Skip to main content
main-content

Über dieses Buch

This book presents a collection of research findings and proposals on computer science and computer engineering, introducing readers to essential concepts, theories, and applications. It also shares perspectives on how cutting-edge and established methodologies and techniques can be used to obtain new and interesting results. Each chapter focuses on a specific aspect of computer science or computer engineering, such as: software engineering, complex systems, computational intelligence, embedded systems, and systems engineering. As such, the book will bring students and professionals alike up to date on key advances in these areas.

Inhaltsverzeichnis

Frontmatter

A Comprehensive Context-Aware Recommender System Framework

Context-Aware Recommender System research has realized that effective recommendations go beyond recommendation accuracy, thus research has paid more attention to human and context factors, as an opportunity to increase user satisfaction. Despite the strong tie between recommendation algorithms and the human and context data that feed them, both elements have been treated as separated research problems. This document introduces MoRe, a comprehensive software framework to build context-aware recommender systems. MoRe provides developers a set of state of the art recommendation algorithms for contextual and traditional recommendations covering the main recommendation techniques existing in the literature. MoRe also provides developers a generic data model structure that supports an extensive range of human, context and items factors that is designed and implemented following the object-oriented paradigm. MoRe saves developers the tasks of implementing recommendation algorithms, and creating a structure to support the information the system will require, proving concrete functionality, and at the same time is generic enough to allow developers adapt its features to fit specific project needs.
Sergio Inzunza, Reyes Juárez-Ramírez

Executive Functions and Their Relationship with Interaction Design

Human Factors have been for several decades one of the main factors of contention when trying to develop a usable system. From physical to cognitive characteristics and everything in between, the attributes of a user can impact on several aspects of the design of said software. Although there are guidelines for some characteristics there is no definitive model with what characteristics to consider, what metrics to use and their effect on the interface. In this chapter we talk about the effect of some of the executive functions (working memory and cognitive flexibility) depending on the interface design pattern used, and the relationship with the cognitive load produced by the design pattern.
Andrés Mejía Figueroa, J. Reyes Juárez Ramírez, David Saldaña Sage

Integrating Learning Styles in an Adaptive Hypermedia System with Adaptive Resources

At present, e-Learning for distance education is increasing, but most of them do not take into account the individualities of the students, such as learning styles, offering the same content to all. This paper presents, an adaptive hypermedia web system based on learning styles; At the start of the session, the user performs the Felder and Soloman learning styles questionnaire to obtain information from the students. These results show learning objects (OA) for the computer programming subject to the students to analyze later the data of interaction of them with the system and thus determine if these objects apply to that style of learning, in case they solve them in many attempts, feedback is sent to the teacher to modify them based on their learning styles and exercises that are resolved in fewer efforts to improve the course, when the system collects more user interaction data will make better recommendations to new users using simple sequencing.
Carlos Hurtado, Guillermo Licea, Mario Garcia-Valdez

On Modeling Tacit Knowledge for Intelligent Systems

In an attempt to support efforts to narrow the gap between current Artificial Intelligence and actual intelligent human behavior, this paper addresses Tacit Knowledge. Tacit Knowledge is analyzed and separated into articulable and inarticulable for ease of scrutiny. Concepts and ideas are taken up from knowledge management literature aiming to understand the scope of knowledge. Among the bailed out concepts “particulars” and “concepts” stand out, and “preconcept” is suggested as an intermediate phase between the former two. These concepts are placed into mental processes of knowledge resulting in an alternative neurological model of knowledge acquisition. The model’s target is to provide a picture as detailed as possible of the processes executed by the brain to make learning achievable. It encompasses from sensing the stimuli that is produced by the environment that are collected by sensory receptors to turn them into electrical impulses that are transmitted to the brain to climax with the emergence of concepts, from which increasingly complex knowledge is built. The model is then expanded to the social level.
Violeta Ocegueda-Miramontes, Antonio Rodríguez-Díaz, Juan R. Castro, Mauricio A. Sanchez, Olivia Mendoza

Influence of the Betweenness Centrality to Characterize the Behavior of Communication in a Group

The behavior of the distribution of a rumor must emerge according to the relations between the individuals. Taking as a reference that human society creates links of friendship through random encounters and conscious decisions, therefore, a rumor can be spread considering the degree of grouping that individuals have, also their location in the network and if they decide to cooperate or not. Considering the analysis of the topology that interconnects a set of individuals, relationships are detected between them that allow recognizing their centrality of degree, betweenness, and closeness. For a rumor to spread it requires that the individual has an incentive by which he decides to cooperate or not in the distribution of it. In this chapter, we propose an agent-based model that allows the identification of the central measures of each of the individuals that integrate a group which has a topology based on the Barbell’s graph.
K. Raya-Díaz, C. Gaxiola-Pacheco, Manuel Castañón-Puga, L. E. Palafox, R. Rosales Cisneros

