Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Intelligent systems and applications : extended and selected results from the SAI Intelligent Systems Conference (IntelliSys) 2015
Bi Y., Kapoor S., Bhatia R., Springer International Publishing, New York, NY, 2016. 450 pp. Type: Book (978-3-319333-84-7)
Date Reviewed: Jan 20 2017

This book is a collection of “extended and selected results” from the 2015 SAI Intelligent Systems Conference, held in London. Each of the 21 papers is about 20 pages in length. While the book is not formally divided into sections, the editors have grouped papers on similar topics together. With one exception, the papers describe concrete applications so the flavor is very much applied rather than theoretical.

The first group of papers all deal with prediction in some way:

  • predicting metal loss defects in pipelines based on magnetic flux leakage makes use of a neural network and a support vector machine;
  • predicting financial time series using support vector regression and a back propagating network (the methods are applied to data from the FTSE100, S&P 500, and the Nikkei 225);
  • predicting bladder cancer recurrence, which describes two methods based on association rules;
  • the use of text mining to improve early warnings in construction project management, based on a naive Bayes classifier; and
  • a quantitative assessment of anomaly detection in annotated datasets from the maritime domain comparing two algorithms, one using kernel density estimation and the other a Gaussian mixture model.

Two papers that use fuzzy techniques follow. One uses a fuzzy version of PROMETHEE to select countries for developmental aid based on OECD criteria. The other describes how fuzzy and ontological techniques can be used in the design of self-configuring robots. The system monitors the subsystems to detect anomalous changes that trigger reconfiguration. Continuing with the robot theme, a paper describes a module for the use of permanent magnets for robots that climb reinforced concrete surfaces, where the reinforcement bars provide the material for magnetic adhesion to work.

The next five chapters deal generally with control. The first deals with the control of an ambidextrous robot hand, ambidextrous because the position of the thumb is sufficiently flexible. In the chapter on task allocation for robots that can do both cleaning and delivery, the cleaning is an ongoing process. Next is a chapter on the use of fractional order controllers, meaning that fractional derivatives or integrals are used to determine the control much as one controls by modifying speed. The application is to a pair of interacting conical tanks. A chapter on a context-aware system for control in the Internet of Things (IoT) follows.

Two chapters that deal with human interactions follow. One is on the prediction of pedestrian behavior based on training data from real urban traffic scenes. It makes use of multilevel perceptron networks and a horizon of 2.5 seconds for the predicted movement. The second of these two chapters describes how augmented reality can be used to guide the assembly of a multipart object.

The next two papers make use of stochastic diffusion search (SDS). In the first, SDS is used in the DNA assembly problem, evaluating the performance on several frequently used benchmarks. The second applies SDS to search for symmetries of patterns on toroidal surfaces, in particular those generated by cellular automata.

The remaining chapters cover a fuzzy application for maintaining power quality in a mixed diesel wind generating system. Next is a chapter on improving the speed of high-efficiency video encoding via a method that uses fewer modes than are used in existing systems. There follows a chapter on the use of intelligent communication for load balancing in a wireless sensor network based on a Khalimsky topology that defines a graph for efficient flooding.

A chapter on learning weights in intuitionistic statement networks is the closest to a theoretical rather than applied chapter. The chapter describes how an intuitionistic statement network can be transformed into a conventional network.

In the last chapter, machine learning is applied to sequences of system calls for intrusion detection. A naive Bayes classifier is used that builds conditional probabilities from Markov models of system call sequences.

All of the papers are interesting and each has a comprehensive bibliography. The reader interested in seeing a wide range of applications can dip into this book for examples.

Reviewer:  J. P. E. Hodgson Review #: CR145017 (1704-0209)
Bookmark and Share
  Featured Reviewer  
 
Applications And Expert Systems (I.2.1 )
 
 
Uncertainty, “Fuzzy,” And Probabilistic Reasoning (I.2.3 ... )
 
 
Deduction And Theorem Proving (I.2.3 )
 
 
Robotics (I.2.9 )
 
Would you recommend this review?
yes
no
Other reviews under "Applications And Expert Systems": Date
Institutionalizing expert systems: a handbook for managers
Liebowitz J. (ed), Prentice-Hall, Inc., Upper Saddle River, NJ, 1991. Type: Book (9780134720777)
Nov 1 1991
Verifying and validating personal computer-based expert systems
Bahill A., Prentice-Hall, Inc., Upper Saddle River, NJ, 1991. Type: Book (9780139574573)
Jun 1 1992
Knowledge-based systems: a manager’s perspective
Tuthill G., Levy S., TAB Books, Blue Ridge Summit, PA, 1991. Type: Book (9780830634798)
Dec 1 1991
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy