Category Archives: MT in business

“Hasn’t it always been there?” – machine translation for patents

I work with patents and I use machine translation at work very regularly, maybe every other day. Mostly I use MT to read patents and patent applications that are in other languages. We have a patent search database where we retrieve the patents and that database has a machine translation function built into it, so it’s very easy to use.

If we have an idea for a new invention, the first thing we have to do is make sure it has not already been invented by someone else. That’s why we need to read through other patents and patent applications – we’re looking for the ones that are closest to our own invention or product, those are the relevant ones. The others we can ignore. Many of the ones we read are already in English, but there are also quite a few that are in some other language, and those are the ones we read through machine translation. We also might read other patents to make sure others are not infringing on our patents (or vice versa), to keep up with what’s going on in our field, or to check on the IPR of other companies.

The languages we are translating from most are Chinese, Japanese and Korean. China is currently the largest producer of patents, by a lot, but Japan and Korea are also high on the list. Then sometimes we translate from German and French. I know some German but none of those other languages. For German patents I often check the original document as well as the machine translation, putting together both to get an understanding. 

If we don’t understand the machine translation well enough, we can resort to other methods. As I mentioned, the patent database has its own machine translation function. But it also offers an alternative, Google Translate, if the first translation we get isn’t good enough for us to understand. And if that’s still not enough to be able to see if the patent is relevant or not, we can order human translation for whatever piece of text we need. We also might order human translation when a patent is clearly very relevant and important to our work, so we have to be sure we have a very good understanding of it. And of course, if we need to use some information from a patent in a legal setting, we don’t use the machine-translated version, we get it translated by a human.

Machine translation saves us time. In situations where you don’t need to understand every detail, it really is sufficient. You can get some kind of understanding quickly.

What was work like before we had machine translation? Hmmm…hasn’t it always been there?! I have to say, I feel like it’s always been there. I think we’ve used it…at least since 2010. Our way of working has changed so dramatically over the years. The work has grown and gotten digitalized, and the MT was there and you just kind of noticed it and started to use it and it just…snuck up on you like a thief, now that I think about it.

Our new employees learn to use machine translation while learning to use our patent search database. And when they first start reading the machine translations, they just laugh. The sentences and the structures just look a bit funny. But once you get past that, then you get used to the level of the language and you learn to pick out the important points. Also, patent publications have a certain structure and there’s a certain format that they’re in. That makes them easier to follow.

And yes, we also make decisions based on machine translations, even though there are risks. We have to. It would simply not be possible to do all of this with human translations. Plus there are many other risks in our work – the first one comes when you go searching for relevant patents, you might use the wrong keywords or miss something. Then there are risks when using MT. But this is how our industry works, everybody uses machine translation to screen patents in other languages. Then when you’ve found the most relevant patents, you weigh the risks, importance, and cost, and decide what to do. Basically, the more risky the situation is, the less you are going to rely on machine translation.


patent text

Screen shot of an example patent picked by me (as you can see by the topic of blogging software) from the European Patent Office’s patent search database. The text in the background is the machine translation and the text in the pop-up blue box is the original paragraph in Chinese.


Interested in learning more about users of patent machine translation? My first academic paper on the group can be found here.

Nora, young world traveler

One of the most comical situations I’ve been in with machine translation was when a colleague and I went to see a local tourist site in a part of China where people don’t tend to speak English. We were trying to buy tickets with the help of Google Translate but for some reason, no one could understand what we wanted. I think they were telling us where we needed to go, and we just kept repeating that we want to buy tickets. Then more and more people gathered around to help and we were all typing furiously on our phones. After 15 minutes we finally managed to get the tickets.

GT screenshot_buytickets

It was in China that I originally got the idea that machine translation would be helpful while traveling. Some colleagues and I were trying to work out something in Chinese when one of them pulled out her phone, translated, and we quickly figured it out. Before that I’d used Google Translate rarely, just to check individual words while I was studying or something.

Nowadays I use the Google Translate app very often. My work in the international travel industry means I travel to a lot of foreign destinations, spending 1-5 days in each. In many locations I get by just fine with English, but especially in China and Japan, I find myself pulling out my phone and using Google Translate many times a day.

One of the main places I use it is in restaurants and it seems to be a very common thing to do that. Especially in China, the minute they realize we don’t share a language, they pull out a phone and start trying to communicate through machine translation. In Japan it’s kind of the opposite – they won’t be the first to start trying with Google Translate. I think they’re quite shy, and maybe a bit embarrassed that they don’t know English. The Chinese seem to want to understand so they jump right in and don’t worry about potentially embarrassing situations.

I actually know a little Chinese so might get by on that and body language, but I also eat vegetarian so often need to explain things a bit more. If I try to say ‘No meat!’ with my few words of Chinese, they might only understand the word ‘meat’ and bring me a meat dish. (In fact that has happened and I was presented with octopus, beef and shrimp. Perfect.) Better to translate what I’m trying to say on the phone.

Most of the time I type what I want to say into the app and then it translates. If the person looks confused then I try writing it again in a slightly different way. Once in a while I use the voice thing where you talk into the phone’s speaker and it translates, but mostly I like the typing better. Somehow with the voice thing I’m less sure if it’s correct or not. Sometimes I use both: first let them read the translated text, then listen to a voice translation.

I also use the Google Translate app in stores to read labels and signs. One of my favorite things to do abroad is to visit local grocery stores, I love spending time in them. And, of course, since I’m often at a destination for several days, I need to buy some very normal, everyday things. Or I might shop for clothing and then I want to good, natural materials. What I use in these situations is that translation app function where I take a photo of a text – a label or a sign – and it tries to translate it for me.

The app has even been helpful at work on occasion. Like when a customer speaks a language I don’t know, or even a dialect of a language that is hard to understand, but they really need something. Then we have sometimes used Google Translate to communicate and resolve their issues.

I would say that my success rate with using apps to communicate with other people is fairly high, maybe 80-100% Usually I get what I want and they understand me, at least I think so. With the picture-taking method I’d say the rate of success is more like 50/50. Sometimes I get a clue what something’s about but sometimes…it’s just random words. The photo thing is just not that good yet.  

I always use English with these machine translation apps. I just think that my own language is not a good language to translate with. It seems to me that English is the key to machine translating. Or should I say it this way: if you understand English, it’s much easier for you to use it.

Gustav: “With machine translation I can contribute better, also in places where I’m not directly asked.”

For the first several years I lived in Finland, I worked in a large, very international company and English was the main language used. Even though that is not my native language, Swedish is, my English is very good. But a few years ago I changed jobs and my new job is in a truly Finnish company with only 3-4 non-Finns working there. Finnish is not a requirement (luckily, or they wouldn’t have hired me), but of course people are more comfortable with Finnish. I think this is a great thing actually, I see it very much as an opportunity for me to learn Finnish.

Some of the texts I need to deal with at work, e-mails and documentation on the software products we make, are in English half the time and Finnish half the time. There is no strict language policy. Other texts I need to understand are always in Finnish, like human resources kind of information – things like, what’s the company’s travel policy? What is the procedure I need to follow to take parental leave?

My Finnish is OK-ish. I find it hard to follow spoken Finnish. But with written things like e-mails or instructions, I can usually work out the basics of what I need. The problem is when I need to understand the details. Then I can very often get lost. So what I do is use Google Translate to translate the parts that I don’t understand. Sometimes I do it to get confirmation that I have understood things correctly.

I mostly translate from Finnish to English, not Swedish, because generally I find it works better than Finnish-to-Swedish. Every once in a while if the English translation is iffy or I don’t understand it, I might try translating into Swedish.  

I would say that machine translation works surprisingly well. I use it pretty much daily and 80-90% of the times I use it, it gives me the information I need. I actually work with speech and language technology and I’ve noticed that in the past couple of years, there have been amazing advances in machine translation in terms of readability. I have been using it more and more since I noticed this. It helps that I know the general gist of things when I translate a text – I’m confident that I can assess whether the translation makes sense or not.

When it doesn’t work – I don’t understand something I’ve translated – I go back to the original text in Finnish and simplify it. The original might have little mistakes in it that I correct, or I simplify the content and structure a bit. And then I put it through machine translation again. This often helps. When it doesn’t, I either ask a colleague for help or I simply decide that the text is not that important so I ignore it.

Machine translation really helps me in getting the missing pieces from everyday e-mails and documents. I might get an e-mail with a long discussion thread, all in Finnish, and finally someone forwards it to me to see if I can help with the solution. With the help of Google Translate, I can get a better understanding of the thread and the context of the problem, and then I can answer more questions and answer the right questions better.

What would I do if I didn’t have the help of machine translation at work? Well, I would probably be more blind to the context of things. I might end up ignoring some things, and I suppose I might end up being less cooperative in a way. I would get away with being more in the background. With the translation I can contribute better, also in places where I wasn’t directly asked. 

On a larger scale, one thing that surprises me is how little visible impact machine translation has had on businesses. Take web shops – you rarely see web shops from other places in Europe that have their pages machine translated. You rarely see them available in, say, 25 languages. It seems to me that businesses, even small ones, could be selling across Europe more than they are now. Machine translation could help.

Gary, developer of MT systems

I use Google Translate to translate Chinese spam messages which regularly get posted on my website – just in case any of them are of interest!

Actually most of my interest in MT is not about free online MT tools like Google Translate, but about professional tools. I created my first rule-based MT system over 25 years ago and have been involved in developing machine translation since then. My main project is a commercial neural MT system for translating  between Dutch and English. But I also experiment in neural systems for other languages, most recently Indonesian and Turkish.

Since my own knowledge of Indonesian and Turkish is very poor, I’ve found a use for online MT in this experimental work too. For example, I could translate something into Indonesian with my own engine, then back-translate it through Google into English to see how well my model has learned the patterns of a new language. Other times I simply translate the same thing using both my own and Google’s engines and compare the results. It helps me check that my neural network building is going in the right direction.

I really do also Google-translate Chinese spam messages. The translations are usually very basic but tell me what I want to know about the spam – mostly that this is not something in which I could be remotely interested. Let’s not forget that before the arrival of online MT, companies would spend large sums of money each year translating documents that turned out to be completely useless. Online MT saves money.

In general I’m happy to recommend MT to friends for general information purposes: double-checking the translation of a menu in a Chinese restaurant, translating the instructions sent with self-assembly furniture, or reading spam. Most general online translators are quite capable of fulfilling such tasks nowadays. I also recommend MT as a language learning tool.

However, I would not rely on unchecked, unedited MT for information I need to make important decisions. Apart from the obvious fact that the machine translation may be fluent yet incorrect, we have to allow for the possibility that a decision may be challenged in the courts. (Example case. See MT use in the news for more links to news stories on the use of MT).

One thing I’d like people to understand and remember is that the best results are obtained when the MT system is “trained” to handle the type of texts being translated. In online tools like Google Translate, this is not the case.

Krista, language and communications professional

My English, French and Swedish are all strong and I studied them at the university, but German, Italian and Spanish are languages I haven’t studied since high school. I have a university degree in translation and I worked as a professional translator for some years, but for the past 15 years I’ve been more of a communications professional. I am the Managing Director of a national non-governmental organization and my job consists mostly of communications.

Rich use of Google Translate

I’ve been using Google Translate for many years now. Not in translation work, of course, but for a variety of other things. First of all, I use it to help me make sure I understand texts that are in languages I am not fully fluent in. I read articles, newspaper stories and research papers that have to do with my industry and are in a variety of languages. When I come across a passage I’m not sure I understand, I google-translate it into my native language of Finnish and also into English (usually both, not sure why…maybe curiosity? Or because I love languages?) I also want to know what similar organizations to mine are doing around Europe so I find myself google-translating web pages or other documents to be able to read them.

Sometimes I use the tool to help me when I have to write something in Swedish, which I struggle with a bit. I might write a piece and test it out in Google Translate to make sure I got the phrasing and spelling right. I also test it in regular Google to see if it’s a common way of expressing what I’m trying to express.

I guess the funny thing is that my most frequent use of Google Translate is not really for its translation function. I most often use it as a dictionary. I don’t have a good electronic dictionary at my disposal and I’m a bit lazy about looking things up in books. So I either google the word or use Google Translate to find the meaning. I don’t really look at the translation but down below, where they give dictionary-type information. Here’s an example:

GT dictionary

I like that it gives me multiple suggestions for a translation. I couldn’t ever see myself simply taking the 1 translation it gives in the box and using that. More context helps me make sure I understand.

This also works very well as a thesaurus, a bilingual one at that. It gives you optional synonyms in both languages. And one more thing, a feature I noticed just this year! The megaphone icon at the bottom of the word boxes? That is a pronunciation audio tool.

These kinds of language tools are included in many online dictionaries in different languages, but this is an easy option for multiple languages. Quick.