Multi-layered Network Modeled with MAS and Network Theory

Complex networks have been widely used to model diverse real-world systems with great success. However, a limitation emerges when it comes to modeling real multilayered systems where nodes belong to different layers at the same time and have multiple interactions. Some of the research conducted under the multilayer approach focuses on the structure and topology of the network; to measure the resilience and robustness simulating the elimination of nodes in the network in this work, the theory of networks is used to model the problem of the European air transport network. Centrality measures will be applied to obtain information from the crucial entities Within a network and the topology formed layer by layer, multi-agent systems are applied to simulate the scenario of a negotiation when the risk of cancellation of flights is presented. By including multiagent systems, we intend to approximate a bit more to real systems since it allows to simulate multiple scenarios, provide them with mechanisms of negotiation and decision making.
Jose Parra, Carelia Gaxiola, Manuel Castañón-Puga

A Fuzzy Inference System and Data Mining Toolkit for Agent-Based Simulation in NetLogo

In machine learning, hybrid systems are methods that combine different computational techniques in modeling. NetLogo is a favorite tool used by scientists with limited ability as programmers who aim to leverage computer modeling via agent-oriented approaches. This paper introduces a novel modeling framework, JT2FIS NetLogo, a toolkit for integrating interval Type-2 fuzzy inference systems in agent-based models and simulations. An extension to NetLogo, it includes a set of tools oriented to data mining, configuration, and implementation of fuzzy inference systems that modeler used within an agent-based simulation. We discuss the advantages and disadvantages of integrating intelligent systems in agent-based simulations by leveraging the toolkit, and present potential areas of opportunity.
Josue-Miguel Flores-Parra, Manuel Castañón-Puga, Carelia Gaxiola-Pacheco, Luis-Enrique Palafox-Maestre, Ricardo Rosales, Alfredo Tirado-Ramos

An Approach to Fuzzy Inference System Based Fuzzy Cognitive Maps

In the search of modeling methodologies for complex systems various attempts have been made, and so far, all have been inadequate in one thing or another leading the pathway open for the next better tool. Fuzzy cognitive maps have been one of such tools, although mainly used for decision making in what-if scenarios, they can also be used to represent complex systems. In this paper, we define an approach of fuzzy inference system based fuzzy cognitive map for modeling dynamic systems, where the complex model is defined by means of fuzzy IF-THEN rules which represent the behavior of the system in an easy to understand format, therefore facilitating a tool for complex system design. Various examples of dynamic systems are shown used as a means to demonstrate the ease of use, design and capability of the proposed approach.
Itzel Barriba, Antonio Rodríguez-Díaz, Juan R. Castro, Mauricio A. Sanchez

Detecting Epilepsy in EEG Signals Using Time, Frequency and Time-Frequency Domain Features

Seizures caused by epilepsy are unprovoked, they disrupt the mantel activity of the patient and impair their normal motor and sensorial functions, endangering the patient’s well being. Exploiting today’s technology it is possible toe create automatic systems to monitor and evaluate patients. An area of special interest is the automatic analysis of EEG signals. This paper presents extensive analysis of feature extraction and classification methods that have reported good results in other EEG based problems. Several methods are detailed to extract 52 features from the time, frequency and time-frequency domains in order to characterize the EEG signals. Additionally, 10 different classification models, together with a feature selection method, are implemented using these features to identify if a signal corresponds to an epileptic state. The experiments were performed using the standard BONN and the proposed method achieve results comparable to those in the state-of-the-art for the three and four classes problems.
D. E. Hernández, L. Trujillo, E. Z-Flores, O. M. Villanueva, O. Romo-Fewell

Big Data and Computational Intelligence: Background, Trends, Challenges, and Opportunities

The boom of technologies such as social media, mobile devices, internet of things, and so on, has generated enormous amounts of data that represent a tremendous challenge, since they come from different sources, different formats and are being generated in real time at an exponential speed which brings with it new necessities, opportunities, and many challenges both in the technical and analytical area. Some of the prevailing necessities lie on the development of computationally efficient algorithms that can extract value and knowledge from data and can manage the noise within in it. Computational intelligence can be seen as a key alternative to manage inaccuracies and extract value from Big Data, using fuzzy logic techniques for a better representation of the problem. And, if the concept of granular computing is also added, we will have new opportunities to decomposition of a complex data model into smaller, more defined, and meaningful granularity levels, therefore different perspectives could yield more manageable models. In this paper, two related subjects are covered, (1) the fundamentals and concepts of Big Data are described, and (2) an analysis of how computational intelligence techniques could bring benefits to this area is discussed.
Sukey Nakasima-López, Mauricio A. Sanchez, Juan R. Castro

