<body>
<p>MaltParser has the possibility of distinguishing between different kinds of null-values when extracting the feature
vector. For input columns like <code>POSTAG</code> it is possible to differentiate two null-values:</p>
<ul>
<li><code>NO NODE</code>: There exists no corresponding dependency graph node (e.g., because the lookahead extend beyond the end of the string),
which means that the feature is really undefined.
<li><code>ROOT NODE</code>: The dependency graph node is a root node, which means that it is not possible to extract an input column value
(for example, the word form or the part-of-speech).
</ul>
<p>In addition to the two null value categories for input columns, there is one more for the output columns:</p>
<ul>
<li><code>NO VALUE</code>: The dependency graph node exists and is not the root, but has not yet been assigned a value for the output column
requested (e.g., has not been assigned a head and therefore does not have a dependency type).
</ul>
With the option <code>--singlemalt-null_value</code> it is possible to specify the degree of differentiation of null-values.
<ul>
<li><code>none</code>: Excludes all kinds of null-values when extracting the feature vector, this option value is not possible for learning
methods that have symbolic feature vector encoding.
<li><code>one</code>: Maps all kinds of null values to one symbol.
<li><code>rootlabel</code>: Maps all kinds of null values to one symbol and for output columns this symbol is the same as the root
label (used to emulate MaltParser 0.4)
<li><code>rootnode</code>: Distinguishes between <code>NO NODE</code> and <code>ROOT NODE</code>, and the <code>NO VALUE</code> null-value case is mapped
to the <code>ROOT NODE</code> null-value for output columns.
<li><code>novalue</code>: Distinguishes between <code>NO NODE</code> and <code>ROOT NODE</code> for both input and output columns, and <code>NO VALUE</code> for output columns.
</ul>
</body>