| D15-1203 | initialized . We then transform the | relative distances | into real valued vectors by looking |
| N04-1023 | up two samples , we compute the | relative distance | between these two samples in |
| C02-1002 | string similarity scoring and | relative distance | . The similarity measure we used |
| A00-2038 | a similar procedure to compute | relative distance | between words from various Dutch |
| D15-1203 | defined as the combination of the | relative distances | from the current word to e1 and |
| C00-2156 | the selective morpheme tag and | relative distance | to the other phrase breaks . |
| C02-1002 | to the pairs with the smallest | relative distance | between the constituent tokens |
| D09-1137 | occurrence is computed as the | relative distance | of the first occurrence of the |
| E06-1043 | since we are concerned with the | relative distance | of several posterior distributions |
| D14-1178 | -- with a high probability -- | relative distances | between vectors are approximately |
| D08-1108 | same document . In general , the | relative distance | between an informal phrase and |
| E06-1019 | so that , among mod - ifiers , | relative distance | from head is maintained . More |
| D14-1037 | EL ( oi ) are filtered by the | relative distance | of nj to oi as given in the NL |
| D15-1110 | text seg - ments , we extract | relative distance | between the segments and their |
| J11-2001 | task should directly reflect the | relative distance | of words on an SO ( semantic |
| J96-1005 | processing considerations ( e.g. , the | relative distance | of potential antecedents to anaphors |
| C00-2156 | morl ) heme tag se.lectively and | relative distance | to the other phrase breaks . |
| E09-1077 | adjectives in WordNet by measuring | relative distance | of the term from exemplars , |
| C04-1011 | and our goal is to compare the | relative distances | between Gp and Mp for di erent |
| J93-1007 | around w. For each of the possible | relative distances | from w , we analyze the distribution |
