Phil Humnicky/GU
The 6th Conference of the Association for Machine Translation in the Americas
September 28 - October 2, 2004
Phil Humnicky/GU |
AMTA-2004 The 6th Conference of the Association for Machine Translation in the Americas |
|
| Georgetown University, Washington DC September 28 - October 2, 2004 |
||
|
Home
|
TutorialsThe AMTA 2004 Tutorials will be held on September 28th from 9 AM to 12 noon and from 2 to 5PM. The workshop titles follow (with organizers). Morning Tutorials (9 AM - Noon, 28 September 2004)
Afternoon Tutorials (2 - 5 PM, 28 September 2004)
Tutorial costs for the tutorials are posted on the Registration page. For any questions, please contact Mike Dillinger mike.dillinger@pobox.com. Tutorial Abstracts:Arabic Natural Language Processing This tutorial will provide MT system developers/researchers with necessary background information for working with Modern Standard Arabic. The goal of the tutorial is to introduce Arabic linguistic phenomena that need to be addressed and review the state-of-the-art on Arabic processing. The focus will be on text-to-text MT from Modern Standard Arabic to English, but other applications with a cross-lingual component will be visited (such as cross-lingual information retrieval). Also some reference will be made to Arabic dialect research. Dr Nizar Habash received his PhD in 2003 from the Computer Science Department, University of Maryland College Park. He is currently a Postdoctoral researcher at the Center for Computational Learning Systems in Columbia University. His research focuses on Machine Translation, Natural Language Generation and Lexical Semantics. His work on Arabic ranges from research in Arabic encoding issues to Arabic-English Machine Translation and computational modeling of Arabic dialects for Machine Translation and Automatic Speech Recognition. Interlinguas & Semantic Markup There has been a resurgence of interest in interlinguas and semantic roles, now focussed on semantic markup using XML, rather than on interlingua-based translation. XML, the Semantic Web, and interlinguas provide increasing amounts of semantic markup within the source text. This trend will have widespread effects on machine translation. In this tutorial, we will review recent efforts to design and use interlinguas and semantic markup, and the new prospects for applying them to translation. Mike Dillinger is Director of Linguistics for Spoken Translation, Inc., where he supervises the creation of semantic knowledge bases for interactive translation, and consults on localization technologies for clients in industry. He was Director of Linguistic Development for Logos Corporation and for Global Words Technologies where he produced their Portuguese>English system, developed innovative workflow processes, and led the effort to integrate MT with tools for controlled language and translation memory. He also contributed to the Universal Networking Language Project of the United Nations University in Tokyo, where he wrote specifications for the Universal Networking Language interlingua as part of a global effort to implement distributed, interlingua-based machine translation. Previously, Dr. Dillinger did his doctoral research at McGill University in Canada on how interpreters carry out semantic processing during real-time speech-to-speech translation and taught semantics and other disciplines at more than a dozen universities in several countries, and has been a visiting researcher on four continents. Participants will be active observers in every step in the translation process at the actual setting of PAHO Translation Services: receiving a request, creating an entry in the translation tracking system, MT processing, MT postediting, revision, final check, delivery, feedback and dictionary updating. Target audience: MT users, translators, developers. Julia Aymerich studied Philology and Applied Linguistics at Universidad de Sevilla, Spain. Came to Washington in 1989 to study Computational Linguistics at Georgetown University. Has been working at PAHO Machine Translation program in 1989, first as a lexicographer, later as computational linguist and currently as senior computational linguist. Responsible for all MT development, user training, and tech support. Is and has been both developer and daily user of MT. This tutorial demonstrates principles on how to master commercial
MT systems with MT dictionary building with the goal of improving translation
output. Jeff Allen is currently the Program Manager of Product Training and Knowledge Transfer at Mycom France in the Product Division of Mycom International. He has also served on the MultiLingual Computing and Technology magazine Editorial Advisory Board for nearly four years. After completing masters and doctoral degrees in general linguistics and computational linguistics at French universities, and teaching at several universities, he then worked on several Natural Language Processing projects/products concerning controlled language, translation systems / software, and speech systems: 1) trainer of technical writing and translation tools on the controlled language authoring and machine translation project at Caterpillar Inc; 2) Research Linguist on wearable speech-to-speech translation systems of the DIPLOMAT project at the Center for Machine Translation (Carnegie Mellon University); 3) Technical Director for several international projects funded by the Evaluation and Language Resources Distribution Agency and the European Commission; and 4) Director of Projects for commercial MT software at Softissimo. Machine Translation: He has been working on MT dictionary building and MT postediting principles for the past decade. He is the author of the chapter"Postediting" in the book "Computers and Translation: A handbook for Translators" (John Benjamins, 2003) and has published extensively on topics related to MT systems and use. What's New in Statistical Machine Translation Accurate translation requires a great deal of knowledge about the usage and meaning of words, the structure of phrases, the meaning of sentences, and which real-life situations are plausible. Recently, there has been a fair amount of research into extracting translation-relevant knowledge automatically from large collections of manually-translated texts, and over the past years, several statistical MT projects have appeared in North America, Europe, and Asia, and the literature is growing substantially. We will overview this progress. Philipp Koehn is the author of papers on natural language processing, machine translation, and machine learning. He received his PhD from the University of Southern California in 2003, and is currently employed as a postdoc at the Massachusetts Institute of Technology, working with Michael Collins. He has worked at AT&T Laboratories on text-to-speech systems, and at WhizBang! Labs on text categorization. Kevin Knight is a Senior Research Scientist at the USC/Information Sciences Institute and an Research Associate Professor in the Computer Science Department at USC. He has written a number of articles on statistical MT, plus a widely-circulated MT workbook (http://www.isi.edu/natural-language/mt/wkbk.rtf). Dr. Knight has also given several MT-related talks at the AMTA and EMNLP conferences, such as "Statistical Machine Translation: Where Did It Go?", "Every Time I Fire a Statistician, I Get a Warm Fuzzy Feeling", and "Deeper Representations for Machine Translation: Ready or Not?" XML and Localization: Optimizing your Content Supply Chain This tutorial focuses on real-world case studies, showing how Fortune 500 companies have dramatically reduced localization and desktop publishing (DTP) costs with the implementation of XML-based technology. These solutions were designed to address the following problem areas related to localization:
The tutorial will cover the following topics:
amta2004@amtaweb.org |