IDENTIFYING OPTIMAL NUMBER OF ORTHONORMALISATION IN LEARNING ALGORITHM USING WEATHER FORECASTING

Abstract

 Artificial neural networks are more powerful than any other traditional expert system in the classification of patterns, which are non linear and in performing pattern classification tasks because they learn from examples without explicitly stating the rules. Multilayered feed forward neural networks possess a number of properties, which make them particularly suited to complex problems. Their applications to some real world problems are hampered by the lack of a training algorithm which finds a globally optimal set of weights in a relatively short time. Genetic algorithms are a class of optimization procedures, which are good at exploring a large and complex space in an intelligent way to find values close to the global optimum. In this study, conversion of random weights into orthonormal weights (orthonormalisation) in feed forward network algorithm is proposed and also optimal number of orthonormalisation of weight is evaluated. Orthonormal weight based artificial neural network algorithms not only succeed in its task but also outperforms the genetic algorithm. This success comes from orthonormal weights instead of random and genetic weights. This paper has documented the evolution and ultimate success of this algorithm using weather forecasting data.

Authors and Affiliations

Dr. Heren Chellam G. *

Keywords

Related Articles

 A Performance of a Window Air Conditioner Using Alternative Refrigerants R22 AND R410A

 This paper presents the experimental performance analysis of a window air conditioner using two refrigerants R22 and R410A.The effect of the different parameters of performance analysis (refrigeration capacity, C...

 [b]Glass Fiber Reinforced Concrete & Its Properties [/b]

Glass fiber reinforced concrete (GFRC) is a recent introduction in the field of civil engineering. So, it has been extensively used in many countries since its introduction two decades ago. This product has advantage of...

 Improving Search Strategy of Search Engine Using Probabilistic Latent Semantic

 Users on the internet uses search engine to find information of their interest. However current search engines on web return answer to a query of user independent of user’s requirement for the information. In this...

A SURVEY ON SUPERVISED METHOD FOR DETECTION OF MALWARE

The Number of Android mobile devices has been increased in recent year. There are so many approaches for detection of android malware has been proposed by using permission or source code analysis or dynamic analysis. In...

 A NOVEL APPROACH OF WEAKLY SUPERVISED MODEL FOR COMPARABLE ENTITY MINING

 To compare things which are having same function but differ in their properties is part of human decision making process.Differentiating products having same job but differing in their capabilities or properties i...

Download PDF file
  • EP ID EP138836
  • DOI 10.5281/zenodo.60844
  • Views 85
  • Downloads 0

How To Cite

Dr. Heren Chellam G. * (30).  IDENTIFYING OPTIMAL NUMBER OF ORTHONORMALISATION IN LEARNING ALGORITHM USING WEATHER FORECASTING. International Journal of Engineering Sciences & Research Technology, 5(8), 996-1003. https://europub.co.uk/articles/-A-138836