Table 4 Features of MTS’s entities identification
From: Building a Chinese discourse topic corpus with a micro-topic scheme based on theme-rheme theory
| Name | Description |
|---|---|
| Predicate | A content word (lemma) of the predicate of each clause |
| Predicate class | The verb class that the predicate belongs to |
| Head word | String representation of head word of one clause |
| POS of head word | Part of speech of head word |
| Phrase type | Syntactic category of the constituent |
| Path of span | The path from the span to the nominal predicate |
| Position | The positional relationship of the span with the predicate, “left” or “right” |
| Focus word | First word and last word of the focus span |
| Focus span space | Is the focus span adjacent to the predicate? Yes or No |
| IsBrothers | Has the predicate brothers? Yes or No |
| IsRightBrother | Has the predicate right brother? Yes or No |
| Head word of right brother | The headword of the predicate’s nearest right brother |
| POS of right brother | The POS of the predicate’s nearest right brother |
| IVerb | Intervening verb itself |
| IVerb class | The verb class that contains IVerb |
| Path of IVerb | The path from the IVerb to the focus constituent |
| IsFocusSpArg | Is the focus span an argument for IVerb? Yes or No |
| Sematic role of focus | The sematic role of the focus span for IVerb |
| IsHNPArg | Is HNP(Hightest NP headed by the nominal predicate) an argument for IVerb? Yes or No |
| Sematic role of HNP | The semantic role of HNP for IVerb |