Skip to content
GitLab
Commit 0feef5a3 authored by Gérard Huet's avatar Gérard Huet
Browse files

Clean-up of XML subsite

parent ae235022
Hide whitespace changes
Inline Side-by-side
BOOK/Heritage.pdf
View file @ 0feef5a3
No preview for this file type
XML/index.html 0 → 100644
View file @ 0feef5a3
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>Sanskrit Linguistic Resources</title>
<meta name="author" content="Gérard Huet">
<meta property="dc:datecopyrighted" content="2017">
<meta property="dc:rightsholder" content="Gérard Huet">
<meta name ="keywords" content="india,dictionary,indology,sanskrit,lexicography,linguistics,indo-european,dictionnaire,sanscrit,panini,indology,linguistics">
<meta name ="date" content="2017-05-18">
<meta name="classification" content="computational linguistics, sanskrit, morphology, lexicography, indology">
<meta name="description" content="This page is for downloading Sanskrit resources.">
<link rel="shortcut icon" href="IMAGES/favicon.ico"/>
<link rel="stylesheet" type="text/css" href="DICO/style.css" media="screen,tv"/>
</head>
<body class="pink_back"> <!-- Pale_rose -->
<table class="body">
<table border="0pt" cellpadding="0" cellspacing="15pt" width="100%">
<tr><td>
<h1 class=b1>Sanskrit linguistic resources</h1>
<br>
<img src="IMAGES/Panini2.jpg" alt="Panini"/>
<br>
<div class="latin12">
<h2 class=b2>Sanskrit Morphology</h2>
<h3 class=b3>Background</h3>
This documents XML data banks of Sanskrit forms given with their morphological
taggings. They are produced mechanically by the declension and conjugation
engines of the Sanskrit Heritage Platform, processing the Sanskrit lexicon
underlying the Sanskrit Heritage Dictionary. The version used for this
generation is available <a href="Heritage.pdf">here</a> as a PDF document.
<p>
These databanks are regularly updated. They are available for public download
as a public git archive in the Sanskrit Heritage development site:
"https://gitlab.inria.fr/huet/Heritage_Resources".
<h3 class=b3>Databanks description</h3>
We provide here inflected forms and morphemes derived from the root forms
defined in the
<a href="DICO/index.html">Sanskrit Heritage Dictionary</a>. These forms are
presented as lemmas linking each form to its stem entry by possible morpho-phonetic
operations. We limit ourselves to classical Sanskrit, and do not cover precative,
subjunctive, injunctive and conditional forms of the verbs.
At present, we provide for two transliteration schemas, respectively
WX, used by the
<a href="http://sanskrit.uohyd.ernet.in/">Department of Sanskrit Studies at
University of Hyderabad</a>
and SLP1, used by the
<a href="http://sanskritlibrary.org/">Sanskrit Library</a>.
The respective data banks are given in compressed archives (6Mo each, gz format)
"WX_morph.tar.gz" (WX) and
"SL_morph.tar.gz" (SLP1. After downloading these documents,
and uncompressing them (typically with the utility gunzip),
you get a UNIX tar archive containing the following data.
The morphological lemmas are distributed in 6 files in
XML format, conformant to a common DTD.
The nominal morphological declensions of nouns, adjectives and numbers,
are covered in "T_nouns.xml" (where T is respectively WX or SL).
Those of pronouns are covered in "T_pronouns.xml".
The conjugated forms of roots in the present, imperfect, imperative, optative,
perfect, aorist
and future tenses, as well as passives of the present system,
for the primary conjugation and for some secondary conjugations
(causative, intensive, desiderative) are covered in "T_roots.xml".
Additional declensions of derived participial forms are given in T_parts.xml.
Absolutives, infinitives and other undeclinable words and particles
are listed in "T_adverbs.xml". In addition, "T_final.xml" gives additional
generative morphemes. The files are conformant to the DTD "T_morph.dtd".
<p>
Finally, the text file "X_preverbs.txt" lists common
preverb sequences, given with their sandhi analysis.
<h3 class=b2>Intellectual Property</h3>
All these linguistic data banks are Copyrighted Gérard Huet 1994-2017.
They are derived from the Sanskrit Heritage Dictionary
version 2.99 dated 2017-05-18.
<p>
Use of these linguistic resources is granted according to the
Lesser General Public Licence for Linguistic Resources.
Copies of this license, in pdf as well as HTML, are provided at
the Heritage_Resources distribution site in its XML subdirectory.
Thank you for referencing the origin of this data if you use it in your own work.
<h2 class=b2>Methodology</h2>
We deal here with a mixture of derivational and inflexional morphology.
For instance, from the roots we generate verbal and propositional stems, and from
these stems we generate in turn inflected forms: conjugated forms from the
verbal stems, and declined forms from the participial stems. But at present
we do not generate mechanically primary nominal stems from roots,
nor secondary nominal stems from primary ones, because of overgeneration.
The nominal stems, as well as the undeclinable forms, are taken from the
lexicon, that lists also some frequent participles.
<p>
This organization entails a different role in our morphological data bases.
The <i>basic</i> morphological categories correspond to lexical phases,
which are atomic letters in the defining grammar of Sanskrit <i>word</i>.
The forms listed in these data bases act as morphemes of this high-level
morphological definition, which is recursive, since compounding may be
iterated, as well as preverb formation, to a certain extent.
But this recursion power is limited, in the sense that the grammar of a word
is a regular one (type 0 in the Chomsky hierarchy), and its recognizer is
a finite automaton, whose states are precisely the lexical categories indexing
the basic data bases. This definition of word implements correctly the geometry
of constructions such as absolutives (which fall in two distinct categories,
the preverb form and the root form) and periphrastic phrases (periphrastic
futures with substantives, and periphratic perfects as prefixes of finite
perfect forms of the auxiliary roots <i>as</i>, <i>bhū</i> and
<i>kṛ]</i> which are duplicated in a specific auxiliary lexicon).
Here is a simplified diagram of the current state space of our lexer.
<div class="center">
<img src="IMAGES/lexer17.jpg" alt="Lexer automaton">
</div>
This automaton is also the top-level view of our Sanskrit Tagger, which
implements Sanskrit analysis from <i>devanagarī</i> text.
The technical exposition of this method, together with its correctness
justification, has been exposed in various scientific journals and conferences,
and the corresponding articles are also available freely on my
<a href="http://pauillac.inria.fr/~huet/bib.html">
<strong>publications page</strong></a>
(papers [78], [87], [88], [94], [95], [105], [106] and [110]
are specially relevant).
This material will not be repeated here. Let us just explain a few difficulties
of the large-scale implementation of this Sanskrit analyser.
<p>
As usual in a non-deterministic search algorithm (here all the possible parsings
of a sentence as a sandhied stream of forms), we have two pitfalls, silence and noise.
Silence (lack of recall) means incompleteness. Some legal Sanskrit sentences
may fail to be recognized.
Typicallly, some root word may be missing from the base lexicon,
or some Vedic form may use some construction rare in the later language,
like precative or subjunctive.
Compounding gives rise to two complications, the raising of new cases by
<i>bahuvrīhi</i> compounding,
and the formation of <i>avyayībhava</i> compounds. Some of these
constructions are treated incompletely.
<p>
The opposite of silence is noise (lack of precision), that is overgeneration.
We deal with overgeneration
in the syntactico-semantic layer of our tagger, which filters out combinations of
tags inconsistent with semantic role assignments.
We shall not discuss this technology
further in this note on morphology, and refer the interested reader to our
<a href="/DICO/reader.html"><strong>Sanskrit reader
demonstration page</strong></a> and its <a href="manual.html">
<strong>Reference manual</strong></a>
<p>
We remark that the respective data bases can be interrogated online by our
<a href="http://sanskrit.inria.fr/DICO/index.html#stemmer"><strong>stemmer
interface</strong></a>. But note that verbal forms prefixed by preverbs
are analysed by the tagger as non-atomic words, and only root forms and
their secondary conjugations are recognized by the stemmer.
<h2 class=b2>Help</h2>
Questions concerning these resources should be addressed to
<a href="mailto:Gerard.Huet@inria.fr">Gérard Huet</a>.
All suggestions for improvements will be gratefully considered.
</td></tr>
</table>
</div>
<table class="pad60"> <!--padding for bandeau -->
<tr><td></td></tr></table>
<div class="enpied">
<table class="bandeau"><tr><td>
<a href="http://ocaml.org">
<img src="IMAGES/icon_ocaml.png" alt="Objective Caml" height="50"></a>
</td><td>
<table class="center">
<tr><td>
<a href="index.html"><strong>Top</strong></a> |
<a href="DICO/index.en.html"><strong>Index</strong></a> |
<a href="DICO/index.en.html#stemmer"><strong>Stemmer</strong></a> |
<a href="DICO/grammar.en.html"><strong>Grammar</strong></a> |
<a href="DICO/sandhi.en.html"><strong>Sandhi</strong></a> |
<a href="DICO/reader.en.html"><strong>Reader</strong></a> |
<a href="faq.en.html"><strong>Help</strong></a> |
<a href="portal.en.html"><strong>Portal</strong></a>
</td></tr>
<tr><td>© Gérard Huet 1994-2017</td></tr>
</table></td><td>
<a href="http://www.inria.fr/">
<img src="IMAGES/logo_inria.png" alt="Logo Inria" height="50"></a>
<br></td></tr></table></div>
</body>
</html>
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment