For everyone’s safety during the Covid-19 pandemic we have put on hold all ‘in-person’ events – but we are pleased to offer the same high standard of education, training, and certification in our online courses. Click here to see a full listing of online options!

Tag Archives: Machine Translation

Machine Translation Master Course Newsletter – Issue 2

Machine Translation Master Course Newsletter – Issue 2

Machine Translation Master Class Newsletter – Issue 2

As the Instructor for the Machine Translation Master Class from The Localization Institute, I am very happy to continue to share with you my ideas about the relationship between humans and machines. In this issue, we will discuss it from a data management perspective.

Why is data management relevant to me?

 

In 2016, Google said in a statement when it corrected a bug translating “Russia” as “Mordor”, “Google Translate is an automatic translator — it works without the intervention of human translators, using technology instead.” (See here)

 

Indeed Google Translate did not have human translators involved in its translation process. But remember, all natural language data comes from humans in their daily life. You might have heard things like “if you hear it enough, you’ll start to believe it”.  The “illusion of truth” effect also applies to machines. A machine will believe what it has seen after looking for patterns in hundreds of millions of documents. You can, of course, try to fix some problems by manually correcting them. Yet, in many cases, in particular in a neural MT system, it is very difficult for humans to manually hit the quantity and complexity that the hidden layers present and thus some features have to be deleted in order to avoid potentially catastrophic mistakes. Therefore, it makes a lot of sense if we can control the quality and quantity of data before feeding it to a machine. Data management is one of the most effective ways to control MT related risks.

 

Why is some data more relevant than others?

Relevance, first and foremost, is based on the comparison. In a localization process, this comparison often happens between your data and the source text. You can compare them from different perspectives. A translator, for example, usually judges the relevance of their reference materials by searching for concepts, words or knowledge about these words that are similar to those appearing in the source text. If a 100-page document did not include any of these, most probably this translator will give up reading. Humans can make such decisions in a split second. Yet it is a daunting task for machines to simulate this process. So typically an MT engine would diligently scan the whole database and analyze the pattern. If a big percentage of data is irrelevant, it is a waste of computing power and you could not achieve your goal. Of course, in the machine world, language data is processed in a different way. For example, neural MT uses embeddings to capture word meaning whereas statistical MT uses n-gram to process corpora. So we cannot judge data relevance only from a human’s perspective. Yet this analogy helps you get a rough picture based on your intuition.

Who is involved in the process of managing MT-driven data?

While IT professionals can communicate your ideas to machines, it is translators, linguists, project managers, and content managers, who can really make sense out of the data from a human perspective. With effective communication that is based on relevant technological knowledge, you will be able to generate a “collective” insight from your team, other teams outside your department, clients, end-users, and last but not least, your machine. This insight will navigate your attention to meet your needs.

Finally, it is important to point out that there are many more aspects regarding data management in an MT deployment process. For example, data quantity, data generated in an interactive MT or an MTPE (Machine Translation Post Editing) process, and data format. It is definitely an intriguing topic we can further explore.

 

Takeaways:

 

  1. Data management is one of the most effective ways to control MT related risks
  2. Data relevance is key to train an MT engine
  3. Communication helps the team make sense out of the data

 

If you want to know more about machine translation, sign up for our next Machine Translation Master Class.

 

Learn More:

 

Mach-Trans-MC-rz

 

If you are interested in learning more about the Machine Translation Master Class please click here.
Peng-cropped

About the Author

Dr. Peng Wang has been teaching, researching and practicing localization in three continents. She is the convener for EDUinLOC, a part-time professor for the School of Translation and Interpretation at the University of Ottawa and a freelance conference interpreter with the Translation Bureau of the Canadian government. Before that, she was a CAT Tools Coordinator at the Graduate Studies in Interpreting and Translation at the University of Maryland. She has chaired the automation/AI track for LocWorldWide conferences since 2020. Her current research interests include human learning vs. machine learning, machine translation risk management, terminology and multilingual data analysis.

Dr. Wang began conducting corpus-based translation studies at the University of Liverpool and later she worked in the Corpus Research Lab at the Northern Arizona University. She has a rich experience of teaching multilingual classes, with students aged from 18 to over 70, in over 10 language combinations, coming from UAE, China, Italy, Spain, Germany, Morocco, Colombia, Mexico, and Haiti, to name just a few. She is an expert in approaching technology in the context of culture and humanities.

 

Connect with Dr. Peng Wang:

Connect with Dr. Peng Wang on LinkedIn: https://www.linkedin.com/in/pengjanewang/.

Contact Us - Machine Translation

    Please select if you would like to register for our mailing list to receive more articles like this.
  • This field is for validation purposes and should be left unchanged.

