期刊名称:International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN:2347-6710
电子版ISSN:2319-8753
出版年度:2015
卷号:4
期号:7
页码:5155
DOI:10.15680/IJIRSET.2015.0407016
出版社:S&S Publications
摘要:In this project we present Trinity Tree Algorithm comparison with Back Propagation Algorithm.Among these the trinity tree algorithm is an unsupervised data extraction and Backpropagation algorithm is asupervised data extraction. Data mining is a growing topic of interest in latest Engineering subject as it has help in theresearch area to extract important information from raw data. Data mining deals with supervised data as well asunsupervised data. Our first method Backpropagation algorithm is being used through artificial neural networks whichrequire a dataset of the desired output for many inputs, making up the training set. It is most useful for feed-forwardnetworks (networks that have no feedback, or simply, that have no connections that loop). Our second method isTrinity is a unsupervised proposal to generate patterns and to separate them by prefix, suffix and separator in order togenerate a regular expression. Aim of our proposed technique is to improve the quality of time in performance. Theaccuracy of result to be much more and with less error. The graph gives the total clustering of two methods.
关键词:Data mining; Backpropagation algorithm; KDD cup dataset; Artificial neural network; SOM maps