Design of a Low-Cost Test Plan for Low-Cost MEMS Accelerometers

The present work proposes a Test Plan to evaluate the performance of low-cost MEMS accelerometers, currently some of these sensors suffer from non-linearities in its outputs caused primarily by scale factors, biases and random noise, some of these factors can be compensated to a certain extent. The Test Plan is divided on three stages of testing, with each one testing different aspects of the sensor; and for those devices that are found suitable, a characterization of a basic accelerometer model is recommended to help increase their performance so that they may be able to be used in high demanding applications, such as Inertial Navigation Systems. It is important to acknowledge that this Test Plan does not ensure that every sensor will be able to be compensated, and should be primarily used for finding suitable sensors.
Jesús A. García López, Leocundo Aguilar

Evaluation of Scheduling Algorithms for 5G Mobile Systems

One of the key elements of the fifth generation (5G) of mobile communication systems is the support of a large number of users communicating through a wide range of devices and applications. These conditions give rise to heterogeneous traffic offered to the network. In order to carry such traffic in a wireless network, the design and development of schedulers capable of considering the conditions of each user is needed. In this chapter a Model Based Design (MBD) and Model Based Testing (MBT) approach are used to implement and evaluate different scheduling algorithms that consider the Quality of Service (QoS) requirements of each user as well as the individual channel conditions. The development process is achieved through a hardware platform consisting of an FPGA and a System on Chip in order to provide an emulation environment. The development process as well as the results obtained (in terms of throughput and fairness) through the evaluation of Maximum Rate (MR), Round Robin (RR), Proportional Fair (PF) and a proposed novel UE-based Maximum Rate (UEMR) scheduling algorithms are presented. Using MBD together with MBT the developed scheduling algorithms can be further enhanced and exported to real world applications.
Christian F. Müller, Guillermo Galaviz, Ángel G. Andrade, Irina Kaiser, Wolfgang Fengler

User Location Forecasting Based on Collective Preferences

With the proliferation of mobile devices and the huge variety of sensors they incorporate, it is possible to register the user location on the move. Based on historical records, it is feasible to predict user location in space or space and time. Studies show that user mobility patterns have a high degree of repetition and this regularity has been exploited to forecast the next location of the user. Furthermore, proposals have been made to forecast user location in space and time; in particular, we present a spatio-temporal prediction model that we developed to forecast user location in a medium-term with good accuracy results. After explaining how collaborative filtering (CF) works, we explore the feasibility of using collective preferences to avoid missing POIs and therefore increase the prediction accuracy. To test the performance of the method based on CF, we compare our spatio-temporal prediction model with and without using the method based on CF.
Jorge Alvarez-Lozano, J. Antonio García-Macías, Edgar Chávez

Unimodular Sequences with Low Complementary Autocorrelation Properties

The design of sequences with constant amplitude in the time domain, that possess specific autocorrelation and specific complementary autocorrelation functions, is here addressed. The proposed sequences can be used in the identification of strictly linear (SL) and widely linear systems (WL) by making use of the second order characteristics of the process observed at the output of the system. Through the analysis developed, the main differences of the disclosed sequences with those commonly used for conventional SL processing are highlighted. Theoretical results are accompanied with numerical simulations to show the performance of the sequences here designed.
Israel Alejandro Arriaga-Trejo, Aldo Gustavo Orozco-Lugo
Weitere Informationen

Premium Partner

BranchenIndex Online

Die B2B-Firmensuche für Industrie und Wirtschaft: Kostenfrei in Firmenprofilen nach Lieferanten, Herstellern, Dienstleistern und Händlern recherchieren.

Whitepaper

- ANZEIGE -

Best Practices für die Mitarbeiter-Partizipation in der Produktentwicklung

Unternehmen haben das Innovationspotenzial der eigenen Mitarbeiter auch außerhalb der F&E-Abteilung erkannt. Viele Initiativen zur Partizipation scheitern in der Praxis jedoch häufig. Lesen Sie hier  - basierend auf einer qualitativ-explorativen Expertenstudie - mehr über die wesentlichen Problemfelder der mitarbeiterzentrierten Produktentwicklung und profitieren Sie von konkreten Handlungsempfehlungen aus der Praxis.
Jetzt gratis downloaden!

Bildnachweise