Table 6.

Performance of classification models using a combination of lexical and different types of non-lexical features according to standard metrics for the task of annotating adolescent interview session transcripts.

Cls.	Model	Acc.	Prec.	Rec.	F1	Kappa
17	CRF	0.682	0.673	0.682	0.677	0.636
	SVM	0.708	0.705	0.708	0.680	0.663
	SVM-PL	0.715	0.711	0.715	0.696	0.673
	SVM-LIWC	0.742	0.740	0.742	0.727	0.704
	SVM-AF	0.751	0.750	0.751	0.739	0.715
20	CRF	0.581	0.579	0.581	0.580	0.540
	SVM	0.610	0.611	0.610	0.592	0.571
	SVM-PL	0.639	0.642	0.639	0.630	0.604
	SVM-LIWC	0.653	0.653	0.653	0.657	0.619
	SVM-AF	0.682	0.685	0.682	0.674	0.651
41	CRF	0.493	0.485	0.493	0.457	0.502
	SVM	0.537	0.513	0.537	0.504	0.502
	SVM-PL	0.565	0.543	0.565	0.542	0.535
	SVM-LIWC	0.538	0.518	0.538	0.507	0.503
	SVM-AF	0.568	0.549	0.568	0.546	0.538