Skip to main content
main-content

Über dieses Buch

This book highlights recent research advances in unsupervised learning using natural computing techniques such as artificial neural networks, evolutionary algorithms, swarm intelligence, artificial immune systems, artificial life, quantum computing, DNA computing, and others. The book also includes information on the use of natural computing techniques for unsupervised learning tasks. It features several trending topics, such as big data scalability, wireless network analysis, engineering optimization, social media, and complex network analytics. It shows how these applications have triggered a number of new natural computing techniques to improve the performance of unsupervised learning methods. With this book, the readers can easily capture new advances in this area with systematic understanding of the scope in depth. Readers can rapidly explore new methods and new applications at the junction between natural computing and unsupervised learning.

Includes advances on unsupervised learning using natural computing techniques

Reports on topics in emerging areas such as evolutionary multi-objective unsupervised learning

Features natural computing techniques such as evolutionary multi-objective algorithms and many-objective swarm intelligence algorithms

Inhaltsverzeichnis

Frontmatter

Advances in Natural Computing

Frontmatter

Chapter 1. Detailed Modeling of CSC-STATCOM with Optimized PSO Based Controller

Today, different types of FACTS devices have been used in the complex transmission power network with many applications to solve the system stability problems. Among all FACTS devices, static synchronous compensators (STATCOMs) are the most important shunt FACTS devices for the vital voltage support and efficient reactive power compensation because of its attractive quick response and wide operating characteristics range. Therefore, in this chapter, first the system description and modeling of three-phase current source converter (CSC)-based STATCOM is presented. Nonlinearity of this modeling is removed with the help of power balance equation and dq-transformation. After that pole-shifting and LQR-based controllers are designed for the CSC-STATCOM. But the best state-feedback gain matrices for different controllers are obtained laboriously through trial and error method, although time-consuming. So this problem is solved with the help of particle swarm optimization (PSO)-based AI method to search the best values of state-feedback gain matrices in a very short time. In this chapter, comparisons among different controller-based CSC-STATCOM devices are also presented with simulation results. The feasibility of the proposed optimized controller-based CSC-STATCOM scheme is demonstrated through simulation in MATLAB. Finally, this chapter shows the detailed modeling of optimized novel controller-based CSC-STATCOM with better simulation outcomes. Design of DC-link reactor for CSC-STATCOM is also discussed in this chapter.
Sandeep Gupta

Chapter 2. A Brief Review and Comparative Study of Nature-Inspired Optimization Algorithms Applied to Power System Control

This work deals with the use of a special class of optimization algorithms called nature-inspired optimization algorithms (NIOA) to improve power system control actions. This work discusses also the optimization issue of the control task in power system. As an example of nature-inspired (NI) algorithm, various swarm intelligence (SI) and bio-inspired (BI) algorithms that mimic the social, living, and hunting behavior of many kinds of animal, insects, and creatures in nature such as wolves, elephants, whale, fishes, spider, bees, ants, bats, and birds were used as an optimization tool. The main aim was to enhance frequency and voltage regulation loops to cope with system fluctuations during disturbances. The purpose was to optimize the Power System Stabilizer (PSS) parameters and the PID controller gains for enhancing both load frequency control (LFC) and automatic voltage regulator (AVR) systems. To satisfy the objective of this work, a series of simulations on single-area power system with standard LFC and AVR loops was performed. To show the contribution of each applied method, a comparative study in view of peak overshoot, peak undershoot, and settling time was carried out.
Nour E. L. Yakine Kouba, Mohamed Boudour

Chapter 3. Self-Organization: A Perspective on Applications in the Internet of Things

A technical overview of self-organization for enabling efficient machine-to-machine (M2M) communication networks for the Internet of things (IoT) is presented. M2M networks consists of intelligent machines often including sensors, actuators, and computing nodes that enable automated sensing, data gathering and processing, and information exchange. Self-organization mechanisms are needed to minimize human intervention and enable distributed intelligence with local information exchange. Thus, self-organizing M2M networks act as building blocks for the IoT. In this chapter, the aim is to give an overview of IoT and important M2M communication network use-cases which motivate the application of self-organized networking approaches and to highlight related challenges and opportunities. We give an introduction to self-organizing systems and discuss self-organization approaches in contemporary cellular networks and its potential for future networks especially M2M communication networks for IoT. Considering the diverse requirements of M2M use-cases in terms of scalability, latency, cost, overhead, and reliability, the focus is on a fundamental question: How can M2M networks in IoT benefit from self-organized networking paradigms? To this end, we discuss M2M networks that involve heterogeneous devices communicating directly to other devices in proximity, without centralized control by network infrastructure. The use-cases discussed here include home networking, vehicle-to-vehicle/roadside networks, and public safety networks. In home M2M networking scenarios based on low-power short-range communication technologies, self-organizing mechanisms can enable autonomous and seamless connectivity between heterogeneous devices. In contrast, vehicular networks such as vehicle-to-vehicle require self-organization due to mission-critical aspects such as stringent latency requirements and changing topology. Likewise, public safety networks are also mission-critical and require not only rapid deployment and resiliency but also self-healing capabilities that are the cornerstone of self-organizing networks. We conclude that self-organization is an effective paradigm for rapid, resource-efficient, and low-cost deployments of heterogeneous M2M networks, which makes it an important enabler for IoT.
Furqan Ahmed

