A Treebank-based Investigation of IPP-triggering Verbs in Dutch

More documents

Recommendations

Info

Sen_id: 729 token_id: 6, 7, 8 Source tokens: den, ersten, Gipfel Alignments: 5-5, 6-6, 7-7, Target equivalent ids: 5, 6, 7 Target sentence: Après six heures j’ atteignais le premier sommet, le Combin de la Tsessette. Target equivalent tokens: le premier sommet Checking for n-gram matches for checkpoint instance: 319 Ref: le premier sommet Hypo: après six heures nous atteignons le premier sommet , le combin de la tsessette. Number of 1-grams in reference: 3 # of matching 1-grams = 3 Number of 2-grams in reference: 2 # of matching 2-grams = 2 Number of 3-grams in reference: 1 Matched 3-gram: le premier sommet # of matching 3-grams = 1 All n_gram matches : le ... le premier premier sommet le premier sommet Total n_gram matches: 6 Total n_gram count in reference: 6 Figure 3: Sample output for a specific checkpoint instance There are cases where a single word can correspond to a whole subtree in the other language. For example, the German adjective constituting the adjectival phrase glücklich (EN: happy) is paraphrased in French by the prepositional phrase avec un sentiment de bonheur (EN: with a feeling of happiness). We can thus use the phrase alignment in our treeebank and transform it into word alignments between the constituent words. In this way, our additional alignment level facilitates the extraction of n-m word alignments. In order to demonstrate our claim, we have automatically computed the alignments for the 1000 sentences in the treebank with a well-established tool, GIZA++ [10]. Because the test set is relatively small for a statistical aligner to suggest accurate results, we have appended it to a considerably bigger corpus. For comparison purposes, we have chosen 200000 sentences from the Europarl corpus and, respectively, the same amount of sentences from an Alpine corpus. We have then used the generated alignments as input for the evaluation tool, along with the automatic translations generated by our SMT system (see section 5). Table 1 shows the results for several checkpoints for the language pair German-French. 150
SUJ NP SENT MOD PP NP PP PP NP NP PP CMP_N VN NP Pourtant , même cette dernière partie de notre aventure au Mont Blanc est vécue avec un sentiment de bonheur . ADV PCT_W ADV D_dem A_qual N_C P D_poss N_C P N_C A_qual V V P D_ind N_C P N_C PCT_S KON Doch ADV auch PDAT dieser ADJA letzte NN Teil PPOSAT unserer NN Montblanc-Abenteuer VAFIN wird ADJD glücklich VVPP überstanden $. . HD AP NK NP HD HD AP MO NK NK HD NP AG MO VP HD JU SB S HD OC Figure 4: Aligned French-German tree pair from the Alpine treebank The scores normally indicate the percentage of overlapping n-grams between the reference phrase (checkpoint instance) and the output produced by the MT system. However, in this context, the scores reported for the automatic alignments do not reflect the quality of the MT system. The evaluation module takes the same input in all three cases, except for the alignments, which are computed in different ways and generate different outcomes accordingly. Therefore the scores should be seen as estimates of the accuracy of the evaluation. The more precise the alignments, the more reliable the evaluation results. We notice that the domain of the texts used for training GIZA++ does not influence significantly the accuracy, since the produced scores are similar (e.g. less than 2% difference between Europarl and the Alpine texts). However, when we compare the evaluation results with automatic alignments to the ones obtained with manual alignments, the latter ones are significantly better (up to 12% increase). This finding demonstrates the validity of our claim, namely that feeding manually proofed alignments from a parallel treebank to the evaluation pipeline generates more reliable results. Checkpoint Alignment type Final score GIZA++: Europarl 0.190 29 Verb GIZA++: Alpine 0.191 78 Parallel Treebank 0.283 65 GIZA++: Europarl 0.228 82 Det+Noun+Adj GIZA++: Alpine 0.240 99 Parallel Treebank 0.480 17 Table 1: Evaluation results for different alignments 151
Page 1 and 2:
A Treebank-based Investigation of I
Page 3 and 4:
participle and a (te-)infinitival c
Page 5 and 6:
Some verbs occur twice, since they
Page 7 and 8:
Profiling Feature Selection for Nam
Page 9 and 10:
prepositional objects (FOPP, OPP).
Page 11 and 12:
the limited size of annotated data
Page 13 and 14:
with high precision typically have
Page 15 and 16:
‘This was “not significantly”
Page 17 and 18:
The preposition durch typically has
Page 19 and 20:
Non-Projective Structures in Indian
Page 21 and 22:
the sequential order of nodes in a
Page 23 and 24:
extra-posed relative clause that ge
Page 25 and 26:
Experiments on Dependency Parsing o
Page 27 and 28:
for mitigating the effects of spars
Page 29 and 30:
obtained with MALTParser is 76.61%
Page 31 and 32:
as a standardised serialisation for
Page 33 and 34:
constituency and dependency, possib
Page 35 and 36:
SynAF and/or in . However, they sha
Page 37 and 38:
shows how some elements that are no
Page 39 and 40:
Example
Page 41 and 42:
‏ Example 8: represent
Page 43 and 44:

Page 45 and 46:
Chinese) as the direct object NP.
Page 47 and 48:
Example 15: Tokens and Word Forms
Page 49 and 50:
In a second experiment, a dataset w
Page 51 and 52:

Page 53 and 54:
[3] Leech G. N., Barnett, R. & Kahr
Page 55 and 56:
Effectively long-distance dependenc
Page 57 and 58:
In French, another case of eLDD is
Page 59 and 60:
elativization, it-clefts or questio
Page 61 and 62:
4.2.3 Annotation methodology Becaus
Page 63 and 64:
Number of occurrences in FTB +SEQTB
Page 65 and 66:
producing treebank based LFG approx
Page 67 and 68:
Logical Form Representation for Lin
Page 69 and 70:
gerundives for a total amount of so
Page 71 and 72:
object+indirect object/object one.
Page 73 and 74:
(VP (VB patch) ) ) ) (VP (VBZ is) (
Page 75 and 76:
Types Adverb. Adject. Verbs Nouns T
Page 77 and 78:
Eventually we may comment that ther
Page 79 and 80:
DeepBank: A Dynamically Annotated T
Page 81 and 82:
from another already existing one,
Page 83 and 84:
to parse now does. The extra manual
Page 85 and 86:
In the derivation show in Figure 1,
Page 87 and 88:
4 Patching Coverage Gaps with An Ap
Page 89 and 90:
will enable further improvements in
Page 91 and 92:
ParDeepBank: Multiple Parallel Deep
Page 93 and 94: 2 The ParDeepBank The PTB has emerg
Page 95 and 96: undergone extensive scientific scru
Page 97 and 98: the second combines the structures
Page 99 and 100: Sofia University). Each sentence wa
Page 101 and 102: data, and improvements in the infra
Page 103 and 104: The Effect of Treebank Annotation G
Page 105 and 106: ut without feature structures. This
Page 107 and 108: only the lexicon is fine-grained to
Page 109 and 110: Automatic Coreference Annotation in
Page 111 and 112: manually annotated Czech sentences.
Page 113 and 114: citizens of Bilbao] are very involv
Page 115 and 116: 4.1.3 Coreference Selector Module T
Page 117 and 118: Nominal P R F1 MUC 75.33% 81.33% 78
Page 119 and 120: [9] G. Doddington, A. Mitchell, M.
Page 121 and 122: Analyzing the Most Common Errors in
Page 123 and 124: Graph 1 shows results of subsequent
Page 125 and 126: in the semantic type). This situati
Page 127 and 128: Will a Parser Overtake Achilles? Fi
Page 129 and 130: annotation is also based on a depen
Page 131 and 132: Set Name Sentences Tokens % Train/T
Page 133 and 134: dency relations (PRED, OBJ, SBJ, AD
Page 135 and 136: ton and Stacklazy), we trained a mo
Page 137 and 138: Feature Function Column Name Addres
Page 139 and 140: Using Parallel Treebanks for Machin
Page 141 and 142: phrases are generated by a shallow
Page 143: Figure 2: A Kybot p
Page 147 and 148: Checkpoint Instances Google PT Our
Page 149 and 150: 6 Conclusions In this paper we have
Page 151 and 152: An integrated web-based treebank an
Page 153 and 154: 16] and we have earlier reported fr
Page 155 and 156: Figure 2: Screenshot of the interfa
Page 157 and 158: ple subcategorization frames may be
Page 159 and 160: 4 Integrated interface for annotati
Page 161 and 162: of the 5th International Conference
Page 163 and 164: Translational divergences and their
Page 165 and 166: posed so far, Cyrus’ work did not
Page 167 and 168: allowed an appropriate coverage of
Page 169 and 170: damentales [...] 10 (the universal
Page 171 and 172: for all lex_pair(s,t) do if head an
Page 173 and 174: the treatment of typical translatio
Page 175 and 176: Building a treebank of noisy user-g
Page 177 and 178: Phenomenon Attested example Std. co
Page 179 and 180: Figure 1: French Social Media Bank
Page 181 and 182: Impact of treebank characteristics
Page 183 and 184: Det Adj N (a) Danish Det Adj N (b)
Page 185 and 186: sv: Bestämmelserna i detta avtal f
Page 187 and 188: 100 90 80 Unlabelled attachment 70
Page 189 and 190: vidner , tilhørere og tiltalte wit
Page 191 and 192: mance is not inconsiderable, as was
Page 193 and 194: Genitives in Hindi Treebank: An Att
Page 195 and 196:
A genitive can occur with complex p
Page 197 and 198:
quite high. This motivates us towar
show all

A Treebank-based Investigation of IPP-triggering Verbs in Dutch

Create successful ePaper yourself

Delete template?

Save as template?