Transcript 발표자료 - 서울시립대학교

Text Mining Tool
- Carabao
서울 시립대학교
전자전기컴퓨터 공학부
안민영
Data Mining Lab
Contents
What is Carabao?
Solutions
Linguistic Data
References
Demo
Company Logo
What is Carabao?(1/2)
Carabao
필리핀을 상징하는 심볼 중의 하나임.
물소(water buffalo)의 일종으로 필리핀에서 자주 볼수 있음
Company Logo
What is Carabao?(2/2)
Carabao Language Kit?
 Main purpose:
• Understand text
• Transform text from any language to any language
 Analytics and automatic translation
•
•
•
•
•
•
Part of speech tagging
Sense disambiguation
Named entity extraction
Deep morphological analysis and synthesis
Idiom extraction
Automatic transliteration between languages
Data Mining Lab
Solutions
Carabao
Text analytic
is a process of
information extraction
ex)
Nice to see you here
VS
Nice is a great place to
relax
Automatic
Translation
Is a sub-field of
computational linguistics
-Transliteration between
scripts
-Automatic translation
between languages
Company Logo
Linguistic Data
 Transformation configuration - Commercial
Cebuano, Czech  English
English  French
Data Mining Lab
References
1
2
www.digitalsonata.c
om
www.wikipidea.org
www.themegallery.
3
Carababomanual.pdf
Company Logo
Demo
Execute Carabao Program
Data Mining Lab