City Block Distance for Identification of Co-expressed MicroRNAs
Citations
39 citations
Cites methods from "City Block Distance for Identificat..."
...In order to include fuzziness, the normalized city-block distance was employed, adopting the probability function Px(U) as a dimension to compare dissimilarities between the two sample sets (M1,M2) (Webb and Copsey, 2003; Paul and Maji, 2014):...
[...]
...In order to include fuzziness, the normalized city-block distance was employed, adopting the probability function Px(U) as a dimension to compare dissimilarities between the two sample sets (M1,M2) (Webb and Copsey, 2003; Paul and Maji, 2014): dNCB = 1 N × N∑ x=1...
[...]
23 citations
Cites methods from "City Block Distance for Identificat..."
...For example microRNA datasets used in [10] or real life gene expression datasets used in [11] are some unlabeled datasets....
[...]
15 citations
10 citations
Additional excerpts
...But most of them utilize either supervised or semi-supervised classification ( An & Doerge, 2012; Saha et al., 2016; Wang & Pan, 2014 ) techniques. In cancer diagnosis, these classification methodologies help in classifying tumor samples as benign or malignant or any other sub types ( Alizadeh et al., 20 0 0; de Souto et al., 20 08; Yeung & Bumgarner, 20 03 ). But in many cases, it may not be possible to have labeled tissue samples. For example microRNA datasets used in ( Paul & Maji, 2013 ) or real life gene expression datasets used in ( Saha, Ekbal, Gupta, & Bandyopadhyay, 2013 ) are some unlabeled datasets. Because of the unavailability of labeled data, it is difficult to apply any supervised classification technique to solve this problem. Thus unsupervised classification techniques become popular in solving this problem. In recent years the use of multi-objective optimization (MOO) ( Saha et al., 2016 ) becomes popular in solving the cancer tissue sample classification problem. Several objective functions related to partitioning the cancer tissues are simultaneously optimized using some MOO-based techniques. In Horng et al. (2009) , the authors have developed a supervised system that selects a small group of gene markers for classification by using all the necessary information on well-defined pathways available from KEGG. They have used C4.5 decision tree for generating the classification model. In Alonso-González, Moro-Sancho, Simon-Hurtado, and Varelarrabal (2012) , the authors propose, relaxing the maximum accuracy criterion, to select the combination of attribute selection and classification algorithm that using less attributes has an accuracy not statistically significantly worst that the best....
[...]
...But most of them utilize either supervised or semi-supervised classification ( An & Doerge, 2012; Saha et al., 2016; Wang & Pan, 2014 ) techniques. In cancer diagnosis, these classification methodologies help in classifying tumor samples as benign or malignant or any other sub types ( Alizadeh et al., 20 0 0; de Souto et al., 20 08; Yeung & Bumgarner, 20 03 ). But in many cases, it may not be possible to have labeled tissue samples. For example microRNA datasets used in ( Paul & Maji, 2013 ) or real life gene expression datasets used in ( Saha, Ekbal, Gupta, & Bandyopadhyay, 2013 ) are some unlabeled datasets. Because of the unavailability of labeled data, it is difficult to apply any supervised classification technique to solve this problem. Thus unsupervised classification techniques become popular in solving this problem. In recent years the use of multi-objective optimization (MOO) ( Saha et al., 2016 ) becomes popular in solving the cancer tissue sample classification problem. Several objective functions related to partitioning the cancer tissues are simultaneously optimized using some MOO-based techniques. In Horng et al. (2009) , the authors have developed a supervised system that selects a small group of gene markers for classification by using all the necessary information on well-defined pathways available from KEGG....
[...]
4 citations
References
[...]
52,705 citations
15,662 citations
14,144 citations
9,470 citations
7,826 citations