,
,
,
,
,
,
,
,
,
these rules predict the whole inflection Their accuracy on unseen data is 6443%: ? (LAST-TWO-TONES = L3H) and (LAST-PHONEME-TYPE = C) => FULL-CHANGE-CLASS=L3H-LHH + e (18 ,
,
,
,
,
,
,
,
,
,
,
,
,
,
, Thus, even with so few rules, it is possible to correctly predict the object form for a good portion of nouns
, References
Fast Effective Rule Induction, Proceedings of the Twelfth International Conference on Machine Learning, pp.115-123, 1995. ,
DOI : 10.1016/B978-1-55860-377-6.50023-2
Incremental Reduced Error Pruning, International Conference on Machine Learning, pp.70-77, 1994. ,
DOI : 10.1016/B978-1-55860-335-6.50017-9
Induction of decision trees, Machine Learning, vol.1, issue.1, pp.81-106, 1986. ,
DOI : 10.1037/13135-000
Generating production rules from decision trees Cite- seer, Proceedings of the Tenth International Joint Conference on Artificial Intelligence, pp.304-307, 1987. ,
C4.5: Programs for machine learning, 1993. ,
Do you speak Kordofanian? 'Fifty Years After Independence: Sudan's Quest for Peace, Stability and Identity, 2006. ,
Benefactive and malefactive verb extensions in the Koalib verb system. Benefactives and malefactives. Typological perspectives and case studies, pp.295-315, 2010. ,
URL : https://hal.archives-ouvertes.fr/hal-01162007
The phonology of Koalib, 2010. ,
URL : https://hal.archives-ouvertes.fr/halshs-00584041
Case in Koalib (a Kordofanian language) and related Heibanian languages ,
A survey of Kordofanian: The Heiban group, 1981. ,
Kordofanian. The Niger-Congo Languages, p.66, 1989. ,
A survey of the phonetics and grammatical structure of the Nuba Mountain languages, with particular reference to Otoro, Katcha and Nyima, 1956. ,
Data mining, ACM SIGMOD Record, vol.31, issue.1, 2011. ,
DOI : 10.1145/507338.507355