Frequently Asked Questions (FAQs)

Q: What is Anusaaraka?

Ans: Anusaaraka is a Language accessing software which shall allow users to access text in any Indian language, after translation from the source language (i.e. English or any other regional Indian language). This machine translation software is based on Information Dynamics, inspired by Panini's grammar.

In the version of Anusaaraka software that is being spoken of here, English is the source language and the target language is Hindi.
The Anusaaraka software is under the General Public License (GPL), and is available for free.

Q: How does Anusaaraka differ from conventional Machine Translation efforts?

Ans: The most distinct feature of Anusaaraka is that it does not merely give a translated output, but a layered output as well. Translation of the source language text to the target language text is shown step-by-step. Given below is a display of how Anusaaraka will show output for translation of a sentence:

A

I

hope

to

understand

what

you

are

saying

B

I

hope

to

understand

what

you

are#be

say#saying

C

PRP{pron}

VBP{v}

TO{to}

VB{v}

WP{wh}

PRP{pron}

VBP{v}

VBG{v}

D

मैं

आशा~कर{tam:ता_है}

->{tam:ने~की}

समझ

जो

आप

-->

कह{tam:0_रहा_है{ब.}}

E

मैं

आशा_करता_हूँ

--

समझने_की

जो

आप

--

कह_रहे_हो.

In the above table: the first row A displays the sentence given for translation. The second row gives the 'root' word for each of the words in the English sentence. The third row C displays the 'parts of speech' i.e. a verb , pronoun, noun etc. Row D gives the Hindi translation for each word and in the final row the tense, gender, number, person of words are taken into account.

The table above shows Anusaaraka output. Thereafter, some more steps are done at the backend and the final output is displayed after prepositions are moved to the correct position. This step is called Machine Translation. Thus, Anusaaraka cum Machine Translation output would be: 'मैं जो आप कह रहे हो समझने की आशा करता हूँ l'

If the translated text is not clearly understood, the user can scan through various layers of the translation output and interpret the translation based on the information in these layers of output and the context in which the sentence appears. An individual shall need some training for reading the various layers in the output.

Q: What kind of training is required?

Ans: The training involves understanding of some basic concepts of grammar and knowledge of specific grammatical features of the source language.

Q: How is this training imparted?

Ans: Resource material for the same will be available on the web. In addition, there are also plans to introduce a course on Applied Grammar based on Paninian concepts as institutional curriculum.

Q: What is the current status of the Anusaaraka project?

Ans: In the past we have developed Anusaarakas for translating Telugu, Kannada, Marathi, Punjabi and Bangla into Hindi.

Currently we are concentrating all our efforts on the English-Hindi Anusaaraka. This is an acid test for the Anusaaraka concept, since the complexity involved in the development of English-Hindi Anusaaraka is much more than that involved in the development of Anusaarakas among Indian languages, as English and Hindi have such varied grammatical structures.

These Anusaarakas are usable but a lot more work needs to be done to improve them qualitatively and quantitatively.

SIGNIFICANCE OF ANUSAARAKA

Indian Traditional dimension:

Q: What is the significance of this activity from the shastric tradition viewpoint?

Ans: Anusaaraka is an attempt to demonstrate the application of ideas from the Indian tradition (viz. use of Paninian concepts in Information Dynamics).

In the current Indian scenario a typical researcher in a technical institute, working for example in the field of Natural Language Processing (NLP), is not even aware that the Indian tradition has deep insights relevant for her / his work.

This practical demonstration will establish the case for relevance of the Indian tradition in the present scenario, which in turn would naturally remove the stagnancy of research in our traditional shastras.
Social dimension:

Q: Apart from the traditional viewpoint, does it have any other social significance?

Ans: The world today is undergoing an Information Revolution. More and more people are getting involved in study, research, development, innovation etc. in the field of IT. It is of paramount importance to make sure that we not only become consumers of the Information technologies but also participate effectively as technology contributors. The underlying philosophy is that of 'svaavalambana' (self-dependence).

Anusaaraka project opens up ample opportunities for all members of the society to actively participate in this Information Revolution as Information Technology contributors.

AN INVITATION FOR PARTICIPATION

Q: How can I participate?

Ans: If you can spare a little time, help in creating awareness among others about the Anusaaraka project. If you can spare some time every week / have special interest in languages and love to work on languages. Please register as a participant on this site. You shall be able to access the Guidelines created for Language Resource Development work and other resources which will familiarize you with the task.

If you do not have time to spare for this activity you can help to raise funds:
    *to reward the participants who are not volunteers;
    *to give scholarships to students taking up Anusaaraka as their research topic;
    *to create a corpus fund for research which aims at concrete application of Indian tradition shastras.

form input

Root/Word