Learning Linear Precedence rules
Vladimir Borissov Pericliev · 1996
A system is described which learns from examples the Linear Precedence rules in an Immediate Dominance/Linear Precedence grammar. Given a particular Immediate Dominance grammar and hierarchies of feature values potentially relevant for linearization (=the system's bias), the learner generates appropriate natural language expressions to be evanaluted as positive or negative by a teacher, and produces as output Linear Precedence rules which can be directly used by the grammar.