Table 8.
Sentence detection.
| Top 10 | 4 | w 2 | 5 | w 2 | 6 | w 2 | 7 | w 2 |
|---|---|---|---|---|---|---|---|---|
| 1 | LT border b2 | 60.82 | Die | 121.98 | St.p. | 415.20 | Abbr | 0.54 |
| 2 | LT border b1 | 60.82 | für | 121.03 | Amb. | 410.82 | - | - |
| 3 | LT border b3 | 60.82 | TE | 94.84 | ca. | 402.62 | - | - |
| 4 | LT | 1.98 | Keine | 94.09 | Pat. | 401.16 | - | - |
| 5 | Mean-LT | 1.98 | Sono | 83.67 | max. | 397.93 | - | - |
| 6 | > b2 | 0.13 | Der | 80.47 | Z.n. | 392.47 | - | - |
| 7 | > b3 | 0.04 | CT | 77.13 | st.p. | 390.62 | - | - |
| 8 | > b1 | 2.27E-4 | E-Nr | 75.40 | n. | 378.70 | - | - |
| 9 | - | - | Im | 71.92 | St. | 377.24 | - | - |
| 10 | - | - | Am | 66.45 | bzw. | 368.27 | - | - |
Top 10 feature rankings per feature set (4 Word length features; 5 Right context word type features; 6 Word type features; 7 Abbreviation feature). Length (LT); w2: Weight based feature relevance criterion.