Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
MTZ-Fachkongress

Antriebe und Energiesysteme von morgen


Austausch von Automobil- und Energiewirtschaft

Am 14./15. Mai geht es in Chemnitz um die Mobilitätswende durch Elektro- und Wasserstofffahrzeuge.
Erweiterte Suche
Anmelden
nach oben

2016 | Buch

Managing Natural Noise in Recommender Systems Kapitel lesen Erstes Kapitel lesen

Theory and Practice of Natural Computing

5th International Conference, TPNC 2016, Sendai, Japan, December 12-13, 2016, Proceedings

herausgegeben von: Carlos Martín-Vide, Takaaki Mizuki, Miguel A. Vega-Rodríguez

Verlag: Springer International Publishing

Buchreihe : Lecture Notes in Computer Science

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

This book constitutes the refereed proceedings of the 5th International Conference on Theory and Practice of Natural Computing, TPNC 2016, held in Sendai, Japan, in December 2016.

The 16 revised full papers presented together with one invited talk in this book were carefully reviewed and selected from 33 submissions. The papers are grouped in topical sections on applications of natural computing, evolutionary computation, formal models, and machine learning.

Inhaltsverzeichnis

Frontmatter

Invited Talk

Frontmatter
Managing Natural Noise in Recommender Systems
Abstract
E-commerce customers demand quick and easy access to suitable products in large purchase spaces. To support and facilitate this purchasing process to users, recommender systems (RSs) help them to find out the information that best fits their preferences and needs in an overloaded search space. These systems require the elicitation of customers’ preferences. However, this elicitation process is not always precise either correct because of external factors such as human errors, uncertainty, human beings inherent inconsistency and so on. Such a problem in RSs is known as natural noise (NN) and can negatively bias recommendations, which leads to poor user’s experience. Different proposals have been presented to deal with natural noise in RSs. Several of them require additional interaction with customers. Others just remove noisy information. Recently, new NN approaches dealing with the ratings stored in the user/item rating matrix have raised to deal with NN in a better and simpler way. This contribution is devoted to provide a brief review of the latter approaches revising crisp and fuzzy approaches for dealing with NN in RSs. Eventually it points out as a future research the management of NN in other recommendation scenarios as group RSs.
Luis Martínez, Jorge Castro, Raciel Yera

Applications of Natural Computing

Frontmatter
Realization of Periodic Functions by Self-stabilizing Population Protocols with Synchronous Handshakes
Abstract
We consider in the following the problem of realizing periodic functions by a collection of finite state-agents that cooperate by interacting with each other. More formally, given a periodic non-negative integer function f that maps the set of non-negative integers \(\mathbf{N}\) to itself, we aim in this paper at designing a distributed protocol with a state set Q and a subset \(S \subseteq Q\), such that, for any initial configuration \(C_0\), with probability 1, there are a time instant \(t_0\) and a constant \(d \in \mathbf{N}\) satisfying \(f(t+d) = \nu _S(C_t)\) for all \(t \ge t_0\), where \(\nu _S(C)\) is the number of agents with a state in S in a configuration C. The model that we consider is a variant of the population protocol (PP) model in which we assume that each agent is involved in an interaction at each time instant t, hence the notion of synchronous handshakes. These additional assumptions on the model are necessary to solve the considered problem. We also assume that the interacting pairs are matched uniformly at random.
Anissa Lamani, Masafumi Yamashita
Localized Load Balancing in RFID Systems
Abstract
Radio Frequency Identification (RFID) technology consists of uniquely identifiable and inexpensive tags as well as readers that monitor these tags with the help of radio frequency signal. Balancing the load of tags among the readers in a multi-reader RFID system is an important issue to successfully collect data from all the tags. In this paper we try to minimize the reading time of the readers by distributing the tags among the readers in the RFID system. We introduce a Cellular Automaton (CA) based localized load balancing algorithm to transfer tags from highly loaded readers to readers with lower number of tags associated with them. Our simulation results exhibit that our algorithm outperforms the existing algorithm.
Ahnaf Munir, Md Sakhawat Hossen, Salimur Choudhury
The Fuel-Efficient Platooning of Heavy Duty Vehicles by Mathematical Programming and Genetic Algorithm
Abstract
As fuel accounts for a significant proportion of the total operational cost of heavy duty vehicles (HDVs) and to alleviate the environmental impacts, companies are seeking for novel practical methods to reduce fuel consumption. Platooning is an effective approach in which a string of HDVs driving close behind each other is formed. This reduces the aerodynamic drag leading to a reduction in the overall resistive force on vehicles which can provide an amount of fuel-saving in each of the following vehicles. These trivial reductions result in a considerable decrease in the total fuel consumption corresponding to all the vehicles. In this paper, we propose a mathematical model for the fuel-efficient platooning problem with a deadline for each vehicle (truck) to reach its destination by then. We create a graph partly based on the German road network considering only 20 important cities and solve 50 instances including 10 to 50 trucks. The small samples are solved by the LINDO solver in real time but as the problem has a high computational complexity, a Genetic Algorithm is applied to obtain fast solutions of a good quality for larger instances. The final results show a satisfactory fuel-saving in all of the cases.
Abtin Nourmohammadzadeh, Sven Hartmann
How to Implement a Random Bisection Cut
Abstract
By using a deck of cards, it is possible to realize a secure computation. In particular, since a new shuffling operation, called a random bisection cut, was devised in 2009, many efficient protocols have been designed. The shuffle functions in the following manner. A sequence of cards is bisected, and the two halves are swapped randomly. This results in two possible cases, depending on whether the two halves of the card sequence are swapped or not. Because there are only two possibilities when a random bisection cut is performed, it has been suggested that information regarding the result of the shuffle could sometimes be leaked visually. Thus, in this paper we propose some methods for implementing a random bisection cut without leaking such information.
Itaru Ueda, Akihiro Nishimura, Yu-ichi Hayashi, Takaaki Mizuki, Hideaki Sone

