Skip to main content

2016 | Buch

Challenging Problems and Solutions in Intelligent Systems

herausgegeben von: Guy de Trė, Przemysław Grzegorzewski, Janusz Kacprzyk, Jan W. Owsiński, Wojciech Penczek, Sławomir Zadrożny

Verlag: Springer International Publishing

Buchreihe : Studies in Computational Intelligence

insite
SUCHEN

Über dieses Buch

This volume presents recent research, challenging problems and solutions in Intelligent Systems– covering the following disciplines: artificial and computational intelligence, fuzzy logic and other non-classic logics, intelligent database systems, information retrieval, information fusion, intelligent search (engines), data mining, cluster analysis, unsupervised learning, machine learning, intelligent data analysis, (group) decision support systems, intelligent agents and multi-agent systems, knowledge-based systems, imprecision and uncertainty handling, electronic commerce, distributed systems, etc. The book defines a common ground for sometimes seemingly disparate problems and addresses them by using the paradigm of broadly perceived intelligent systems. It presents a broad panorama of a multitude of theoretical and practical problems which have been successfully dealt with using the paradigm of intelligent computing.

Inhaltsverzeichnis

Frontmatter

Foundations of Intelligent Computing

Frontmatter
SMT-Based Parameter Synthesis for Parametric Timed Automata
Abstract
We present a simple method for representing finite executions of Parametric Timed Automata using Satisfiability Modulo Theories (SMT). The transition relation of an automaton is translated to a formula of SMT, which is used to represent all the prefixes of a given length of all the executions. This enables to underapproximate the set of parameter valuations for the undecidable problem of parametric reachability. We introduce a freely available, open-source tool PTA2SMT and show its application to the synthesis of parameter valuations under which a timed mutual exclusion protocol fails.
Michał Knapik, Wojciech Penczek
On the Identification of $$\alpha $$ α -Asynchronous Cellular Automata in the Case of Partial Observations with Spatially Separated Gaps
Abstract
In this paper we present a statistical method, based on frequencies, for identifying so-called \(\alpha \)-asynchronous Cellular Automata from partial observations, i.e. pre-recorded configurations of the system with some cells having an unknown (missing) state. The presented method, in addition to finding the unknown Cellular Automaton, is able to unveil the missing state values with high accuracy.
Witold Bołt, Barbara Wolnik, Jan M. Baetens, Bernard De Baets
k-Arithmetic Sequences—Theory and Applications
Abstract
The notion of an arithmetic progression was extended to embrace the class of polynomials of degree \(k>1\). Some properties of difference sequences are analyzed and their connections with some number-theory problems are studied. In particular, a certain aspects of Fermat’s Last Theorem and the Fibonacci numbers are revisited.
Adam Kołacz

Intelligent Techniques in Data Mining

Frontmatter
Forecasting of Short Time Series with Intelligent Computing
Abstract
Although time series analysis and forecasting have been studied since the seventeenth century and the literature related to its statistical foundations is extensive, the problem arises when the assumptions underlying statistical modeling are not fulfilled due to the shortness of available data. In such cases, additional expert knowledge is needed to support the forecasting process. The inclusion of prior knowledge may be easily formalized with the Bayesian approach. However, the proper formulation of prior probability distributions is still one of the main challenges for practitioners. Hopefully, intelligent computing can support the formulation of the prior knowledge. In this paper, we review recent trends and challenges of the interdisciplinary research on time series forecasting with the use of intelligent computing, especially fuzzy systems. Then, we propose a method that incorporates fuzzy trends and linguistic summaries for the forecasting of short time series. Experiments show that it is a very promising and human-consistent approach.
Katarzyna Kaczmarek, Olgierd Hryniewicz
An Improved Adaptive Self-Organizing Map
Abstract
We propose a novel adaptive Self-Organizing Map (SOM). In the introduced approach, the SOM neurons’ neighborhood widths are computed adaptively using the information about the frequencies of occurrences of input patterns in the input space. The neighborhood widths are determined independently for each neuron in the SOM grid. In this way, the proposed SOM properly visualizes the input data, especially, when there are significant differences in frequencies of occurrences of input patterns. The experimental study on real data, on three different datasets, verifies and confirms the effectiveness of the proposed adaptive SOM.
Dominik Olszewski, Janusz Kacprzyk, Sławomir Zadrożny
Support Vector Machines in Fuzzy Regression
Abstract
This paper presents methods of estimating fuzzy regression models based on support vector machines. Starting from the approaches known from the literature and dedicated to triangular fuzzy numbers and based on linear and quadratic loss, a new method applying loss function based on the Trutschnig distance is proposed. Furthermore, a generalization of those models for fuzzy numbers with trapezoidal membership function is given. Finally, the proposed models are illustrated and compared in the examples and some of their properties are discussed.
Paulina Wieszczy, Przemysław Grzegorzewski
Quantified Quality Criteria of Contextual Bipolar Linguistic Summaries
Abstract
In our previous work we have proposed the concept of a contextual bipolar linguistic summary. It is an extension of the seminal concept of a linguistic summary proposed by Yager which is based on the application of Zadeh’s calculus of linguistically quantified propositions to more intuitive and human consistent data mining. The original Yager’s concept evolved over the years and our recent contribution to this theory is the inclusion of the concept of bipolarity of information and preferences. This enrichment of the notion of the linguistic summary calls for specialized measures of its quality, interestingness, etc. We further study this problem and in this paper we propose a new approach to assessing the quality of this type of summaries.
Mateusz Dziedzic, Janusz Kacprzyk, Sławomir Zadrożny, Guy De Tré

