File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/04/c04-1114_concl.xml
Size: 965 bytes
Last Modified: 2025-10-06 13:54:00
<?xml version="1.0" standalone="yes"?> <Paper uid="C04-1114"> <Title>Improving Statistical Machine Translation in the Medical Domain using the Unified Medical Language System</Title> <Section position="6" start_page="0" end_page="0" type="concl"> <SectionTitle> 5 Conclusion </SectionTitle> <Paragraph position="0"> We carried out four different experiments in order to improve a Spanish-English medical domain translation system. After sequentially applying different ideas the final system shows an 11% improvement in BLEU and 6% improvement in NIST score.</Paragraph> <Paragraph position="1"> Table 4 compares the different experiments and scores (the 500k dictionary refers to the dictionary that was first extracted from the UMLS with With more investigation and the ongoing effort of the National Library of Medicine to extent the UMLS databases it will hopefully be possible to further improve the translation performance.</Paragraph> </Section> class="xml-element"></Paper>