Data
----
We have collected English-Odia parallel and monolingual data from the
available public websites for NLP research in Odia.
The parallel corpus consists of English-Odia parallel Bible, Odia
digital library, and Odisha Goverment websites. It covers bible,
literature, goverment of Odisha and its policies. We have processed the
raw data collected from the websites, performed alignments (a mix of
manual and automatic alignments) and release the corpus in a form ready
for various NLP tasks.
The Odia monolingual data consists of Odia-Wikipedia and Odia e-magazine
websites. Because the major portion of data is extracted from
Odia-Wikipedia, it covers all kinds of domains. The e-magazines data
mostly cover the literature domain. We have preprocessed the monolingual
data including de-duplication, text normalization, and sentence
segmentation to make it ready for various NLP tasks.
Corpus Formats
--------------
Both corpora are in simple tab-delimited plain text files.
The parallel corpus files have three columns:
- the original book/source of the sentence pair
- the English sentence
- the corresponding Odia sentence
The monolingual corpus has a varying number of columns:
- each line corresponds to one *paragraph* (or related unit) of the
original source
- each tab-delimited unit corresponds to one *sentence* in the paragraph
Data Statistics
----------------
The statistics of the current release is given below.
Parallel Corpus Statistics
---------------------------
Dataset Sentences #English tokens #Odia tokens
------- --------- ---------------- -------------
Train 27136 706567 604147
Dev 948 21912 19513
Test 1262 28488 24365
------- --------- ---------------- -------------
Total 29346 756967 648025
Domain Level Statistics
------------------------
Domain Sentences #English tokens #Odia tokens
------------------ --------- ---------------- -------------
Bible 29069 756861 640157
Literature 424 7977 6611
Goverment policies 204 1411 1257
------------------ --------- ---------------- -------------
Total 29697 766249 648025
Monolingual Corpus Statistics
-----------------------------
Paragraphs Sentences #Odia tokens
---------- --------- ------------
71698 221546 2641308
Domain Level Statistics
-----------------------
Domain Paragraphs Sentences #Odia tokens
-------------- -------------- --------- -------------
General (wiki) 30468 (42.49%) 102085 1320367
Literature 41230 (57.50%) 119461 1320941
-------------- -------------- --------- -------------
Total 71698 221546 2641308
Citation
--------
If you use this corpus, please cite it directly (see above), but please cite also the following paper:
Title: OdiEnCorp: Odia-English and Odia-Only Corpus for Machine Translation
Author: Shantipriya Parida, Ondrej Bojar, and Satya Ranjan Dash
Proceedings of the Third International Conference on Smart Computing & Informatics (SCI) 2018
Series: Smart Innovation, Systems and Technologies (SIST)
Publisher: Springer Singapore
Data
-----
We have collected English-Odia parallel data for the purposes of NLP
research of the Odia language.
The data for the parallel corpus was extracted from existing parallel
corpora such as OdiEnCorp 1.0 and PMIndia, and books which contain both
English and Odia text such as grammar and bilingual literature books. We
also included parallel text from multiple public websites such as Odia
Wikipedia, Odia digital library, and Odisha Government websites.
The parallel corpus covers many domains: the Bible, other literature,
Wiki data relating to many topics, Government policies, and general
conversation. We have processed the raw data collected from the books,
websites, performed sentence alignments (a mix of manual and automatic
alignments) and released the corpus in a form suitable for various NLP
tasks.
Corpus Format
-------------
OdiEnCorp 2.0 is stored in simple tab-delimited plain text files, each
with three tab-delimited columns:
- a coarse indication of the domain
- the English sentence
- the corresponding Odia sentence
The corpus is shuffled at the level of sentence pairs.
The coarse domains are:
books ... prose text
dict ... dictionaries and phrasebooks
govt ... partially formal text
odiencorp10 ... OdiEnCorp 1.0 (mix of domains)
pmindia ... PMIndia (the original corpus)
wikipedia ... sentences and phrases from Wikipedia
Data Statistics
---------------
The statistics of the current release are given below.
Note that the statistics differ from those reported in the paper due to
deduplication at the level of sentence pairs. The deduplication was
performed within each of the dev set, test set and training set and
taking the coarse domain indication into account. It is still possible
that the same sentence pair appears more than once within the same set
(dev/test/train) if it came from different domains, and it is also
possible that a sentence pair appears in several sets (dev/test/train).