Evolutionary Computation

Frontmatter
A Discrete Artificial Bee Colony Algorithm Based on Similarity for Graph Coloring Problems
Abstract
In this paper, a novel non-hybrid discrete artificial bee colony (ABC) algorithm is proposed for solving planar graph coloring problems. The original ABC intends to handle only continuous optimization problems. To apply ABC to discrete problems, the original ABC operators need to be redefined over discrete space. In this work, a new algorithm based on Similarity is introduced. Compared with HDPSO, the experiment shows that the proposed method matches the competitive results and obtains higher success rate and lower average evaluation times when solving planar graph coloring problems.
Kui Chen, Hitoshi Kanoh
A Multi-objective Evolutionary Approach to Pareto Optimal Model Trees. A Preliminary Study
Abstract
Decision tree induction is inherently a multi-objective task. However, most of the conventional learning algorithms can only deal with a single-objective that may possibly aggregate multiple objectives. This paper proposes the multi-objective evolutionary approach to Pareto optimal model trees. We developed a set of non-dominated model trees for a Global Model Tree framework using efficient sort and specialized selection. Performed study covers variants with two and three objectives that relate to the tree error and the tree comprehensibility. Pareto front generated by the GMT system allows the decision maker to select desired output model according to his preferences on the conflicting objectives. Experimental evaluation of the proposed approach is performed on three real-life datasets and is confronted with competitive model tree inducers.
Marcin Czajkowski, Marek Kretowski
A Sampling-Based Metaheuristic for the Orienteering Problem with Stochastic Travel Times
Abstract
In this paper we propose a new metaheuristic approach based on sampling for the Orienteering Problem with Stochastic Travel Times (OPSTS). As in many Stochastic Combinatorial Optimization Problems, the computational bottleneck of OPSTS is in the objective function evaluation. For this reason, this study is mainly devoted to the development and integration of on-purpose, sampling-based fast objective function evaluations into metaheuristic methods. In details, we show how a Variable Neighbourhood Search Metaheuristic can be enhanced by adopting such evaluators. Experimental results show that the new sampling-based method is faster than conventional methods for the given problem, and the improvement is particularly relevant for large-scale instances.
Vassilis Papapanagiotou, Roberto Montemanni, Luca Maria Gambardella
Real Time Traffic Intersection Management Using Multi-objective Evolutionary Algorithm
Abstract
With the advent of autonomous vehicles, the field of traffic intersection management has changed. Most of the current methods for intersection management use either stochastic methods for optimizing single scheduling scenarios or deterministic algorithms to optimize parameters for intersection traffic lights. This paper proposes and explores the application of multi-objective evolutionary algorithm (MOEA) to manage a traffic intersection in real time. To achieve this goal, this work implements an intersection manager (IM) that divides the continuous problem into smaller discrete time steps. The vehicular behaviour in single time steps is then optimized, considering several optimization objectives with different goals in terms of overall performance.
Kazi Shah Nawaz Ripon, Håkon Dissen, Jostein Solaas

Formal Models

Frontmatter
Natural and Efficient Subtraction Operation in Carry Value Transformation (CVT)-Exclusive OR (XOR) Paradigm
Abstract
Carry value transformation (CVT) and Exclusive OR (XOR) operations on two non-negative integers have been defined previously in several articles. In this paper, the definition of CVT and XOR operations are extended from non-negative integer to integer domain. Thereafter various cases of integer pairs towards their convergence behaviour are thoroughly discussed. Our analyses through the convergence behavior of integer pairs are easily directed to capture the natural subtraction operation in this paradigm by representing negative integer in 2’s complement form. The average time complexity of the addition/subtraction operation is seen to be highly competitive in any bulk computation in real life scenario. In other words, in the event of bulk addition/subtraction operation to be performed, the average time complexity is seen to be highly efficient.
Jayanta Kumar Das, Pabitra Pal Choudhury, Ayesha Arora
Decreasing Entropy: How Wide to Open the Window?
Abstract
On the basis of the literature about human sentence processing we examined the parsing process from two aspects. With the help of a sentence-completion experiment we show that there is a strong relationship between the entropy of the words in the sentence and the look-ahead window of a two-phase sentence processing model. The result of our experiment showed that people intend to close the verbal complex and the noun phrase as soon as possible and our corpus-measurements support that it happens in a trigram window.
Balázs Indig, Noémi Vadász, Ágnes Kalivoda
Simulating Stochastic Dynamic Interactions with Spatial Information and Flux
Abstract
We present a conservative extension to rule based modeling languages with constructs for component attributes and functions that modify these attributes; the language has a stochastic semantics, and it is equipped with flux analysis. We show that the constructs of this language, called \(\mathsf {M}\), bring an ease especially in modeling biological systems, where spatial information is of essence. We discuss the language on models from molecular biology such as membrane diffusion systems, and actin polymerization networks, as well as models from ecology, where spatial behavior of animals as in bird flocks and fish schools are studied.
Ozan Kahramanoğulları
Implementation of Turing Machine Using DNA Strand Displacement
Abstract
The computational capability of biochemical systems is one of the major interest in the area of nanotechnology. Since Bennett proposed his thought experiment of chemical Turing machine using DNA-like molecules, many attempts for DNA Turing machine have been made. However, they are based on some hypothetical assumptions or require laboratory manipulations for each step. Here we propose an implementation of Turing machine by using DNA strand displacement cascades.
Wataru Yahiro, Masami Hagiya

