Table 6.

Results of systems described in this paper

System	Official test			Internal evaluation
	F score (%)	Precision (%)	Recall (%)	F score (%)	Precision (%)	Recall (%)
1: Traditional	89.04	89.57	88.52	86.75	86.03	87.49
2: Minimalist	88.71	88.04	89.38	86.85	85.10	88.68
3: Ensemble of 1 and 2	90.11	88.69	88.02	88.02	86.89	89.18
4: Traditional with custom embeddings	89.19	90.05	87.93	86.91	87.10	86.72
5: Minimalist with transfer training	89.32	90.49	88.18	87.18	87.58	87.38
6: Ensemble of 4 and 5	90.33	91.47	89.21	88.17	87.99	88.17

Entries in italics are the best results in that column