Results 1 - 10 of 51,418

Table 3. E ect of vocabulary size (seconds)

in 2-D-S tree: An index structure for content-based retrieval of images
by Youping Niu, M. Tamer Özsu, Xiaobo Li
"... In PAGE 8: ....2. Vocabulary Sizes In the second set of experiments, we xed the values of D (= 5000) and L (= 10) and varied S from 20 to 500. Table3 shows that the e ect of the number of di erent symbols, on both the sequential matching and the 2-D-S-tree, is not very signi cant. The reasons that the performance of the 2-D-S-tree improves slightly as the vocabulary size grows are as follows.... ..."

Table 3. Vocabulary size of each test set.

in Using Partial Morphological Analysis In Language Modeling Estimation For Large Vocabulary Portuguese Speech Recognition
by Ciro Martins, João P. Neto, Luís B. Almeida
"... In PAGE 3: ...resulting from the texts used as training material for the language models (TRAIN). On Table3 we can compare the size of the vocabulary of each test set before and after the application of the decomposition method. As we can verify the reduction on the vo- cabulary size increases for larger vocabularies.... In PAGE 3: ... Vocabulary size of each test set. To evaluate the e ect of the decomposition on the number of OOVs we took all the training texts of the language models (which yields a vocabulary of 139,758 words as we can see from Table3 ) and esti- mated the percentage of words not covered by the vocabulary sets of 5K and 20K. From Table 4 we see that this percentage gets lower with the decom- posed vocabulary.... ..."

Table 6 Language Model Vocabulary Size and Out of Vocabulary Ratio

in Language Model Based Arabic Word Segmentation
by Young-suk Lee, Kishore Papineni, Salim Roukos, Ossama Emam, Hany Hassan 2003
"... In PAGE 6: ... Word error rate reduction due to the unsupervised stem acquisition is 38% for the segmenter developed from the 10K word manually segmented corpus and 32% for the segmenter developed from 110K word manually segmented corpus. Language model vocabulary size (LM VOC Size) and the unknown stem ratio (OOV ratio) of various segmenters is given in Table6 . For unsupervised stem acquisition, we have set the frequency threshold at 10 for every 10-15 million word corpus, i.... ..."
Cited by 15

Table 6 Language Model Vocabulary Size and Out of Vocabulary Ratio

in Language Model Based Arabic Word Segmentation
by Young-suk Lee, Kishore Papineni, Salim Roukos, Ossama Emam, Hany Hassan 2003
"... In PAGE 6: ... Word error rate reduction due to the unsupervised stem acquisition is 38% for the segmenter developed from the 10K word manually segmented corpus and 32% for the segmenter developed from 110K word manually segmented corpus. Language model vocabulary size (LM VOC Size) and the unknown stem ratio (OOV ratio) of various segmenters is given in Table6 . For unsupervised stem acquisition, we have set the frequency threshold at 10 for every 10-15 million word corpus, i.... ..."
Cited by 15

Table 1: Vocabulary sizes for two Turkish corpora.

in Statistical Morphological Disambiguation for Agglutinative Languages
by Dilek Z. Hakkani-Tür, Dilek Z. Hakkani-tfir, Gökhan Tür, Kemal Oflazer, Gskhan Tiir 2000
Cited by 1

Table 2: Vocabulary sizes and number of trig-

in Applying Extrasentential Context To Maximum Entropy Based
by Tagging With Large, Ezra Black, Andrew Finch, Ruigiang Zhang

Table 5.1: Feature Vocabulary Sizes.

in INVISTOR - A Distributed MultiMedia Indexing System
by Mike Westmacott

Table 1: Transcription quality vs vocabulary size

in Effects of Out of Vocabulary Words in Spoken Document Retrieval
by P. C. Woodland , S. E. Johnson, P. Jourlin, K. Spärck Jones

Table 4. F score for 11 datasets for each of the vocabulary size. Vocabulary Sizes

in Incremental hierarchical clustering of text documents
by Nachiketa Sahoo 2006
"... In PAGE 24: ... Normal) We compared the clustering results using both F score and H score (computed as described in Expressions (34) and (37) respectively). The scores at different vocabulary sizes are given in the Table4 and Table 58. Vocabulary Sizes F Data k1a k1b re0 re1 tr11 tr12 tr23 tr31 tr41 tr45 wap Winner in 50 K N 0.... ..."
Cited by 2

Table 1: Result of bench mark test (V. size: Vocabulary size)

in Voice-Activated Home Banking System And Its Field Trial
by Toshihiro Isobe Masatoshi, Masatoshi Morishima, Fuminori Yoshitani, Nobuo Koizumi
Results 1 - 10 of 51,418