Advances in Unsupervised Learning

Frontmatter

Chapter 4. Applications of Unsupervised Techniques for Clustering of Audio Data

Audio data mining is an important area of research in the field of computing technology. These techniques can be used for automatically analyzing and searching/indexing the content of an audio data. The growing amount of digital contents in the world of computers has increased the challenges in managing and retrieving the audio data [8]. The advancements in technology in this area of audio data mining have made the scientists to develop more beneficial applications helpful to society [6]. Content-based music retrieval is one of the significant applications of audio data mining, which makes use of the content such as melody, instrument, rhythms, etc. for efficient management of music repository [17]. The study of audio data mining includes the study of various fields like statistics, pattern recognition, neural networks, artificial intelligence, signal processing, machine learning, and data mining. Machine learning is a part of artificial intelligence, and to be intelligent, a system that is in a changing environment should have the ability to learn. If the system can learn and adapt to such changes, the system designer need not foresee and provide solution for all possible situations [5]. Classification and clustering are the major techniques used under audio data mining.
Surendra Shetty, Sarika Hegde

Chapter 5. Feature Extraction and Classification in Brain-Computer Interfacing: Future Research Issues and Challenges

Brain-computer interfacing (BCI) is a communication bridge between human brain and computer. BCI system consisted of four sections (signal acquisition, signal processing, feature extraction and classifications, application Interface). In this survey paper, we try to elaborate the entire structure of BCI process especially emphasizing on feature extraction and classification area. We have briefly described different types of brain signals and their properties. For the stationary type of signal, we have used autoregressor and Fourier transform, and for nonstationary signal, we have used wavelet transformation as feature extraction policy. There have been various techniques introduced for EEG signal classification in the literature from low-cost methods (LDA, logistic regression, KNN) to computationally expensive techniques (SVM, artificial neural networks). We have also discussed ensemble and complex classifiers. In this paper, we have explained the basic concepts of all the classifiers and describe their key properties and applications. We have thoroughly analyzed all possible types of comparisons between classifiers using statistical plotting (bar chart, line chart) so that future researchers can identify the suitable classifier for a specific task. Finally, this paper deals with the various open challenges and future research issues with respect to feature extraction and classification in BCI system.
Debashis Das Chakladar, Sanjay Chakraborty

Chapter 6. A Multilingual Framework for a Question Answering System for the Support of Ticket Data: An Extended Approach

This chapter describes about a multilingual framework for semantic matcher where the task is to obtain the top K semantically matched results for a user query. The proposed framework makes use of word embeddings of different languages under consideration, with a help of the pre-trained models to detect the word senses of search queries and show the top best matches which belong to the same class. Word-level to context-level semantics are achieved through the trained semantic model with word embeddings. This semantic matching approach is evaluated against Troubleshooting queries on Tickets data of various languages such as English, Spanish, French, etc.
Suyog Trivedi, Balaji Jagan, Gopichand Agnihotram

Natural Computing for Unsupervised Learning

Frontmatter

Chapter 7. Chemical Reaction-Based Optimization Algorithm for Solving Clustering Problems

Heuristic algorithms are widely used in the diverse fields of engineering and sciences and prove its efficiency over classical algorithms. In the analysis of chemical process, it is observed that the formation of new product consists of a proficient computational procedure among chemical reactions. These chemical reactions consist of objects, events, states, and process. In this work, an efficient and robust algorithm, called artificial chemical reaction optimization algorithm, is adopted for solving the partitional clustering problems. The performance of the proposed algorithm is investigated on well-known clustering datasets. Further, the simulation results of the CRO-based clustering algorithm are compared with some state-of-the-art clustering algorithms. It is seen that proposed clustering algorithm provides better performance than other algorithms in terms of intra-cluster distance and f-measure.
Yugal Kumar, Neeraj Dahiya, Sanjay Malik, Geeta Yadav, Vijendra Singh

Chapter 8. A Novel Artificial Bee Colony Algorithm for Robust Permutation Flowshop Scheduling

Because of the importance of the permutation flowshop scheduling problem, a variety of researches are being focused on this issue and numerous algorithms have been proposed. However, many uncertainties may exist in real production environment of the permutation flowshop problem. In order to solve this obstacle, an optimization criterion of maximizing the probability of ensuring the makespan not surpass the expected finish time is considered for M-machine permutation flowshop in this chapter. A schedule for this optimization criterion is called the robust schedule. A novel artificial bee colony (ABC) algorithm integrated with an efficient local search is proposed. The local search introduces into a probability model to determine whether the new generated solution should be accepted. For experiment analysis, the performance of proposed ABC is evaluated on the well-known Car and Rec permutation flowshop problems which are taken from OR library and is compared with an improved genetic algorithm and NEH heuristic. The comparison results indicate that the proposed ABC performs well and can give better robust schedules for M-machine permutation flowshop problem.
Shijing Ma, Yunhe Wang, Mingjie Li

