric is considered. This paper aims to analyze connections between both approaches along with some properties of the first...

Since its launch in December 2013, the Gaia space mission has collected and continues to collect tremendous amounts of information concerning the objects that populate our Galaxy and beyond. The international Gaia Data and Analysis Consortium (DPAC) is in charge of developing computer algorithms that...

This paper presents an automatic method to design interval type-2 fuzzy systems for load forecasting applications using a memetic algorithm. This hybridisation of a variable-length genetic algorithm and a gradient descent method allows for concurrent learning of the system’s parameters and structure...

The aim of the 2015 IFSA-EUSFLAT International Time Series Competition, Computational Intelligence in Forecasting (CIF), is to evaluate the performance of computational intelligence-based approaches to forecast time series of different nature. The participants must propose a unique consistent methodology...

Currently there are not many data mining method available to solve the classification task in datasets with low quality values. In this paper we propose a method of imputation/classification based on neighborhood that can work with nominal and numerical attributes which can contain low quality values....

We propose linguistic associations mining as a technique to create the models of the multivariate time series. We define various linguistic evaluative expressions on the range of the values of the time series and variables derived from them. We mine linguistic associations then and interpret them as...

Transfer learning provides an approach to solve target tasks more quickly and effectively by using previously-acquired knowledge learned from source tasks. Most of transfer learning approaches extract knowledge of source domain in the given feature space. The issue is that single perspective can t mine...

classification based on the One-vs-One decomposition strategy has shown a high quality for addressing those problems with multiple classes, even if the learning model enables the discrimination among several concepts. The main phase of the pairwiselearning is the decision process, where the outputs of...

The aim of this contribution is to show the opportunities of applying of fuzzy time series models to predict multiple heterogeneous time series, given at International Time Series Forecasting Competition [http://irafm.osu.cz/cif/main.php]. The dataset of this competition includes 91 time series of different...

Redundancy is a handicap in association rules. It becomes a limitation to use rules models in order to support the decision-making process. A technique based on user knowledge has been proposed recently, which aims at eliminating redundancy. However, it ignores the imprecise nature of knowledge. In this...

Evolving classification models are designed to solve online tasks with demands restricting computational power and memory. The present paper proposes an evolving version of an established fuzzy classification approach based on fuzzy pattern classes. The approach incorporates a novel type of virtual sample...

While online product reviews are valuable sources of information to facilitate consumers’ purchase decisions, it is deemed meaningful and important to distinguish helpful reviews from unhelpful ones for consumers fac-ing huge amounts of reviews nowadays. Thus, in light of review classification, this...

This paper presents a comparative analysis of the performance of fuzzy approaches on the task of predicting customer review ratings using a computational intelligence framework based on a genetic algorithm for data dimensionality reduction. The performance of the Fuzzy C-Means (FCM), a neurofuzzy approach...

Synchrony among neuronal impulses (or spikes) plays, according to some of the most prominent neural coding hypotheses, a central role in information processing in biological neural networks. When dealing with multiple electrode recordings (i.e., spike trains) modelers generally characterize synchrony...

In this paper the results of rule-base construction parameter optimization for a multi-stroke fuzzy character recognizer are compared. The experiment covers the investigation of the optimal number of samples used to build the rule-base and the parameter of the method to generate fuzzy sets from the training...

We present methods to find (significant) frequent synchronous patterns in event sequences, using a graded notion of synchrony that captures both the number of instances of a pattern as well as the precision of synchrony of its constituting events. Since transferring earlier work (using a binary notion...

In recent years, Fuzzy Time Series have been considered a promising tool to deal with forecasting problems due to the ease to model the problems, the satisfactory results obtained and also to the low computational cost required. However, the long experience with traditional methods coming from statistics,...

In order to to predict regime duration in a given chaotic system, for a set of output prototypes are available, we propose to use a clustering technique for the definition of classes of regime duration, which are then used by a chosen classifier. In this way, the exact boundaries between classes are...

In current literature, the degrading performances of iris recognition systems is put in user’s responsibility (Biometric Menagerie-BM), or explained through a va-gue mix of time-related changes in biometric pattern, its presentation and the acquisition sensor (Template Age-ing-TA). Actually, BM and TA...

Facilitating tasks such as image search is one of the goals of image annotation methods that automatically assign keywords to images. In order to achieve as accu-rate annotation on object level as possible, and to reduce negative influence of misclassified objects on the infer-ence of scenes, a knowledge...

Usually, the expected loss minimization criterion is used in order to look for the optimal model that expresses a certain response variable as a function of a collection of attributes. We generalize this criterion, in order to be able to deal also with those situations where a numerical loss function...

Fuzzy c-means is a well known fuzzy clustering algorithm. It is an unsupervised clustering algorithm that permits us to build a fuzzy partition from data. The algorithm depends on a parameter m which corresponds to the degree of fuzziness of the solution. Large values of m will blur the classes and all...

The introduction of Resource Description Framework (RDF) as a fundamental data representation format of Semantic Web is changing a way how data is stored on the Internet. The intrinsic features of RDF data, i.e., its interconnections and simplicity of expressing information as triples: two entities connected...

Mining emerging patterns aims at contrasting data sets and identifying itemsets that characterise a data set by contrast to a reference data set, so as to capture and highlight their differences. This paper considers the case of emerging gradual patterns, to extract discriminant attribute co-variations....