Deep Learning- and Word Embedding-Based Heterogeneous Classifier Ensembles for Text Classification

Table 4

The list of classification accuracies of single classifiers and heterogeneous ensemble systems with majority voting and stacking integration strategies on datasets represented with the average of Word2vec vectors.

Methods	20News 18828	20News 19997	Mini-news	Web KB4	1150haber	Milliyet	Hurriyet	Aahaber	Computation time
MVNB	74.23	61.01	76.18	79.77	87.27	81.12	77.58	83.11	56 m
MNB	89.50	80.44	81.67	87.05	96.13	81.55	82.39	84.36	1 h 17 m
SVM	88.51	62.78	90.35	91.71	93.89	89.27	82.00	82.23	1 h 50 m
RF	87.34	76.65	89.27	90.06	97.56	90.30	84.45	89.78	2 h 44 m
CNN	90.17	76.91	92.55	92.22	95.14	76.10	88.07	93.26	4 h 55 m
Heter-MV	89.62	78.20	91.45	93.45	95.81	90.93	87.18	92.78	6 h 30 m
Heter-Stck	90.85	80.94	93.78	94.56	96.34	92.05	88.96	93.80	7 h 23 m