Chapter 9. Semantic Image Retrieval Using Point-Set Topology and the Ant Sleeping Model

Due to its widespread practical applications to image database management, semantic image retrieval has received a lot of attention in past decades. In particular, relevance feedback-based methods have been a popular approach toward bridging the gap between low-level features and high-level semantic concepts. Nature-inspired algorithms have shown a lot of potential to solve problems which are complex and require discovering of patterns under changing environment. However, lack of sound mathematical foundations has been considered a drawback toward better analysis of these algorithms. In this chapter, we propose a novel general topological model for semantic image retrieval using relevance feedback. We use point-set topology to develop mathematical constructs for modeling the semantic retrieval. In particular, we develop an image retrieval algorithm based on the ant sleeping model and extend the topological model to analyze it. Through experiments we show that our algorithm performs well in indexing an image database for relevance feedback. With our indexing procedure, the average response time to access image results from a storage device is lower when compared to vector quantization techniques. We also evaluate our algorithm, theoretically and empirically, against PicSOM (a CBIR system based on relevance feedback). Our ASM-based technique shows a very efficient retrieval performance using relevance feedback.
Deepak Karunakaran, Shrisha Rao

Chapter 10. Filter-Based Feature Selection Methods Using Hill Climbing Approach

Feature selection remains one of the most important steps for usability of a model for both supervised and unsupervised classification. For a dataset, with n features, the number of possible feature subsets is 2n. Even for a moderate size of n, there is a combinatorial explosion in the search space. Feature selection is a NP-hard problem; hence finding the optimal solution is not feasible. Typically various kinds of intelligent and metaheuristic search techniques can be employed for this purpose. Hill climbing is arguably the simplest of such techniques. It has many variants based on (a) trade-off between greediness and randomness, (b) direction of the search, and (c) size of the neighborhood. Consequently it might not be trivial for the practitioner to choose a suitable method for the task in hand. In this paper, we have attempted to address this issue in the context of feature selection. The descriptions of the methods are followed by an extensive empirical study over 20 publicly available datasets. Finally a comparison has been done with genetic algorithm, which shows the effectiveness of hill climbing methods in the context of feature selection.
Saptarsi Goswami, Sanjay Chakraborty, Priyanka Guha, Arunabha Tarafdar, Aman Kedia

Others

Frontmatter

Chapter 11. Social Networking Awareness in Indian Higher Education

Nowadays, few people can imagine life without the Internet. The utilization of the Internet has altered the way in which people communicate and share activities. The Internet is now being utilized in almost every field of human activity, including health, education, business and commerce, to name just a few. Currently, besides the Internet users started to use various tools from Web 2.0 and 3.0 in government, organizations and higher education to connect, communicate, cooperate and collaborate. This study will examine the social networking (SN) awareness in Indian higher education. An online survey was developed based on the current literature and distributed to more than 142 respondents from 100 completed the whole survey. The study outcomes confirmed that using social networking in the Indian higher education produced positive awareness of developing personal skills and communicate and accessible service. On the other hand, using SN created negative awareness in Indian higher education from negative emotions created, decreased productivity in social activities, laziness and depression issues, lack of communication and IP and decrease in study skills. Future research directions are suggested, which can consolidate the findings of this study by means of various surveys and increased sample sizes. Moreover, recommendations are made for improvement in SN research work by means of training and spreading awareness through seminars and other information avenues in the Indian higher education sector.
Shamdeep Brar, Tomayess Issa, Sulaiman Ghazi B. Alqahtani

Backmatter

Weitere Informationen

BranchenIndex Online

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

Whitepaper

- ANZEIGE -

Globales Erdungssystem in urbanen Kabelnetzen

Bedingt durch die Altersstruktur vieler Kabelverteilnetze mit der damit verbundenen verminderten Isolationsfestigkeit oder durch fortschreitenden Kabelausbau ist es immer häufiger erforderlich, anstelle der Resonanz-Sternpunktserdung alternative Konzepte für die Sternpunktsbehandlung umzusetzen. Die damit verbundenen Fehlerortungskonzepte bzw. die Erhöhung der Restströme im Erdschlussfall führen jedoch aufgrund der hohen Fehlerströme zu neuen Anforderungen an die Erdungs- und Fehlerstromrückleitungs-Systeme. Lesen Sie hier über die Auswirkung von leitfähigen Strukturen auf die Stromaufteilung sowie die Potentialverhältnisse in urbanen Kabelnetzen bei stromstarken Erdschlüssen. Jetzt gratis downloaden!

Bildnachweise