Machine Learning

Frontmatter
A Quantum Annealing Approach to Biclustering
Abstract
Several problem in Artificial Intelligence and Pattern Recognition are computationally intractable due to their inherent complexity and the exponential size of the solution space. One example of such problems is biclustering, a specific clustering problem where rows and columns of a data-matrix must be clustered simultaneously. Quantum information processing could provide a viable alternative to combat such a complexity. A notable work in this direction is the recent development of the D-Wave™  computer, whose processor is able to exploit quantum mechanical effects in order to perform quantum annealing. The question motivating this work is whether the use of this special hardware is a viable approach to efficiently solve the biclustering problem. As a first step towards the solution of this problem, we show a feasible encoding of biclustering into the D-Wave™  quantum annealing hardware, and provide a theoretical analysis of its correctness.
Lorenzo Bottarelli, Manuele Bicego, Matteo Denitto, Alessandra Di Pierro, Alessandro Farinelli
Determining Player Skill in the Game of Go with Deep Neural Networks
Abstract
The game of Go has recently been an exuberant topic for AI research, mainly due to advances in Go playing software. Here, we present an application of deep neural networks aiming to improve the experience of humans playing the game of Go online. We have trained a deep convolutional network on 188,700 Go game records to classify players into three categories based on their skill. The method has a very good accuracy of 71.5 % when classifying the skill from a single position, and 77.9 % when aggregating predictions from one game. The performance and low amount of information needed allow for a much faster convergence to true rank on online Go servers, improving user experience for new-coming players. The method will be experimentally deployed on the Online Go Server (OGS).
Josef Moudřík, Roman Neruda
Flexible Generalized Fuzzy Petri Nets for Rule-Based Systems
Abstract
In 2015, the modified generalised fuzzy Petri nets (mGFP-nets) were proposed. This paper describes an extended class of mGFP-nets called flexible generalised fuzzy Petri nets (FGFP-nets). The main difference between the latter net model and the mGFP-net concerns transition operator \(Out_1\) appearing in a triple of operators \((In, Out_1, Out_2)\) in a mGFP-net. The operator \(Out_1\) for each transition is determined automatically by the GTVC algorithm, using the value of In and the value of truth degree function \(\beta \) in the net. This modification has significant influence on optimization of the modelled system by the FGFP-nets. The choice of suitable operators for the modelled system is very important, especially in systems described by incomplete, imprecise and/or vague information. The proposed approach can be used both for control design as well as knowledge representation and modelling of reasoning in decision support systems.
Zbigniew Suraj, Piotr Grochowalski, Sibasis Bandyopadhyay
Learning Grammar Rules in Probabilistic Grammar-Based Genetic Programming
Abstract
Grammar-based Genetic Programming (GBGP) searches for a computer program in order to solve a given problem. Grammar constrains the set of possible programs in the search space. It is not obvious to write an appropriate grammar for a complex problem. Our proposed Bayesian Grammar-Based Genetic Programming with Hierarchical Learning (BGBGP-HL) aims at automatically designing new rules from existing relatively simple grammar rules during evolution to improve the grammar structure. The new grammar rules also reflects the new understanding of the existing grammar under the given fitness evaluation function. Based on our case study in asymmetric royal tree problem, our evaluation shows that BGBGP-HL achieves the best performance among the competitors. Compared to other algorithms, search performance of BGBGP-HL is demonstrated to be more robust against dependencies and the changes in complexity of programs.
Pak-Kan Wong, Man-Leung Wong, Kwong-Sak Leung
Backmatter
Metadaten
Titel
Theory and Practice of Natural Computing
herausgegeben von
Carlos Martín-Vide
Takaaki Mizuki
Miguel A. Vega-Rodríguez
Copyright-Jahr
2016
Electronic ISBN
978-3-319-49001-4
Print ISBN
978-3-319-49000-7
DOI
https://doi.org/10.1007/978-3-319-49001-4

Premium Partner

    Bildnachweise