File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/05/w05-1008_concl.xml
Size: 1,326 bytes
Last Modified: 2025-10-06 13:55:03
<?xml version="1.0" standalone="yes"?> <Paper uid="W05-1008"> <Title>Bootstrapping Deep Lexical Resources: Resources for Courses</Title> <Section position="8" start_page="75" end_page="75" type="concl"> <SectionTitle> 7 Conclusion </SectionTitle> <Paragraph position="0"> We have proposed three basic paradigms for deep lexical acquisition, based on morphological, syntactic and ontological language resources, and demonstrated the effectiveness of each strategy at learning lexical items for the lexicon of a precision English grammar. We discovered surprising variation in the results for the different DLA methods, with each learning method performing particularly well for at least one basic word class, but the best overall methods being syntax- and ontology-based DLA.</Paragraph> <Paragraph position="1"> The results presented in this paper are based on one particular language (English) and a very specific style of DLR (a precision grammar, namely the English Resource Grammar), so some caution must be exercised in extrapolating the results too liberally over new languages/DLA tasks. In future research, we are interested in carrying out experiments over other languages and alternate DLRs to determine how well these results generalise and formulate alternate strategies for DLA.</Paragraph> </Section> class="xml-element"></Paper>