Disclaimer
Copyright © 2021 The Localization Institute. All rights reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published, and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this section are included on all such copies and derivative works. However, this document itself may not be modified in any way, including by removing the copyright notice or references to The Localization Institute, without the permission of the copyright owners. This document and the information contained herein is provided on an “AS IS” basis and THE LOCALIZATION INSTITUTE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY OWNERSHIP RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Machine Translation Master Course Newsletter – Issue 1

Machine Translation Master Course Newsletter – Issue 1

Machine Translation Master Course Newsletter – Issue 1

As the Instructor for the new Machine Translation Master Class from The Localization Institute, I’ve been passionate about the relationship between humans and machines for many years. Since 2004 when I completed my PhD thesis entitled Harry Potter and its Chinese Translation using corpus linguistic methodology and translation theories, I have constantly been reflecting on how we can convert our thoughts to machine readable representations and make use of machine power. Machine Translation, in particular, Neural Machine Translation, is definitely a fascinating manifestation of human intelligence in machines. Having said that, I would rather say I am a researcher and practitioner on cultures and humanities in the name of machine learning and artificial intelligence, as I consider such models as neural networks are significant attempts to demonstrate or simulate human intelligence whereas my ultimate goal is to work with you to explore what’s deep in our mind and what our common humanity is. After all, machine is part of human, demonstrating the representable and decipherable part of all of us. On the other hand, all the unexpected aspects that machines are not capable of can be considered risks, which will eventually be taken care of by humans. That’s also the reason why we focus on risk management in this Machine Translation Master Course.

How can I manage the risks related to Machine Translation?

All risk management considerations are built on knowledge and experience. To successfully manage MT-related risks, first and foremost, you have to possess basic knowledge about machine translation so that every decision you are going to make is well balanced between your intuition based on your past experience in the industry and sufficient (not necessarily all) knowledge about machine translation and machine learning. This Machine Translation Master Course covers some of this basic knowledge, for example, some fundamental classifications associated with Machine Translation risk management, including:

  • Two basic types of MT-related risks
  • Two basic types of MT-related technology
  • Two primary purposes of using technology

1. How do I classify the risks related to Machine Translation?

There are two types of risks when we implement machine translation systems: intrinsic and extrinsic ones. In order to control extrinsic risks, you have to possess a good knowledge of what relevant intrinsic factors are. Thus it is important for us to understand the basic inner workings of machine translation as well as its supporting technology. In this Machine Translation Master Course, we mainly focus on the architectural designs of three types of MT systems, namely, rule-based MT, statistical MT and neural MT, as well as relevant CAT tools that are directly useful for various Machine Translation deployment solutions.

2. How do I classify the technology related to Machine Translation?

Fundamentally speaking, there are two types of technology: tool-based technology and intelligent technology. Like why we use bicycles, we use tool-based technological tools to help us improve productivity.  Under these circumstances, human intelligence is the key to success. In terms of intelligent technology, on the other hand, humans are more in a position to monitor and correct machine generated results, which in turn supports machine learning and improves artificial intelligence. Does Machine Translation belong to tool-based or intelligent technology? It depends on such factors as how you deploy Machine Translation systems, the relationship between human & machine, and your purpose of using it.

3. Why do we use technology?

Technology can serve both humans and machines. In essence, our ultimate goal is always to have machines serve us better. Yet nowadays we can see more human-machine interaction (HMI) activities have aimed to train machines more than human beings. This poses new opportunities and challenges for us. Do you know that convincing evidence in cognitive science, computer science and learning theories indicates that human brains learn better than any machine… at least for now? So rather than resisting change, it makes more sense for us to empower ourselves to better prepare for the machine-human revolution. After all, it is all up to each one of you!

 

If you want to know more about machine translation, sign up for our next Machine Translation Master Class.

 

Learn More:

 

Mach-Trans-MC-rz

 

If you are interested in learning more about the Machine Translation Master Class please click here.
Peng-cropped

About the Author

Dr. Peng Wang is a part-time professor for the School of Translation and Interpretation at the University of Ottawa and a freelance conference interpreter with the Translation Bureau of the Canadian government. Before that, she was a CAT Tools Coordinator at the Graduate Studies in Interpreting and Translation at the University of Maryland. She was the coach and curator for the automation track for LocWorldWide42. Her current research interests include cognitive interpreting/translation studies and AI, risk management of NMT implementation, terminology and multilingual data analysis.

Dr. Wang began conducting corpus-based translation studies at the University of Liverpool and later she worked in the corpus linguistic program at Northern Arizona University. She has a rich experience of teaching multilingual classes, with students aging from 22 to 75, in over 10 language combinations, coming from UAE, China, Italy, Spain, Germany, Morocco, Colombia, Mexico, and Haiti, to name just a few. She is an expert in approaching technology in the context of culture and common core humanity.

 

Connect with Dr. Peng Wang:

Connect with Dr. Peng Wang on LinkedIn: https://www.linkedin.com/in/pengjanewang/.

Contact Us - Machine Translation

    Please select if you would like to register for our mailing list to receive more articles like this.
  • This field is for validation purposes and should be left unchanged.

Disclaimer
Copyright © 2021 The Localization Institute. All rights reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published, and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this section are included on all such copies and derivative works. However, this document itself may not be modified in any way, including by removing the copyright notice or references to The Localization Institute, without the permission of the copyright owners. This document and the information contained herein is provided on an “AS IS” basis and THE LOCALIZATION INSTITUTE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY OWNERSHIP RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

--