<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-0501">
  <Title>Parsing Akkadian Verbs with Prolog</Title>
  <Section position="6" start_page="6" end_page="6" type="concl">
    <SectionTitle>
5 Conclusion and future work
</SectionTitle>
    <Paragraph position="0"> The idea behind this research was not to simulate the evolution of Akkadian, either historically as is done in comparative Semitic studies, or to use some sort of theoretical, two-stage morphological rules as in [Kataja, 1988]. Rather the goal was to create a system which could be used by researchers and students to help in understanding Akkadian texts. The complexities and many weak variants of the verb stems prevent the two-stage rules from ever yielding a useful parser, and there are many unneeded complexities in the historical evolution. This system as it evolved is, perhaps, needlessly intricate, but it works quickly and well and adding new verb stems can be done in about 8 hours of work.</Paragraph>
    <Paragraph position="1"> Work continues on the parser, in teaching Assyriologists to use it in their daily work, in checking for errors in the rules, and in adding more verbal forms. Currently the effort is directed to add the $-stem, the last of the simple stems, to the parser and teach members of the Department of Near Eastern and Judaic Studies at Brandeis University to allow them to make use of the program. Future plans begin with work to add wildcards to the generation part of the web interface, to allow the generation of, for instance, all G preterite forms. This would, among other thing, aid in testing the accuracy of the generated forms, and by extension, the parser itself. They also include adding some of the more common derived stems, from each of the four simple stems there are three derived stems, a 't', an 'n' and a 'tn', for example, the G yields the Gt, Gn and Gtn. Suffix generation needs to be added to the Web interface and along with that some assimilation rules should be added. In the language, if a finite form ending with a semi-weak is followed by a suffix beginning with a Shin, the weak assimilates, for example 'm$' becomes '$$'.</Paragraph>
  </Section>
class="xml-element"></Paper>