期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2012
卷号：3
期号：2
页码：600-603
语种：English
出版社：Ayushmaan Technologies
摘要：One of the foremost tasks of biology is to understand the relationships between the various genes and the protein, they encode. Understanding the protein structures is vital to determine the function of proteins. The information about its confirmation can provide essential information for ‘Drug Design’,’ Protein Engineering’, ‘disease diagnoses, ‘Breast Cancer Prognosis’. The approaches, like Artificial Neural Network in soft-computing and Statistical methods like Support Vector Machines are used for protein structure prediction. Artificial Neural Network is only capable of capturing vectorial data and for nonlinear classification its architecture is very complex and time-consuming. In contrast to it, Support Vector Machine is a linear classifier which generates a representation of non-linear mapping from protein sequence to protein fold space. SVM is characterized by fast training, so computationally efficient. This work is an investigation of protein secondary structure prediction problem by traditional learning techniques such as Artificial Neural Network where Back propagation algorithm is used for learning. It measures the efficiency and accuracy of the machine learning methods through Mean Square Error.
