CAJ | 학술논문

随着 DNA 微列阵技术的发展，利用基因表达谱数据进行生物信息的有效挖掘已经成为研究热点。因此，该文中提出将点的代数连通强度与非负矩阵分解相结合的方法对基因表达数据进行分类处理。首先利用点的代数连通强度剔除受外界因素影响过大的基因数据并用修正的特征计分准则进行计分排序，选取具有高计分的基因子集；接着利用近来流行的非负矩阵分解将该基因子集映射到极低维的特征空间；最后利用SVM 分类器实现分类实验。通过几组公开的基因表达谱数据集的实验结果以及与其他方法的对比分析，验证了该方法是有效的、可行的。
수착 DNA 미렬진기술적발전，이용기인표체보수거진행생물신식적유효알굴이경성위연구열점。인차，해문중제출장점적대수련통강도여비부구진분해상결합적방법대기인표체수거진행분류처리。수선이용점적대수련통강도척제수외계인소영향과대적기인수거병용수정적특정계분준칙진행계분배서，선취구유고계분적기인자집；접착이용근래류행적비부구진분해장해기인자집영사도겁저유적특정공간；최후이용SVM 분류기실현분류실험。통과궤조공개적기인표체보수거집적실험결과이급여기타방법적대비분석，험증료해방법시유효적、가행적。
With the rapid development of DNA micro array ,using the existing data processing technology to excavate gene data efficiently has become a hot research in bioinformatics . Therefore ,this paper puts forward a method which combines the algebraic connectivity strength of point with nonnegative matrix factorization to categorize the gene expression data processing .First ,use the algebraic connectivity strength of point to eliminate the useless genes expression data ; then , score and rank the genes expression data based on revised feature score criterion ;next ,use nonnegative matrix factorization to map the gene subset to a feature space of very low dimension ;at last ,categorize the gene expression by using SVM classifier .Some representative groups of gene expression data are used for test ,and the feasibility and effectiveness of this algorithm has been well proved by contrast tests between other methods .