Parallel Corpus Statistics
--------------------------
Dev Dev Dev Test Test Test Train Train Train
Sents # EN # OD Sents # EN # OD Sents # EN # OD
books 3523 42011 36723 3895 52808 45383 3129 40461 35300
dict 3342 14580 13838 3437 14807 14110 5900 21591 20246
govt - - - - - - 761 15227 13132
odiencorp10 947 21905 19509 1259 28473 24350 26963 704114 602005
pmindia 3836 70282 61099 3836 68695 59876 30687 551657 486636
wikipedia 1896 9388 9385 1917 21381 20951 1930 7087 7122
Total 13544 158166 140554 14344 186164 164670 69370 1340137 1164441
"Sents" are the counts of the sentence pairs in the given set (dev/test/train)
and domain (books/dict/...).
"# EN" and "# OD" are approximate counts of words (simply space-delimited,
without tokenization) in English and Odia
The total number of sentence pairs (lines) is 13544+14344+69370=97258. Ignoring
the set and domain and deduplicating again, this number drops to 94857.
Citation
--------
If you use this corpus, please cite the following paper:
@inproceedings{parida2020odiencorp,
title={OdiEnCorp 2.0: Odia-English Parallel Corpus for Machine Translation},
author={Parida, Shantipriya and Dash, Satya Ranjan and Bojar, Ond{\v{r}}ej and Motlicek, Petr and Pattnaik, Priyanka and Mallick, Debasish Kumar},
booktitle={Proceedings of the WILDRE5--5th Workshop on Indian Language Data: Resources and Evaluation},
pages={14--19},
year={2020}
}
We define "optimal reference translation" as a translation thought to be the best possible that can be achieved by a team of human translators. Optimal reference translations can be used in assessments of excellent machine translations.
We selected 50 documents (online news articles, with 579 paragraphs in total) from the 130 English documents included in the WMT2020 news test (http://www.statmt.org/wmt20/) with the aim to preserve diversity (style, genre etc.) of the selection. In addition to the official Czech reference translation provided by the WMT organizers (P1), we hired two additional translators (P2 and P3, native Czech speakers) via a professional translation agency, resulting in three independent translations. The main contribution of this dataset are two additional translations (i.e. optimal reference translations N1 and N2), done jointly by two translators-cum-theoreticians with an extreme care for various aspects of translation quality, while taking into account the translations P1-P3. We publish also internal comments (in Czech) for some of the segments.
Translation N1 should be closer to the English original (with regards to the meaning and linguistic structure) and female surnames use the Czech feminine suffix (e.g. "Mai" is translated as "Maiová"). Translation N2 is more free, trying to be more creative, idiomatic and entertaining for the readers and following the typical style used in Czech media, while still preserving the rules of functional equivalence. Translation N2 is missing for the segments where it was not deemed necessary to provide two alternative translations. For applications/analyses needing translation of all segments, this should be interpreted as if N2 is the same as N1 for a given segment.
We provide the dataset in two formats: OpenDocument spreadsheet (odt) and plain text (one file for each translation and the English original). Some words were highlighted using different colors during the creation of optimal reference translations; this highlighting and comments are present only in the odt format (some comments refer to row numbers in the odt file). Documents are separated by empty lines and each document starts with a special line containing the document name (e.g. "# upi.205735"), which allows alignment with the original WMT2020 news test. For the segments where N2 translations are missing in the odt format, the respective N1 segments are used instead in the plain-text format.
This corpus contains annotations of translation quality from English to Czech in seven categories on both segment- and document-level. There are 20 documents in total, each with 4 translations (evaluated by each annotator in paralel) of 8 segments (can be longer than one sentence). Apart from the evaluation, the annotators also proposed their own, improved versions of the translations.
There were 11 annotators in total, on expertise levels ranging from non-experts to professional translators.
Parsito is a fast open-source dependency parser written in C++. Parsito is based on greedy transition-based parsing, it has very high accuracy and achieves a throughput of 30K words per second. Parsito can be trained on any input data without feature engineering, because it utilizes artificial neural network classifier. Trained models for all treebanks from Universal Dependencies project are available (37 treebanks as of Dec 2015).
Parsito is a free software under Mozilla Public License 2.0 (http://www.mozilla.org/MPL/2.0/) and the linguistic models are free for non-commercial use and distributed under CC BY-NC-SA (http://creativecommons.org/licenses/by-nc-sa/4.0/) license, although for some models the original data used to create the model may impose additional licensing conditions.
Parsito website http://ufal.mff.cuni.cz/parsito contains download links of both
the released packages and trained models, hosts documentation and offers online
demo.
Parsito development repository http://github.com/ufal/parsito is hosted on
GitHub.
PAWS is a multi-lingual parallel treebank with coreference annotation. It consists of English texts from the Wall Street Journal translated into Czech, Russian and Polish. In addition, the texts are syntactically parsed and word-aligned. PAWS is based on PCEDT 2.0 and continues the tradition of multilingual treebanks with coreference annotation. PAWS offers linguistic material that can be further leveraged in cross-lingual studies, especially on coreference.
Statistical component of Chimera, a state-of-the-art MT system. and Project DF12P01OVV022 of the Ministry of Culture of the Czech Republic (NAKI -- Amalach).
Wikipedia plain text data obtained from Wikipedia dumps with WikiExtractor in February 2018.
The data come from all Wikipedias for which dumps could be downloaded at [https://dumps.wikimedia.org/]. This amounts to 297 Wikipedias, usually corresponding to individual languages and identified by their ISO codes. Several special Wikipedias are included, most notably "simple" (Simple English Wikipedia) and "incubator" (tiny hatching Wikipedias in various languages).
For a list of all the Wikipedias, see [https://meta.wikimedia.org/wiki/List_of_Wikipedias].
The script which can be used to get new version of the data is included, but note that Wikipedia limits the download speed for downloading a lot of the dumps, so it takes a few days to download all of them (but one or a few can be downloaded fast).
Also, the format of the dumps changes time to time, so the script will probably eventually stop working one day.
The WikiExtractor tool [http://medialab.di.unipi.it/wiki/Wikipedia_Extractor] used to extract text from the Wikipedia dumps is not mine, I only modified it slightly to produce plaintext outputs [https://github.com/ptakopysk/wikiextractor].
System for querying annotated treebanks in PML format. The querying uses it own query language with graphical representation. It has two different implementations (SQL and Perl) and several clients (TrEd, browser-based, command line interface).
Texts
The Prague Czech-English Dependency Treebank 2.0 (PCEDT 2.0) is a major update of the Prague Czech-English Dependency Treebank 1.0 (LDC2004T25). It is a manually parsed Czech-English parallel corpus sized over 1.2 million running words in almost 50,000 sentences for each part.
Data
The English part contains the entire Penn Treebank - Wall Street Journal Section (LDC99T42). The Czech part consists of Czech translations of all of the Penn Treebank-WSJ texts. The corpus is 1:1 sentence-aligned. An additional automatic alignment on the node level (different for each annotation layer) is part of this release, too. The original Penn Treebank-like file structure (25 sections, each containing up to one hundred files) has been preserved. Only those PTB documents which have both POS and structural annotation (total of 2312 documents) have been translated to Czech and made part of this release.
Each language part is enhanced with a comprehensive manual linguistic annotation in the PDT 2.0 style (LDC2006T01, Prague Dependency Treebank 2.0). The main features of this annotation style are:
dependency structure of the content words and coordinating and similar structures (function words are attached as their attribute values)
semantic labeling of content words and types of coordinating structures
argument structure, including an argument structure ("valency") lexicon for both languages
ellipsis and anaphora resolution.
This annotation style is called tectogrammatical annotation and it constitutes the tectogrammatical layer in the corpus. For more details see below and documentation.
Annotation of the Czech part
Sentences of the Czech translation were automatically morphologically annotated and parsed into surface-syntax dependency trees in the PDT 2.0 annotation style. This annotation style is sometimes called analytical annotation; it constitutes the analytical layer of the corpus. The manual tectogrammatical (deep-syntax) annotation was built as a separate layer above the automatic analytical (surface-syntax) parse. A sample of 2,000 sentences was manually annotated on the analytical layer.
Annotation of the English part
The resulting manual tectogrammatical annotation was built above an automatic transformation of the original phrase-structure annotation of the Penn Treebank into surface dependency (analytical) representations, using the following additional linguistic information from other sources:
PropBank (LDC2004T14)
VerbNet
NomBank (LDC2008T23)
flat noun phrase structures (by courtesy of D. Vadas and J.R. Curran)
For each sentence, the original Penn Treebank phrase structure trees are preserved in this corpus together with their links to the analytical and tectogrammatical annotation. and Ministry of Education of the Czech Republic projects No.:
MSM0021620838
LC536
ME09008
LM2010013
7E09003+7E11051
7E11041
Czech Science Foundation, grants No.:
GAP406/10/0875
GPP406/10/P193
GA405/09/0729
Research funds of the Faculty of Mathematics and Physics, Charles University, Czech Republic, Grant Agency of the Academy of Sciences of the Czech Republic: No. 1ET101120503
Students participating in this project have been running their own student grants from the Grant Agency of the Charles University, which were connected to this project. Only ongoing projects are mentioned: 116310, 158010, 3537/2011
Also, this work was funded in part by the following projects sponsored by the European Commission:
Companions, No. 034434
EuroMatrix, No. 034291
EuroMatrixPlus, No. 231720
Faust, No. 247762