alphabetic character subset

Keyword(s):  
Author(s):  
BRENT FERGUSON ◽  
RANADHIR GHOSH ◽  
JOHN YEARWOOD

This paper reports on an experimental approach to find a modularized artificial neural network solution for the UCI letters recognition problem. Our experiments have been carried out in two parts. We investigate directed task decomposition using expert knowledge and clustering approaches to find the subtasks for the modules of the network. We next investigate processes to combine the modules effectively in a single decision process. After having found suitable modules through task decomposition we have found through further experimentation that when the modules are combined with decision tree supervision, their functional error is reduced significantly to improve their combination through the decision process that has been implemented as a small multilayered perceptron. The experiments conclude with a modularized neural network design for this classification problem that has increased learning and generalization characteristics. The test results for this network are markedly better than a single or stand alone network that has a fully connected topology.


Author(s):  
Oleg Dmitrochenko ◽  
Marko Matikainen ◽  
Aki Mikkola

A single-line symbolic notation is proposed for description of an arbitrary mechanism or a multibody system. The kinematics is represented by a sequence of elementary transformations; each of those being marked by a reserved alphabetic character. Force and constraint links between the bodies are also defined by reserved characters. The parameters of the system, such as identifiers of degrees of freedom, inertia parameters and others, are assigned default names if not specified. However, user-defined names, parameters and functions can be placed instead, if needed. The proposed description in its shortest form is suitable for academic purpose to identify only the essential properties of a multibody system. In an extended form, by explicit mentioning names of variables and parameters and other data like initial conditions, this description can serve as input data for multibody analysis software. Examples from academic area and technical applications are given to show the applicability of the proposed description.


Author(s):  
A. T. M. Fazle Rabbi Mojumder ◽  
Abu Zafar Md. Imran ◽  
Kallol Biswas ◽  
Muhammad Sifatul Alam Chowdhury ◽  
Md. Shakowat Zaman Sarkar

2015 ◽  
Vol 18 (2) ◽  
pp. 208-227 ◽  
Author(s):  
Mark Sebba

This paper discusses three processes relating to the social meaning of scripts and orthographies, all of which are potentially mediated by the role of script-as-image. One of these processes, iconisation, was introduced to the field by Irvine and Gal (2000) and is widely known. Attribution is a process which precedes iconisation, whereby a group of people associate a linguistic feature or language-related practice with a group of people who (supposedly) use that feature or engage in that practice. Orthographic branding involves a specific visual/graphical element of written language such as an alphabetic character. Through ‘branding,’ this element becomes an emblem of a group of people who use the element in question in their writing practices. Branding may involve iconisation, but the processes are distinct. This paper describes and distinguishes the three processes and provides examples from different languages and user communities.


Sign in / Sign up

Export Citation Format

Share Document