Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMenon, M Supriya-
dc.contributor.authorRajarajeswari, Pothuraju-
dc.identifier.issn0975-1084 (Online); 0022-4456 (Print)-
dc.description.abstractData Mining Techniques has attained its momentum in several areas, and its efficient performance in decision support has outperformed and made it a reliable choice. The medical world is one such empirical domain in which a perfect decision at right time would turn out to be a lifesaver. Medical data figures out to be majorly multi-dimensional, where relevant feature extraction is a challenging factor. Several classification approaches like SVM, Decision Trees, and Naive Based are considered to handle these profound challenges. One such challenge discussed in our paper emphasizing on Medical decision support system with Machine Learning (ML) Methodology considering diseases and treatments with their semantic relations in the document of Pub med abstracts. The proposed Multi variant classification framework aims at reducing data into attributes using PCA Transformation infusion with an efficient classification Algorithm - CNB. Our computed results are comparatively successful in attaining ultimate outcomes concerning performance metrics like Accuracy, Precision, Recall, and Time. The strength of our work lies in presenting an efficient approach for elevating enhanced decisions in Health care.en_US
dc.publisherNISCAIR-CSIR, Indiaen_US
dc.sourceJSIR Vol.80(05) [May 2021]en_US
dc.subjectComplement Naïve Bayes (CNB) Algorithmen_US
dc.subjectData mining, Multi variant classificationen_US
dc.subjectPrincipal Component Analysis (PCA)en_US
dc.titleA Novel Approach for Multi Variant Classification of Medical Data in Short Texten_US
Appears in Collections:JSIR Vol.80(05) [May 2021]

Files in This Item:
File Description SizeFormat 
JSIR 80(5) 457-462.pdf483.14 kBAdobe PDFView/Open

Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.