| 1 |
An EM algorithm for context-based searching and disambiguation with application to synonym term alignment
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/PACLIC.2009/EM.CBS.C2C.PACLIC09.2009.1019.v18.15.+proof.pdf (2010)
|
|
BASE
|
|
Show details
|
|
|
| 2 |
A Preliminary Study on Probabilistic Models for Chinese Abbreviations
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/acl04/acl04.sighan.v08R.Jing.Shin.Chang.PDF (2004)
|
|
BASE
|
|
Show details
|
|
|
| 3 |
A preliminary study on probabilistic models for Chinese abbreviations
|
|
|
|
In: http://www.aclweb.org/anthology/W/W04/W04-1102.pdf (2004)
|
|
BASE
|
|
Show details
|
|
|
| 4 |
A customizable, self-learnable parameterized MT system: the next generation
|
|
|
|
In: http://www.mt-archive.info/MTS-1999-Su.pdf (1999)
|
|
Abstract:
In this paper, the major problems of the current machine translation systems are first outlined. A new direction, highlighting the system capability to be customizable and self-learnable, is then proposed for attacking the described problems, which are mainly resulted from the very complicated characteristics of natural languages. The proposed solution adopts an unsupervised two-way training mechanism and a parameterized architecture to acquire the required statistical knowledge, such that the system can be easily adapted to different domains and various preferences of individual users. However, most current machine translation systems still cannot adapt to the individual's preference. Lacking such adaptation capability will render the MT system making repetitive errors and thus inducing user frustration, because they have no way to teach the dumb system to stop generating those stupid errors. In contrast, many applications in other fields (e.g., speech recognition and on-line character recognition) are usually user adaptable. You can let the computer learn what you prefer and significantly improve the performance. 1 Current Status Most MT systems, currently, adopt a generalpurpose kernel for every application. When different domains are encountered, they only swap the corresponding technical compound dictionaries, without tuning the system with the associated domain knowledge. Due to lacking the capability to economically acquire the huge and fine-grained knowledge 1 (both linguistic and real-world) required for different domains, the quality of those general purpose machine translation systems does not show much improvement during the last 50 years. As a result, the MT output quality usually cannot match user's expectations, and the threshold for user acceptance has not yet been reached. Since the usefulness of an MT system will be appreciated only when its quality exceeds the threshold, machine translation systems are still not widely used as a useful translation tool by many people. In addition, various users and companies usually have different preferences in technical term translation, and show different tastes on translation output styles.
|
|
URL: http://www.mt-archive.info/MTS-1999-Su.pdf
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.332.5461
|
|
BASE
|
|
Hide details
|
|
|
| 5 |
An Unsupervised Iterative Method for Chinese New Lexicon Extraction
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/ukw9706f.ps.gz (1997)
|
|
BASE
|
|
Show details
|
|
|
| 6 |
A Multivariate Gaussian Mixture Model for Automatic Compound Word Extraction
|
|
|
|
In: http://www.bdc.com.tw/~shin/doc/rocling/cpnr.RocX.final.ps.Z (1997)
|
|
BASE
|
|
Show details
|
|
|
| 7 |
An Overview of Corpus-Based Statistics-Oriented (CBSO) Techniques for Natural Language Processing
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/overview.cbso.tech.ps.gz (1996)
|
|
BASE
|
|
Show details
|
|
|
| 8 |
Automatic construction of a Chinese electronic dictionary
|
|
|
|
In: http://www.mt-archive.info/VLC-1995-Chang.pdf (1995)
|
|
BASE
|
|
Show details
|
|
|
| 9 |
A corpus-based two-way design for parameterized mt systems: Rationale, architecture and training issues
|
|
|
|
In: http://www.mt-archive.info/TMI-1995-Su.pdf (1995)
|
|
BASE
|
|
Show details
|
|
|
| 10 |
A Corpus-Based Statistics-Oriented Two-Way Design for Parameterized MT Systems: Rationale, Architecture and Training Issues
|
|
|
|
In: http://www.bdc.com.tw/~shin/doc/cbso.2way.mt.ps.gz (1995)
|
|
BASE
|
|
Show details
|
|
|
| 11 |
Introduction to Corpus-based Statistics-oriented (CBSO) Techniques
|
|
|
|
In: http://www.aclclp.org.tw/clclp/v1n1/v1n1a4.pdf (1994)
|
|
BASE
|
|
Show details
|
|
|
| 12 |
A Corpus-Based Statistics-Oriented Transfer and Generation Model for Machine Translation
|
|
|
|
In: http://www.mt-archive.info/TMI-1993-Chang.pdf (1993)
|
|
BASE
|
|
Show details
|
|
|
| 13 |
Statistical models for word segmentation and unknown resolution
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/sws/sws.PDF (1992)
|
|
BASE
|
|
Show details
|
|
|
| 14 |
GPSM: A Generalized Probabilistic Semantic Model for Ambiguity Resolution
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/archtran/PDF/gpsm.acl92.2c.PDF (1992)
|
|
BASE
|
|
Show details
|
|
|
| 15 |
GPSM: A Generalized Probabilistic Semantic Model for Ambiguity Resolution
|
|
|
|
In: http://acl.ldc.upenn.edu/P/P92/P92-1023.pdf (1992)
|
|
BASE
|
|
Show details
|
|
|
| 16 |
Why Corpus-Based Statistics-Oriented Machine Translation
|
|
|
|
In: http://www.mt-archive.info/TMI-1992-Su.pdf (1992)
|
|
BASE
|
|
Show details
|
|
|
| 17 |
The Semantic Score Approach to the Disambiguation of PP Attachment Problem
|
|
|
|
In: http://www.cs.nccu.edu.tw/~chaolin/papers/rocling90.pdf (1990)
|
|
BASE
|
|
Show details
|
|
|
| 18 |
Mining Atomic Chinese Abbreviation Pairs with a Probabilistic Single Character Word Recovery Model
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/LRE/AAP.LRE.0612.v08R.29.short.21.Jing.Shin.Chang.pdf
|
|
BASE
|
|
Show details
|
|
|
| 19 |
A Chinese-to-Chinese Statistical Machine Translation Model for Mining Synonymous Simplified-Traditional Chinese Terms
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/crs/doc/summit/MTS-2007-Chang-2.pdf
|
|
BASE
|
|
Show details
|
|
|
| 20 |
Computational Tools and Resources for Linguistic Studies
|
|
|
|
In: http://nlp.csie.ncnu.edu.tw/~shin/doc/rocling/CLTools.ps.Z
|
|
BASE
|
|
Show details
|
|
|
|
|
