Skip to main content

Table 2 List of the features engineered from the complex networks in this study

From: CNFE-SE: a novel approach combining complex network-based feature engineering and stacked ensemble to predict the success of intrauterine insemination and ranking the features

j = 1, 2, 3 where j = 1 indicates the index of the complex network (CN) made up of all training instances. J = 2 (or 3) are indices of complex networks consisting of all training instances excluding data records belonging to negative (or positive) class

 

F1 = (node degree in CN2 – node degree in CN3) / node degree in CN1

(4)

F2 = (node weighted degree in CN2 – node weighted degree in CN3) / node weighted degree in CN1

(5)

F3 = (node closeness centrality in CN2 – node closeness centrality in CN3) / node closeness centrality in CN1

(6)

F4 = (node Eigen value centrality in CN2 – node Eigen value centrality in CN3) / node Eigen value centrality in CN1

(7)

F5 = (node betweenness centrality in CN2 – node betweenness centrality in CN3) / node betweenness centrality in CN1

(8)

F6 = (node clustering coefficient in CN2 – node clustering coefficient in CN3)/ node clustering coefficient in CN1

(9)

F7 = minimum length of the shortest path from the node in CN2 / minimum length of the shortest path from the node in CN3

(10)

F8 = the number of 2-hop neighbors of the node in CN2 / the number of 2-hop neighbors of the node in CN3

(11)

F9 = node degree in CN2 / node degree in CN3

(12)

F10 = node closeness centrality in CN2 / node closeness centrality in CN3

(13)

F11 = node Eigen value centrality in CN2 / node Eigen value centrality in CN3

(14)

F12 = node betweenness centrality in CN2 / node betweenness centrality in CN3

(15)

F13 = (normalized node degree in CN2—normalized node degree in CN3)/ max (normalized node degree in CN2, normalized node degree in CN3)

(16)

F14 = (normalized node closeness in CN2 – normalized node closeness in CN3)/ max (normalized node closeness in CN2, normalized node closeness in CN3)

(17)

F15 = (normalized node Eigen value in CN2 – normalized node Eigen value in CN3)/ max (normalized node Eigen value in CN2, normalized node Eigen value in CN3)

(18)

F16 = (normalized node betweenness in CN2 – normalized node betweenness in CN3)/ max (normalized node betweenness in CN2, normalized node betweenness in CN3)

(19)

F17 = (normalized node clustering coefficient in CN2 – normalized node clustering coefficient in CN3)/ max (normalized node clustering coefficient in CN2, normalized node clustering coefficient in CN3)

(20)

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us