Background
Currently,
we have to admit that the inability to access to computers skillfully and to
access to the information on computer network causes a lot of disadvantages.
The opportunity in the information society is differentiated by the ability in
accessing and expressing each own knowledge. The Internet connecting people
dramatically changes the way we communicate with each other, i.e. computer to
computer and human to computer. By means of computer and the Internet, many
research and development projects have been continuously conducted to
facilitate the interaction between human and computer in a natural and smart
way. However, language as a communication tool that we can efficiently use to
communicate and record our experiences. Many advanced countries have been
spending a lot of attempts to enable their people to communicate with computers
by their natural communication skills, mostly by their own mother tongue. To
maintain the language diversity, we realize that the technology of digitizing
and processing the natural language plays an important role.
Thai
Computational Linguistics Laboratory of NICT Asia Research Center, NECTEC,
Asian Language Resource Network of NUT and SIIT, who are in the forefront of
the research in natural language processing (NLP), are cooperating in
organizing a series of applied NLP lectures for enlarging the NLP expert community
especially through the viable experiences. The series of lectures covers the
broad range of NLP research realized in many practical projects. Experts in the
successful projects are invited to provide the fundamental theories and their
implementations. Many systems and development tools will be introduced in
hands-on exercises aiming at future adoption of the technologies. Enlargement
of the NLP expert community, collaboration, knowledge and resource sharing are
our primary interests.
Objectives
The
initiative of the School for Asian Applied NLP is fundamentally aiming at
sharing the experiences among the experts in the field of NLP. Bridging the
experts to the less experienced region will not only help leveraging the
technology of using one's mother tongue in the region, but also strengthen the
language research. The activities range from technology transferring to sharing
the research and language resource development through a series of lectures and
hands-on exercises. Followings are the objectives.
- Build experts
in NLP
- Build a human
network of NLP expert for sharing the experience, expertise, and
collaboration in studying and applying NLP
- Support the
development of language resources for studying and evaluating the
technology
- Support the
development of standards for language resource development
- Support the
research and development of NLP common utilities
- Support the
implementation of the existing NLP utilities
Organizer and Supporter
By
means of technology transfer, experts, and financial assistance, a series of
Applied NLP lectures is supported by
- NICT Asia Research
Center
- Asian Language
Resources Network Project
- National
Electronics Computer and Technology
Center (NECTEC)
- Sirindhorn International
Institute of Technology (SIIT)
- Asia-Pacific
Association for Machine Translation (AAMT)
- Asian
Federation of Natural Language Processing (AFNLP)
- PAN
Localization Project, CRULP
Activities
In
order to increase the number of NLP experts and language resources, we propose
a framework of activities consisting of the following two phases for
integration.
1.
A
series of lectures on applied NLP
2.
Workshop
on applied NLP and language resource development
1. A series of lectures on applied NLP
Topics of great
relevance to researchers and developers in applied NLP have been selected for
the program.
The
program is provided in 3 courses. Each course will end up in 9 days together
with a hands-on training. A workshop to update each other works will be
consecutively held in a period of 1 day after each course.
Course 1: Introduction to NLP
v
Overview
NLP
o
State
of the art of NLP
v
Morphological
Analysis for Non-segmenting Languages
o
Hidden
Markov model
o
Conditional
Random Field
o
Word
Segmentation
§
Longest
matching approach
§
Maximal
matching approach
§
Part-of-speech
tri-gram approach
o
Word
Extraction
v
Search
Engine
o
History
o
Search
Engine Types
o
Crawler
o
Indexer
o
Searcher
o
Lucene Agreement
o
Lucene Example
o
Lucene API
o
Eclipt
o
Analyzer
o
Administration
Tools
o
Evaluation
v
Corpus
Development
o
Introduction
to Corpus Linguistics
§
What
is a corpus?
§
What
is in a corpus?
§
What
is (are) purpose(s) of corpora?
§
How
many kinds of corpora are there?
§
How
to acquire source data?
o
Corpus
for Machine Translation
§
How
does corpus relate to MT?
§
How
to build a corpus for MT?
§
Frequently
found problems in corpus development
§
What
is corpus tool(s) and what does it do?
§
Part
of speech tagger and practice and ORCHID
v
KUI
for Asian Wordnet Initiative
o
KUI
the Knowledge Unifying Initiator
o
Wordnet
o
Euro
Wordnet
o
Asian
Wordnet Initiative
v
Web
Language Engineering
o
Collaborative
crawler
o
Language
identification
§
N-gram
model
§
String
kernel approach
§
Web
document archiving
o
Multi-lingual
search engine
v
Machine
Translation
o
Overview
§
History
of Machine translation
§
Problems
on machine translation
§
Techniques
for machine translation
§
Trend
of research on machine translation
o
Parsing
§
Introduction
to parsing techniques
§
Parser's
components: "parser = mechanism + grammar rules"
§
General
characterization of parsing techniques
§
Apply
parsing to MT
o
Corpus
based MT
§
Introduction
to corpus based MT
§
Developing
corpus
§
Example
based approach
§
Statistical
based approach
§
Improving
quality of MT by using corpus
o
Experiment
§
Develop
their own corpus
§
Test
on toolkit
v
Speech
Recognition
o
Fundamentals
of language & speech
o
Overview
of ASR
o
Acoustic
modeling by HMM
o
Language
modeling
o
Speech
decoding
o
Introduction
to toolkits for ASR
o
Corpus
& ASR component preparation
o
Building
your own ASR
o
Evaluation
& improvement
v
Speech
Synthesis
o
Overview
of TTS
o
Text
processing
o
Linguistic/prosodic
processing
o
Waveform
synthesis
o
Introduction
to toolkits for TTS
o
Corpus
preparation
o
Building
your own TTS
v
Optical
Character Recognition
o
Overview
(Pre-processing, Processing, Post-processing)
o
Pre-processing
§
Image
Enhancement
§
Alignment
§
Page
Layout
§
Binarization
§
Character
Segmentation
o
Processing
§
Feature
Extraction (Thinning, Bitmap, Edge, Structural Extraction, Projection, Run
Length Coding)
§
Training
(SVM, Neural Network)
o
Post-processing
§
Pasting
§
Language
Model
§
Spelling
Correction
v Advanced Research
Course 2: Advanced NLP
v
Information
retrieval and information extraction
v
Search
engine
v
Text
summarization
v
Machine
Translation
v
Interlingua
and knowledge representation
v
Text,
web and data mining
v
Text
classification and language identification
Course 3: Image and Speech processing
v
OCR
v
Image
recognition
v
Image
search
v
Image
corpus
v
Text-to-speech
v
Speech
recognition
v
Speech
search
v
Spoken
language identification
v
Speech
corpus
2. Workshop on Asian Applied NLP and
language resource development
Workshop on Asian
Applied NLP and language resource development is planned to host in the second
phase, after the school. The workshop aims to bring together researchers in NLP
and computational Linguistics, updating and exploring novel ideas on NLP.
Venue
Sirindhorn International
Institute of Technology, Pathumthani, Thailand
Accommodation
Pathumthani Place
Date
The school will be organized in a sequence
of the above 3 courses.
Schedule of the 1st School of Asian ANLP (Course 1)
June 1 - July 21, 2006 Call
for participation
July 21, 2006 Due
date of application submission
July 31, 2006 Notification
of participant acceptance
August 21 - September 1, 2006 The
1st School of Asian ANLP (Course 1: Introduction to NLP)
Participation
The primary participants/trainees are researchers, developers and
integrators from Asian region who have a high potential to be a leader and to
continue establishing R&D on this field in the country. Each course has a
capacity of 30 participants. For each course, the project prepares some
financial support to subsidize only 12 participants whom will be nominated from
the institutional representatives of each country and will be selected by the
program committee on profile basis. The person who has good knowledge
background of NLP, strong interest in NLP and language study, good computer
programming skill, and strong willing to participate and make contribution in
NLP implementation projects is appreciated.
Application
Please complete the application form (
) and send to
virach@tcllab.org (subject: application of ADD)
or to the fax number 66-2564-7992
Achievement
Participants are
expected to have a general background on NLP together with some familiar with
advanced NLP via theoretical and practical experiences. After the completion of
the school, participants should be able to function as a leader in country to
implement and establish NLP related research and development to cope with at
least each own native language. Collaboration and language resource sharing is
expected on the expert network.
Program committee
Yoshimoto Shigetoshi NICT
Asia Research Center, Thailand
Isahara Hitoshi AAMT
Yoshiki Mikami Nagaoka University
of Technology, Japan
Thaweesak Koanantakool NECTEC,
Thailand
Benjamin Tsou AFNLP
Sarmad Hussain CRULP,
Pakistan
Thanaruk Theeramunkong SIIT,
Thailand
Virach Sornlertlamvanich TCL,
NICT, Thailand
Contact person
Dr. Virach Sornlertlamvanich
Thai Computational Linguistics Laboratory
(TCL), NICT Asia Research
Center
112
Phahon Yothin Road, Klong 1, Klong Luang
Pathumthani 12120, Thailand
Email: virach@tcllab.org
Tel: +66-2564-7990 Fax:
+66-2564-7992
