CiteSeerX — Table search results

Results 1 - 10 of 31,146

Table 1: The distribution of word forms

in BottomUp Tagset Design from Maximally Reduced Tagset

by Peter Dienes, Csaba Oravecz

"... In PAGE 2: ... The initial assumption is that this large number of word forms contain all possible ambiguity classes that can occur in the language. Table1 presents some basic statis- tics on the range of word form variation found... ..."

Table 1: The distribution of word forms

in Principled hidden tagset design for tiered tagging of Hungarian

by Péter Dienes, Csaba Oravecz, Tamás Váradi 2000

"... In PAGE 2: ....1. The morphological analysis and morphosyntactic descriptions (MSD) The language resource of our analysis consisted of the whole current stock of the Hungarian National Corpus (ap- proximating 80m words) compiled into a word frequency list as input to the morphological analysis. Table1 presents some basic statistics on the range of word form variation... In PAGE 5: ...00% 2.51% Table1 0: Error rate with the tagsets tagger. The increase of error rate in this case might be attributed the lack of contextual information which could have been provided by features already missing from the R tagset.... ..."

Cited by 5

Table 1: The distribution of word forms

in Principled Hidden Tagset Design for Tiered Tagging of Hungarian

by Dan Tufis, Peter Dienes, Csaba Oravecz, Tamas Vradi

Table 1: The distribution of word forms

in Principled Hidden Tagset Design for Tiered Tagging of Hungarian

by Dan Tufis Pter, Dan Tufis, Pter Dienes, Csaba Oravecz, Tams Vradi

Table 3: Word forms with several lemmata.

in A Freely Available Morphological Analyzer, Disambiguator and Context Sensitive Lemmatizer for German

by Wolfgang Lezius, Manfred Wettler

"... In PAGE 4: ...0%) had an unambiguous lemma. Of the remaining 695 word forms, 667 had two possible lemmata and 28 were threefold ambi- guous ( Table3 gives some examples). Using the large tag set, 616 out of the 695 ambiguous word forms were correctly lemmatized (88.... ..."

Table 5: Relation: word forms to the length of the text

in DESAM - Approaches to Desambiguation

by Fi Mu, Pavel Rychly, Pavel Smrz, Karel Pala, Karel Pala

"... In PAGE 7: ... The second graph displays the beginning of the displayed relation. Similarly, Table5 demonstrates the relation the length of the corpus and its coverage by the different word forms and the second graph offers again the closer look at the coverage of DESAM texts by the word forms. It is obvious that both presented graphs tell us something about the inflectional nature of Czech.... ..."