Analysis of miRNA expression profile based on SVM algorithm

2018 2nd International Workshop on Renewable Energy and Development

能源学;经济学

Ting-Ting, Dai^1 ; Chang-Ji, Shan^2 ; Yan-Shou, Dong^3 ; Yi-Duo, Bian^3

School of Mathematics and Statistics, Zhaotong University, Yunnan

657000, China^1

School of Physics and Information Engineering, Zhaotong University, Yunnan

657000, China^2

School of Foreign Languages, Zhaotong University, Yunnan

657000, China^3

关键词: Classification ability;    Classification accuracy;    Data mining algorithm;    K-nearest neighbors;    k-NN algorithm;    miRNA expressions;    Recognition accuracy;    Unified measurements;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/153/3/032009/pdf
DOI  :  10.1088/1755-1315/153/3/032009

来源: IOP

【 摘 要 】

Based on mirna expression spectrum data set, a new data mining algorithm - tSVM - KNN (t statistic with support vector machine - k nearest neighbor) is proposed. the idea of the algorithm is: firstly, the feature selection of the data set is carried out by the unified measurement method; Secondly, SVM - KNN algorithm, which combines support vector machine (SVM) and k - nearest neighbor (k - nearest neighbor) is used as classifier. Simulation results show that SVM - KNN algorithm has better classification ability than SVM and KNN alone. Tsvm - KNN algorithm only needs 5 mirnas to obtain 96.08 % classification accuracy in terms of the number of mirna » tags» and recognition accuracy. compared with similar algorithms, tsvm - KNN algorithm has obvious advantages.