Skip to main content

Table 2 Comparison with standard outlier detection and unsupervised clustering methods

From: Gene expression data classification using topology and machine learning models

Dataset

Decision tree

Naive Bayes

Local outlier factor

DBSCAN

Topo-curated cohort

Local outlier factor

DBSCAN

Topo-curated cohort

Droso breeding

32.14

–

79.34

57.14

–

42.24

Droso parasitod

36.84

100

83.50

47.36

66.67

–

Mouse E Coli

30.91

21.05

89.29

40.00

21.05

59.21

Mouse prion

35.89

39.53

61.62

39.74

39.53

57.84

Mouse liver cancer

46.67

56.21

72.95

60.00

56.52

72.32

Human bowel disease

17.74

19.41

51.09

48.39

53.40

–

  1. Bold values indicate the highest accuracy obtained for different classifiers and datasets under consideration