摘要 |
Parsing natural language comprising <SL> <LI>a) receiving a tokenised and part-of-speech tagged utterance comprising n tokens <LI>b) for the first token; <SL> <LI>i) calculating a partial parse consisting of one dependency relation by assigning a role and a head for the first token; <LI>ii) calculating the probability of the partial parse from step (i) <LI>iii) repeating steps (b)(i) and (b)(ii) for all heads and roles of the token and storing the A most likely partial parses <LI>c) advancing to the next token and, for each of the A partial parses <LI>iv) calculating a next extension to the partial parse by one dependency relation <LI>v) calculating the probability of the extended partial parse from (c)(i) <LI>vi) repeating steps (c)(i) and (c)(ii) for all possible heads and roles of the token and storing the A most likely resulting partial parses <LI>d) repeating step (c) until all n tokens have been parsed. </SL> </SL> |