Docstoc

Call for Chapters

Document Sample
Call for Chapters Powered By Docstoc
					                             Call for Chapters
           “Diagnostic Test Approaches to Machine Learning and
                  Commonsense Reasoning Systems”
                       to be published by IGI Global (www.igi-global.com)
                                 scheduled for release in 2012

Editors:
    Xenia A. Naidenova, Senior Researcher at Military Medical Academy, Russian Federation, Saint-
Petersburg
    Viktor L. Shagalov, Programmer at Speech Modules Ltd., Israel, Rehovot



Introduction, Aims & Topics
Introduction

Research on supervised symbolic machine learning is developing in many directions which are
sometimes unconnected with each other. Discussions in the literature, for example, tend to deal with
creating algorithms for solving separate, well-formulated practical problems.

A more complex approach which considers symbolic supervised machine learning methods as a class
of multifunctional inductive-deductive reasoning is important and long overdue. The processes of
symbolic machine learning are like the processes of plausible reasoning, in which the synthesis of the
logical operations—such as induction, the deduction, abduction, reasoning from the contrary,
hypotheses generation and refutation—are achieved. The interconnection between learning algorithms
and the results of diagnostic tasks based on acquired knowledge presents many problems which need
to be addressed. These problems include approaches to analyzing the results from the users’
perspective and reconstructing learning algorithms in order to improve the subsequent pattern
recognition results.

The consideration of symbolic machine learning algorithms as an entire class will make it possible, in
the future, to generate algorithms (with the aid of some parameters) depending on the initial users’
requirements and the quality of solving targeted problems in domain applications.

Aims

The objectives of this book are:
    1. To survey, analyze, and compare the existing and most effective algorithms for mining all
kinds of logical rules, including Apriori-like bottom-up search, Formal Concept Analysis, closure
operations of Galois connections, and Diagnostic Test Approach and some others.
    2. To show how these approaches use shared concepts for mining logical rules, including item,
item set, transaction, frequent itemset, maximal itemset, generator (non-redundant or irredundant
itemset), closed itemset, support, and confidence.
    3. To consider all these approaches based on the same mathematical language (the lattice theory)
and to analyze all the techniques in constructing rule mining algorithms.
    4. To describe the genesis of rule mining algorithms and to give the classification of these
algorithms with respect to 1) various kinds of branching ordering methods, 2) various kinds of
pruning methods, 3) various techniques (inductive methods) of algorithm constructing (for example,
simultaneously exploring both the itemset space and transaction space, decomposing the main task in
the sub-problems using item- or itemset- projection, item or itemset skipping techniques, itemset-tid-
set tree search space, and FP-tree search space).
     5. To focus on using language technologies in learning by trying to identify what
questions and problems are solved, but also show how well the algorithms developed assist in
the provision of support and the construction of feedback for learning.

Audience

The primary goal of the book is to bring together experts in the related fields in order to:

       Share knowledge (i.e., approaches, models, issues, solutions) acquired in the domain
        of using supervised symbolic machine learning methods;
       Present applications in different domains;
       Debate on directions and possibilities of future research in the domain;
       Create a forum for further collaboration and develop an international community on
        this field of study.

Topics
    Chapters should cover three directions in the development of symbolic supervised machine
learning methods:
    1. Theoretical problems in constructing supervised symbolic machine learning algorithms
        elaborated in the framework of formal model of conceptual reasoning, based on an algebraic
        lattice;
    2. Presentation and description of integrated multifunctional machine learning systems; and
    3. Applying the described machine learning systems and algorithms for solving some practical
        and rather complex diagnostic tasks.

Suggested topics include, but are not limited to, the following:

The Logical or Symbolic Methods of Machine Learning: Current Trends

       Machine Learning and Extracting Conceptual Knowledge from Data
       Supervised Symbolic Learning as Inferring Logical Dependencies from Data
       Unsupervised Symbolic Machine Learning (Conceptual Clustering) as Inferring Natural
        Classification from Data
       The Interconnection Supervised with Unsupervised Symbolic Learning
       Ontology Application to Knowledge Discovery Process Based on Integrating Supervised and
        Unsupervised Conceptual Machine Learning
       The synthesis of cognitive procedures and the problem of induction
       Learning Logic Formulas
       Logic formulas based knowledge discovery
       Feature Selection for Data Mining
       Classification Algorithm for filtering Information of the Web Pages in Search Engines
       Data Mining and Knowledge Discovery Via Logic-Based Methods

An Analytical Survey of Current Approaches to Mining Logical Rules from Data

       The Main Current Approaches to Mining Logical Rules from Data
      Apriori-Like Approach to Inferring Functional, Implicative Dependencies, and Association
       Rules
      Formal Concept Analysis Approach to Inferring Functional, Implicative Dependencies, and
       Association Rules
      JSM Method of Machine Learning (automated hypotheses generation)
      Galois Connections and Closure Operations for Inferring Functional, Implicative
       Dependencies, and Association Rules
      Diagnostic Test Approach to Inferring Functional, Implicative Dependencies, and Association
       Rules
      An approach to Machine Leaning Based on Inferring Boolean Functions from Positive and
       Negative Examples
      Integration of Mining All Types of Logical Rules Based on Operations of Data-Knowledge
       Lattice Construction

Effective Current Algorithms and Techniques for Symbolic Machine Learning

      Incremental Algorithms for All Kind of Frequent Itemsets Construction
      Constructing Galois Lattices as a Commonsense (plausible) Reasoning Process
      Incremental Construction of Concept Lattices
      Plausible reasoning in the systems of JSM type
      The genesis of rule mining algorithms and the classification of these algorithms with
       respect to 1) various kinds of branching ordering methods, 2) various kinds of pruning
       methods, 3) various techniques (inductive methods) of algorithm constructing, for
       example, simultaneously exploring both the itemset space and transaction space,
       decomposing the main task in the sub-problems using item- or itemset- projection,
       item or itemset skipping techniques, itemset-tid-set tree search space, and FP-tree
       search space and some others methods
      Fast Heuristics for Inferring a Boolean Function from Examples
      An Incremental Learning Algorithm for Inferring Boolean Functions

Symbolic Machine Learning Program Systems

      Machine Learning as an Instrument for Intelligent System Organization
      Diagnostic Test Machine (DTM)
      Program Realization of the systems of JSM type
      Mechanisms of Data and Knowledge Management in Machine Leaning Systems

Application of Symbolic Machine Learning Methods

      Concept Lattices and Their Applications
      Application of DTM in Psycho-Diagnostics
      Application of DTM for Diagnosing Schizophrenia and Mania Disorders Based on Patient
       Texts Analysis
      Some Realistic Applications of Inferring a Boolean Function from Examples (Breast
       Cancer Diagnosis, Predicting Muscle Fatigue from EMG Signals)
      Intelligent System of JSM Type for Analyzing Clinical Data in Oncology
      Machine learning Methods for Analyzing Medical Data
      Intelligent Analysis of Data in Sociology
      Logic Classification and Feature Selection for Biomedical Data
      Logic Based Methods for SNPs Tagging and Reconstruction
      Machine Learning in Web Search Engine
        Web Mining in Thematic Search Engine
        The Analysis of Service Quality via Stated Preferences and Rule-Based Classification

Learning and Reasoning

        Ontology Environment for Knowledge Acquisition
        Incorporating Machine Learning in a Plausible Reasoning Environment for Realizing
         Conceptual Communication of Domain Experts with Intelligent Machine Learning Systems
        Feedback from Quality of Decision Making Results to Machine Learning Algorithm
         Construction
        The Building of High-Performance Web search engines Capable to Adapt to the Evolution of
         Users' Information Needs, Expectations and Behavior
        Natural Language Interface to Symbolic Machine Learning Algorithms


Chapter Format Requirements and Publication Details
Submission Types

Submitted chapters should describe substantial and unpublished work and should be
submitted in English. Submissions are expected to have either 7,000-9,000 words for short
papers (presenting very new and not so formalized work) or 17,000-19,000 words for long
chapters, (presenting theoretical or completed applied work).

Chapter Submission Process

Researchers and practitioners are invited to submit on or before November 28, 2010, a 2-3 page
chapter proposal clearly explaining the mission and concerns of his or her proposed chapter.

Authors of accepted proposals will be notified about the status of their proposals and sent chapter
guidelines. Full chapters are to be submitted by February 6, 2011.

All submitted chapters will be reviewed on a double-blind review basis. Contributors may also be
requested to serve as reviewers for this project.


Important Dates
        Submission Deadline : November 28, 2010
        Notification of acceptance: December 16, 2010
        Full chapter due: February 6, 2011
        Publication-ready papers due date: May 31, 2011

    .............

Editorial Advisory Board:
Giovanni Felici, Istituto Dianalisi dei Sistemi ed Informatica “Antonio Ruberti”, Consiglio Nazionale
Delle Ricerche, Rome, Italy

Evangelos Triantaphyllou, Department of Computer Science, Louisiana State University, USA
Sergei O. Kusnetsov, Department of Applied Mathematics, State University Higher School of
Economics, Russian Federation, Moscow

Viktor K. Finn, Department of Theoretical and Applied Problems of the Information Theory, Russian
Academy of Science, Moscow

Sergei A. Obiedkov, Department of Applied Mathematics, State University Higher School of
Economics, Moscow, Russian Federation

Omar Larouk, National School of Advanced Studies in Information Science and Libraries, France

Publisher

This book is scheduled to be published by IGI Global (formerly Idea Group Inc.), publisher
of the “Information Science Reference” (formerly Idea Group Reference) imprint. For
additional information regarding the publisher, please visit www.igi-global.com. This
publication is scheduled for release in early 2012.

Inquiries and submissions can be forwarded electronically (Word document):

Xenia A. Naidenova, naidenovaxen@gmail.com
Viktor L. Shagalov, shagalovv@gmail.com

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:6
posted:12/23/2010
language:English
pages:5