|
Clients:
Several Clients, mostly in International software
business with a couple of end-users (governmental
agencies).
Location:
Throughout the world
Introduction:
In 1987, in the very beginnings of NLP (Natural Language
Processing) research Era, work began with a large
team of linguitions, developers and consultants to
try to come up with a mathematical formula to deal
with the Arabic language. This team at its peak, reached
twenty people, divided into three groups (Linguitions,
Data gathering and entry, in addition to research
and development)
Objective:
The initial goal of the project was to launch the
first ever Arabic Spell Checker. Something that did
not exist in the Arabic language under any platform.
This goal then was extended to develop the first Arabic
morphological analyzer, grammar checker, dictionary
and thesaurus.
Challenges:
To try to derive the mathematical formulas related
to the unique nature of the Arabic language, such
as the complex nature of prefixing and suffixing,
and the enormous number of words in its lexicon.
Being pioneers in the field of creating the first
ever Arabic Spell Checker was the greatest challenge
especially since the natural language processing science
was still not mature enough.
Different Arabic schools of writing roles was to be
handled.
Scope of Work:
>>Professional development teams spent enormous
amount of time and much effort to come up with a mathematical
structure to understand and use Arabic language in
order to premier spell checker in 1989, and to emphasize
the leadership in this area of technology.
Results:
>>Developed first ever Arabic Spelling Checker
in 1989, followed by the first ever Arabic Grammar
Checker, Arabic Morphological Analyzer, and Arabic/English,
English/Arabic dictionaries.
>>Fourteen of the world’s leading software companies
own licenses of our pioneering linguistic technologies
in all the above programs.
Several large to huge corporate/organizations were
granted license to use our linguistic technologies
within their applications in sophisticated areas of
Arabic textual databases as morphological analyzers.
>>Tens of scientific papers were presented in
international conferences and conventions talking
about computerized Arabic linguistic technologies.
>>Indisputable reputation in the field of natural
language processing (NLP).
Duration and Staff:
190 person/years were invested in different areas
of research and development, including linguistic
research, programming, and study, artificial intelligence,
computer programming. Several technical platforms
were used.
Technical Skills:
The main bulk of the technology were developed using
strict ANSI C enabling the components to successfully
run under many platforms including Windows XX (16
& 32 bits)/ MAC OS/ Unix as well as DOS.
|