Tutorials

This tutorial is for people who are beginning to evaluate how well machine translation will fit their needs or who are curious to know more about how it is used. We assume no previous knowledge of machine translation. We focus on background knowledge that will help you both get more out of the rest of the MT Summit and to make better decisions about how to invest in machine translation. Past participants have ranged from tech writers and freelance translators who want to keep up to date to VPs and CEOs who are evaluating technology strategies for their organizations. We cover questions such as: Can machines really translate?! Can we fire our translators now? Why is the output so bad? What is MT good for? Why buy MT if it.s free on the internet? What other kinds of translation automation are there? How do we use it? Which MT system should we buy? What do we do next?

High quality translation output in Statistical machine translation (SMT) is dependent on the availability of massive amounts of parallel text in the source and target language. There are a large number of languages that are considered "low-density", either because the population speaking the language is not very large, or even if millions of people speak the language, insufficient online resources are available in that language. This tutorial covers machine learning approaches for dealing with such situations in SMT where the amount of available bilingual data is limited.

This tutorial is designed for researchers and tool developers of Fully Automated Machine Translation (FAMT) and/or of Computer Assisted Machine Translation (CAMT). It focuses on the tools that are currently available for human translators, on how those approaches and tools could improve machine translation, and on how FAMT techniques and tools could help CAMT. The tutorial will also cover relevant standards, including the ISO Lexical Markup Framework which enables the exchange of lexical data between FAMT and CAMT systems (e.g., electronic dictionary systems for translators).

In this tutorial, DePalma presents the business drivers, metrics, and best practices associated with successful MT implementations. Based on current research at Common Sense Advisory and interviews with owners of deployed MT applications, he analyzes the reasons most frequently advance for MT usage, categorizes the selection criteria used by practitioners to determine their choice of rules-based or statistical engines, and reviews future business-driven extensions of MT strategies intended to increase the return on MT investment. Attendees will learn what they need in order to build a business case for introducing MT to their organizations, whether they choose to implement it behind the firewall or work with translation agencies and other language service providers.

This tutorial seeks to drill a hole in the wall that seems to divide the MT development community from the translation community and, specifically, the post-editing community. The tutorial will involve contributions from three people who are experienced in post-editing, but from different perspectives, i.e. that of the publisher, that of the translator trainer and researcher and that of the post-editing practitioner.

You may register on site for this tutorial. All participants must bring their own laptops to the tutorial. All participants will also need to have a Google account. If you have one already, you can use this. If you don't, we would ask you to please set up a username and password in advance.

This tutorial is for researchers and developers who are working on Computer Assisted Technology (CAT), and want to know more about how translators do their work, and what they need from such technology. It will teach them how they can achieve this goal using a simple Human Computer Interaction Technique called Contextual Inquiry.

The tutorial will be highly interactive and allow attendees to get hands on experience with the technique, by observing and actual translator doing his work. Attendees will also get to ask probing questions, and collectively analyse what they saw and heard during the interview, in terms of what it might mean for CAT development. For many developers and researchers in the audience, this may be their very first exposure to the work of actual end-users, and this in itself, has been shown to produce long-lasting and valuable changes in developers attitudes towards end users (Beyer and Holtzblatt). In addition, attendees will know the basics of how to carry out Contextual Interviews, so that they can do it themselves if they see fit.