Optimal Cascade Linguistic Attribute Hierarchies for Information Propagation

He, Hongmei; Lawry, Jonathan
May 2009
IAENG International Journal of Computer Science;May2009, Vol. 36 Issue 2, p1
Academic Journal
A hierarchical approach, in which a highdimensional model is decomposed into series of low-dimensional sub-models connected in cascade, has been shown to be an effective way to overcome the 'curse of dimensionality' problem. The upwards propagation of information through a cascade hierarchy of Linguistic Decision Trees (LDTs) based on label semantics forms a process of cascade decision making. In order to examine how a cascade hierarchy of LDTs works compared with a single LDT for multiple attribute decision making, we developed genetic algorithm with linguistic ID3 in wrapper to find optimal cascade hierarchies. Experiments have been carried out on the two benchmark databases, Pima Diabetes and Wisconsin Breast Cancer databases from the UCI Machine Learning Repository. It is shown that an optimal cascade hierarchy of LDTs has better performance than a single LDT. The use of attribute hierarchies also greatly reduces the number of rules when the relationship between a goal variable and input attributes is highly uncertain and nonlinear. Moreover, the cascade linguistic attribute hierarchy presents cascade transparent linguistic rules, which will be useful for analyzing the effect of different attributes on the decision making as a reference in a special application.


Related Articles

  • Finite Automata Approach to Computing All Seeds of Strings with the Smallest Hamming Distance. Guth, Ondřej; Melichar, Bořivoj // IAENG International Journal of Computer Science;May2009, Vol. 36 Issue 2, p1 

    Seed is a type of a regularity of strings. A restricted approximate seed w of string T is a factor of T such that w covers a superstring of T under some distance rule. In this paper, the problem of all restricted seeds with the smallest Hamming distance is studied and a polynomial time and space...

  • Using Genetic Programming to Evolve a Team of Data Classifiers. Morrison, Gregor A.; Searson, Dominic P.; Willis, Mark J. // World Academy of Science, Engineering & Technology;Feb2011, Issue 50, p261 

    No abstract available.

  • Enabling multi-level relevance feedback on PubMed by integrating rank learning into DBMS. Hwanjo Yu; Taehoon Kim; Jinoh Oh; Ilhwan Ko; Sungchul Kim; Wook-Shin Han // BMC Bioinformatics;2010 Supplement 2, Vol. 11, p1 

    Background: Finding relevant articles from PubMed is challenging because it is hard to express the user's specific intention in the given query interface, and a keyword query typically retrieves a large number of results. Researchers have applied machine learning techniques to find relevant...

  • Combining Univariate and Multivariate Bottom-up Discretization. Yu Sang; Keqiu Li // Journal of Multiple-Valued Logic & Soft Computing;2013, Vol. 20 Issue 1/2, p161 

    Most inductive learning methods require that the training data set contains only discrete attributes, which makes it necessary to discretize continuous numeric attributes. Current efforts mainly focus on discretizing data for individual attributes, without taking into account the correlations...

  • Integration of Metaheuristics: A Way to Improve Search Performance. Phen Chiak See; Kuan Yew Wong; Komarudin // AIP Conference Proceedings;8/13/2009, Vol. 1148 Issue 1, p868 

    It is generally stated that hybridizing metaheuristics helps to achieve a better search performance when solving combinatorial optimization problems such as Quadratic Assignment Problems (QAPs). In this paper, the integration of two metaheuristics (Max-Min Ant System and Genetic Algorithm) is...

  • Heart Diseases Diagnosis via Improved Classifier System (XCSLA). Panahi, Masoud Shariat; Yazdani, Navid Moshtaghi; seqerloo, arezoo yazdani // Iranian Journal of Medical Informatics;2014, Vol. 3 Issue 2, p1 

    Today, with advances in use of intelligent systems technology for data analysis, processing in the diagnosis of various diseases is of particular importance. In this paper, an intelligent method that combines xcs and LA are offered to diagnose heart disease. Classification system (XCS) which is...

  • Learn the Nuances of Dealing With Imperfect Data. Koopmann, James F. // Database Trends & Applications;Dec2004, Vol. 18 Issue 12, p22 

    This column explains that through the use of mathematical models, Netrics Solutions learned the nuances of error-laden data and provides effective data matching and record lookup that transforms unreliable data into true information, increasing the integrity and quality of databases. Netrics...

  • A Comparison of three evolutionary strategies for multiobjective genetic programming. Zhang, Yang; Rockett, Peter // Artificial Intelligence Review;Mar2007, Vol. 27 Issue 2/3, p149 

    We report what we believe to be the first comparative study of multi-objective genetic programming (GP) algorithms on benchmark symbolic regression and machine learning problems. We compare the Strength Pareto Evolutionary Algorithm (SPEA2), the Non-dominated Sorting Genetic Algorithm (NSGA-II)...

  • Average Concept of Crossover Operator in Real Coded Genetic Algorithm. Rosshairy Abd Rahman; Razamin Ramli // International Proceedings of Economics Development & Research;2013, Vol. 63, p73 

    As the most important search operator in a Genetic Algorithm (GA) approach, many procedures have been proposed to accomplish the idea of a crossover. As a result, knowledge in crossover has incorporated special features such as statistical elements (i.e. arithmetic crossover) and natural...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics