Conference Proceeding

A novel modular neuro-fuzzy controller driven by natural language commands

Fac. of Sci. & Eng., Saga Univ.
02/2001; DOI:10.1109/SICE.2001.977857 ISBN: 0-7803-7306-5 In proceeding of: SICE 2001. Proceedings of the 40th SICE Annual Conference. International Session Papers
Source: IEEE Xplore

ABSTRACT A method of interpreting imprecise natural language commands to
machine understandable manner is presented in this paper. The proposed
method tries to ease the process of man-machine interaction by combining
the theoretical understanding of artificial neural networks and fuzzy
logic. Both fields are very popular to mimic the human behavior in
different research areas in artificial intelligence. The proposed system
tries to understand the natural language command rather than mere
recognition. The distinctive features of the artificial neural networks
in pattern recognition and classification and the abilities of
manipulating imprecise data by fuzzy systems are merged to recognize the
machine sensitive words in the natural language command and then to
interpret them to machine in machine identifiable manner. Modularity of
the design tries to break up the complete task into manageable parts
where the presence of individual part is vital to bridge the so-called
man-machine gap

0 0
 · 
0 Bookmarks
 · 
17 Views
  • Source
    Article: Intelligent steganalytic system: application on natural language environment‖
    Roshidi Din, Azman Samsudin
    [show abstract] [hide abstract]
    ABSTRACT: This paper presents a consolidated view of the computational intelligence used in the natural language steganalysis. In order to understand the human intelligence on natural language, four major computational intelligence methods have been identified. They are bayesian, fuzzy logic, neural network, and genetic algorithm. This paper also presents a measurement tool to measure the natural language intelligent system properties based on steganalysis objectives. It can be learned that the more suitable intelligent systems to be applied in steganalysis domain properties are: neural network, genetic algorithm and fuzzy logic.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.

Keywords

artificial intelligence
 
artificial neural networks
 
complete task
 
different research areas
 
fuzzy systems
 
human behavior
 
individual part
 
interpreting imprecise natural language commands
 
machine identifiable manner
 
machine sensitive words
 
machine understandable manner
 
man-machine interaction
 
manageable parts
 
manipulating imprecise data
 
pattern recognition
 
proposed system
 

K. Pulasinghe