Multi-agent Based Technologies

Frontmatter
Microgrids and Management of Power
Abstract
The advancements in technology, changes in power usage patterns and the pressure on the renewable technologies are forcing changes in the electric power grids and the electric infrastructure. The new challenges appear and also new ways of dealing with problems. The concepts of prosumer and microgrid emerged. To make these feasible and safe, the management of power usage and production is required to maintain the balance of power. Demand side management and production side management consider techniques to deal with cost-effective power balancing problems. In this article, the concept of complex energy management system is presented; a specific case study for a research and education center is considered. This limits the scope of the management, as the demand side management should not limit the users in performing their professional duties, this restriction is less present in the case of households. The outline of the system is presented with the short description of its elements.
Weronika Radziszewska, Zbigniew Nahorski
Transaction Protocol and Mechanisms for Adaptive Management of Long-Running Tasks
Abstract
Execution of real-world services can lead to many unexpected events that need to be handled. So that failures of tasks composed of such services frequently occur. Mechanisms for automated task accomplishment and failure handling in open and heterogeneous systems are proposed. They are based on general protocols derived from the well known OASIS Web Services Transaction standard WS-TX for business transactions. The protocols and mechanisms are implemented in the prototype Autero multi-robot system in which the robots cooperate so as to accomplish complex tasks.
Marcin Stępniak
A Hybrid Approach to Parallelization of Monte Carlo Tree Search in General Game Playing
Abstract
In this paper, we investigate the concept of a parallelization of Monte Carlo Tree Search applied to games. Specifically, we consider General Game Playing framework, which has originated at Stanford University in 2005 and has become one of the most important realizations of the multi-game playing idea. We introduce a novel parallelization method, called Limited Hybrid Root-Tree Parallelization, based on a combination of two existing ones (Root and Tree Parallelization) additionally equipped with a mechanism of limiting actions available during the search process. The proposed approach is evaluated and compared to the non-limited hybrid version counterpart and to the Tree Parallelization method. The advantages over Root Parallelization are derived on a theoretical basis. In the experiments, the proposed method is more effective than Tree Parallelization and also than non-limited hybrid version in certain games.
Maciej Świechowski, Jacek Mańdziuk

Intelligent Computing in Decision Support Systems

Frontmatter
A Consensus Reaching Support System for Multi-criteria Decision Making Problems
Abstract
We present an extension of a consensus reaching support system presented in our previous works to additionally accommodate a multi-criteria evaluation of options and importance weights of all criteria given by each agent (individual). The multi-criteria setting implies a need for some modification of concepts, tools and techniques proposed in our previous works with a single criterion. To improve the efficiency of the process we use some additional suggestions/hints provided for the moderator in the form of linguistic summaries, modified to the multi-criteria setting. We present an application for a real world problem which involves reaching of a sufficient agreement in the small group of human agents. The results obtained are intuitively appealing, and promising in terms of time and costs of opinion changes to reach a sufficient consensus.
Dominika Gołuńska, Janusz Kacprzyk
Improving Spatial Estimates of Greenhouse Gas Emissions at a Fine Resolution: A Review of Approaches
Abstract
The paper presents a review of the methods which can be useful in quantification of greenhouse gas (GHG) emissions at a fine spatial resolution. The discussed approaches include: spatial disaggregation of GHG emissions based on proxy data and/or statistical modeling of spatial correlation, an estimation of fossil fuel emission changes from measuring rates (mixing ratios) of tracer (like \(^{14}\)CO\(_{2}\)) concentrations in the atmosphere, the atmospheric inversion methods, and flux tower observations.
Joanna Horabik-Pyzel, Zbigniew Nahorski
A New Approach to the Multiaspect Text Categorization by Using the Support Vector Machines
Abstract
In our earlier work we introduced the concept of the multiaspect text categorization (MTC) task which has its roots in relevant practical problems of managing collections of documents at many, if not all, commercial companies and, above all, public institutions. Specifically, it is a well defined general problem which boils down to the classification of textual documents at two levels: first, to a general category, and—second—to a specific sequence of documents within such a category. While the former task may be dealt with the use of some standard text categorization techniques, the latter one is more challenging due to, first of all, a limited number of training documents. On the other hand, it is assumed that there is some natural logic, for instance, resulting from rules and regulations, behind the succession of documents within the sequences which can be exploited to make a decision as to the assignment of a new document to a proper sequence. We have studied the MCT problem in a number of papers and proposed some solutions to it. Here we propose a new solution which is based on the use of the support vector machines (SVMs) which are known as a very effective technique to solve various classification tasks. We consider the application of SVMs in a specific context, determined by the characteristics of the MTC problem, and by a specific data set used for the experimentation. The use of the SVMs has implied a new, more sophisticated representation of the documents and their sequences which has made it possible to obtain promising results in computational experiments. Moreover, the proposed approach is flexible and may be considerably modified and extended to cover many possible problem versions.
Sławomir Zadrożny, Janusz Kacprzyk, Marek Gajewski

Intelligent Text and Data Retrieval: Towards a Better Representation of Users Intensions

Frontmatter
Content Data Based Schema Matching
Abstract
A novel automatic method for detecting corresponding attributes in schemas based on content data is studied. More specifically, our proposed method for the detection of coreferent attributes in schemas is based on a statistical and lexical comparison of content data and detected coreferent tuples across multiple datasets, which increase the possibility of correct schema matching. We will show that knowledge of even a small number of coreferent tuples is sufficient to establish correct matching between corresponding attributes of heterogeneous schemas. The behaviour of the novel schema matching technique has been evaluated on several real life datasets, giving a valuable insight in the influence of the different parameters of our approach on the results obtained.
Marcin Szymczak, Antoon Bronselaer, Sławomir Zadrożny, Guy De Tré
Prior-Art Relevance Ranking Based on the Examiner’s Query Log Content
Abstract
This work belongs to the domain of technical information retrieval (IR) and, more specifically, patent retrieval. We show that the recorded history of patent examiner’s search queries can be used to create a more effective method of finding prior art patents than search methods based on titles and claims. We verify the performance of the proposed method experimentally. Our experiments show that we can almost double the recall measure, compared to classical techniques based on titles and claims. The other contribution of our work is the creation of a database of over half a million patent examiners queries (recorded search activity over the patents prosecution process). The paper also discusses the limitations of the current work and the ongoing research to further improve the proposed approach.
Jakub Wajda, Wlodek Zadrozny
What NEKST?—Semantic Search Engine for Polish Internet
Abstract
We introduce a new semantic search engine, developed at our institute. Its unique feature is the automatic construction of semantic resources, like discovery of millions of facts, IS-A relations and automated generation of sentimental analysis dictionaries. We developed a new method of document categorization. The engine can be queried in natural language and possesses interfaces to be used not only by humans but also by machines.
Dariusz Czerski, Krzysztof Ciesielski, Michał Dramiński, Mieczysław Kłopotek, Paweł Łoziński, Sławomir Wierzchoń
Metadaten
Titel
Challenging Problems and Solutions in Intelligent Systems
herausgegeben von
Guy de Trė
Przemysław Grzegorzewski
Janusz Kacprzyk
Jan W. Owsiński
Wojciech Penczek
Sławomir Zadrożny
Copyright-Jahr
2016
Electronic ISBN
978-3-319-30165-5
Print ISBN
978-3-319-30164-8
DOI
https://doi.org/10.1007/978-3-319-30165-5