Bengali and English/Anubadok

Anubadok is an open source English to Bengali MT system developed by G M Hossain, currently in experimental stage.

The program is licensed under GPL. Its accessible from here.

Inflection Rules (BnSondhi.pm)

Legend

• C Consonant
• V Vowel
• _ Any Letter
• K Kar (Short form of Vowel)
• G General Rule - The consonants and vowels in the example are exchangeable with any other consonants and vowels
• S Special Rule - The consonants and vowels in the example are NOT exchangeable with any other consonants and vowels
• H Hasanth - The joiner, e.g. ঙ+্+গ = ঙ্গ, as in মঙ্গল

__CC means the last two consonants of a word, and CV__ means the first letter of this word is consonant followed by a vowel

Rules

Verb Rules (Applies if the length of the first word is >= 3, rules are not exclusive e.g. one word can qualify for multiple rules)

• G (C+ো+C)+(তে) = (C+ু+C+তে) e.g. খোল + তে = খুলতে

• G (া+C+া) + (তে) = (া+C+া+তে) e.g. পাঠা + তে = পাঠাতে

• G (C+ি)/(C+ে) + (ে+র)্(া+র) = (C+ে+ও+য়+া+র) e.g. নি + ের = নেওয়ার, নে + ার = নেওয়ার

• G (C+া) + (তে) = (C+ে+তে) e.g. পা + তে = পেতে

• G (C+ে) + (তে) = (C+ি+তে) e.g. দে + তে = দিতে

• G (C+C) + (ে+র) = (C+C+া+র) e.g. কর + ের = করার

• G (__K1) + (K2__) = (__K1__) [needs review]

Preposition Rules

(Applies if the length of the first word is >= 2, rules are not exclusive e.g. one word can qualify for multiple rules)

• G (ল+ে+C) -> (ল+ি+C) e.g. লেখ -> লিখ

• G (এ+ই) + (টা) = (এটি) e.g. এই + টি = এটি

• S (_+ক+ে) + (ে+র) = (_+র) e.g. আমাকে + ের = আমার

• G (C+C) + (ৈ+র) = (C+C+া+র) e.g. কর + এর = করার

• G (C/V/H/K + C) + (তে) = (C/V/H/K + C + ে) [needs review]

• S (ং) + (ে+র) = (ং+এ+র) e.g. সং + ের = সংএর

• G (__K1) + (K2__) = (__K1__) [needs review]

Main Rules

(Applies if the length of the first word is >= 2, rules are not exclusive e.g. one word can qualify for multiple rules)

• G (K+C) + (তে) = (K+C+ে) e.g. মার + তে = মারে [need review]

• G (__K1) + (K2__) = (__K1__) [needs review]

Sondhi for progressive tag POS

• G (K) + (_) = (K)

Basic Verb Shondhi

• G (_+ি) + (_) = (_+ে+ও+য়+_)

Verb Shondhi for passive sentence

• G (C/K + C) + (_) = (C/K+C+_) else

• G (_+ি) + (_) = (_+ে+ও+য়+_)

Verb Shondhi for active sentence

(Applies if the length of the first word is >= 1, rules are not exclusive e.g. one word can qualify for multiple rules)

• G (_ে) + (ো) = (_া+ও) e.g. নে + ো = নাও

• G (C+ে) + (_) = (C+ি+_) e.g. দে -> দি

• G (ল+ে+C) -> (ল+ি+C) e.g. লেখ -> লিখ [why anubadok repeats this rule?]

3rd person, present simple

• S (ল+ি+C) + (ে) = (ল+ে+C+ে) e.g. লিখ + ে-> লেখ + ে-> লেখে
• S (দি) + (ে) = দেয় e.g. দি + ে-> দে + ে-> দেয়
• G (C/K+C) + (ে) = (C/K+C+ে) else
• G (_) + (ে) = (_য়) e.g. হ + ে-> হয়, খা + ে-> খায়

2nd person, present simple

1st/2nd/3rd person, future simple

1st/2nd/3rd person, past simple

1st person, present simple

1st/2nd/3rd person, present/past continuous

1st/2nd/3rd person, present/past, perfect

Tag Set

Anubadok uses Penn Treebank Tag Set, the tag set is as follows:

Tag	Gloss	Example
CC	Coordinating conjunction
CD	Cardinal number
DT	Determiner
EX	Existential there
FW	Foreign word
IN	Preposition or subordinating conjunction
JJ	Adjective
JJR	Adjective, comparative
JJS	Adjective, superlative
LS	List item marker
MD	Modal
NN	Noun, singular or mass
NNS	Noun, plural
NP	Proper noun, singular
NPS	Proper noun, plural
PDT	Predeterminer
POS	Possessive ending
PP	Personal pronoun
PP$	Possessive pronoun
RB	Adverb
RBR	Adverb, comparative
RBS	Adverb, superlative
RP	Particle
SYM	Symbol
TO	to
UH	Interjection
VB	Verb, base form
VBD	Verb, past tense
VBG	Verb, gerund or present participle
VBN	Verb, past participle
VBP	Verb, non-3rd person singular present
VBZ	Verb, 3rd person singular present
WDT	Wh-determiner
WP	Wh-pronoun
WP$	Possessive wh-pronoun
WRB	Wh-adverb