Machine Learning Approach for Answer Detection in Discussion Forums: An Application of Big Data Analytics

Table 4

Top 15 features selected by the univariate technique for Ubuntu dataset.

Code	Abbreviation
F1	ThrdCntrodRplyWMDistance
F2	ThrdCentrodRplyCosnSmlrty
F3	TtlRplyCosnSmlrtyWholCrps
F5	TtlRplyCosnSmlrty
F6	QustionRplyCosnSmlrty
F7	UnqWrds
F8	IsRplyByCrtrOfInitlPost
F10	NoThrdsUsrParticipated
F11	ReWrdsOvrlpInitialPost
F13	IsRplyContan5WHWrds
F15	IsRplyHvHyperlnk
F16	“WMDbtwnTitlRpl”
F17	“WMDbtwnQustionRpl”
F19	TotlNoIntialPstsByUser
F20	NoWrdsRply