Oh, and I don’t know if you’ve noticed or not, but they want people to help develop Google Translate. I’ve done some volunteering there, mostly in English since that’s the easiest. You click through expressions and words and see if they’re correct or not. It’s kind of a game and helps to develop it. It’s fun.

Double-checking

Google Translate is hardly ever the only thing I use to verify the meaning of a text; normally I double-check it some other way. Even though it’s just for my own use, I work hard to get a good translation.

First of all, I only use it with languages I know well or at least a little. By comparing the original text and the translation, I know enough to know if the translation is something it’s supposed to be. I would never use it with languages like Arabic or Chinese because I would have no way of knowing how ‘right’ the translations are. I also try out translations with different language combinations, often including English somewhere. Because I have background knowledge on how machine translation works, I know that it works best when English is one of the languages used.

Another way I verify things is by putting them into regular Google. If I can see the same phrase and how it’s used in different contexts, then it can help me fully understand it. Usually when this happens it’s because I’m reading things that are not as familiar to me, like academic or technical texts.

The question of machine translation

I don’t go around recommending the use of machine translation to other people, no. Most often that’s because people need a translation for a text they want to give to someone else or publish. Then I recommend that they use a human translator, of course, because that is what is needed. I think it’s important to make sure people understand that. Another reason I don’t do it is that, if I were to recommend it to someone outside of the language industry, I can’t be sure that person understands the limits of machine translation.

If someone asks me for advice on using it, say they just want to understand a short paragraph, then I tell them it is useful for that if they understand the risks involved. But I’ve never recommended it as such.

When I first started using machine translation it was a bit of a shameful thing. The attitude in the translation community was definitely negative. I didn’t dare tell people I was using it. Nowadays it seems more acceptable for the purposes I use it for. I suppose in earlier years, up to 2010, people were afraid that machine translation would take all the work from translators and eventually they wouldn’t be needed. But now I think people understand it’s just a tool that can’t do everything we need humans to do. I noticed the shift in attitude sometime around 2014 or 2015. It was discussed in conferences in a different way and people were much more interested in the topic.

 

Dr. C. Koby, Physicist*

Oak Ridge National Laboratory, Tennessee, USA
October 15, 1971

Part of my work as a physicist is to keep abreast of research being carried out in my field. I try to follow activities in all parts of the world, but I am particularly interested in the research being conducted in the Soviet Union. I would guess that something like 15% of the articles and books I need to read in my field are written in Russian. I have no competence in Russian so have to rely on translations to access those materials.

A very positive development in the past few years is a service that my laboratory offers which machine translates articles from Russian to English. They don’t have anyone correct the articles,  you get what the machine puts out. I have personally used the service 10 or so times over the past 3½ years.

I have one simple reason for using the MT service: it is fast. After a request, I can expect to get a machine-translated article back in 1 month, much more quickly than the 9-10 weeks some colleagues report waiting for articles translated by humans.

More time and effort is required to understand machine-translated texts as compared to those written in English – I’d say about double the time. One reason for that is that formulas and diagrams from the original articles do not get translated, so I usually have both the machine-translated text and the original Russian article, and I go back and forth between them. In general, however, I’d say that the quality of the translations is acceptable. In a funny way, over time you get used to the style. You find yourself mentally correcting awkward passages.

I think it’s important to bring up the fact that the machine translation is understandable to myself and my colleagues mostly because we are familiar with the contexts being discussed. We know our own fields well and that allows us to accommodate texts that are not perfect. It is also important that technical terms are translated correctly or at least understandably, and the majority of the time, they are. The most common problem is words that are left untranslated. I don’t know why that happens but it does, and it can affect understanding.

I’ve been asked if it would be possible for someone to completely misunderstand the machine translations and arrive at the opposite meaning from the original. Yes, sometimes the meaning is a bit distorted, but since I know the context of what I’m reading so well, I do not think that it is possible to arrive at the opposite meaning. I have never heard of it happening anyway.

Overall, I’d say that I’m happy to use the service. It allows me to read things I would normally have to wait far longer to get access to. I would, and actually have been known to, recommend the use of machine translation to my colleagues.


*       Instead of being about 1 real person, the persona in this story is a composite that is based on a group of 58 real people who were surveyed by Bozena Henisz-Dostert in the early 1970s (see reference below). The group consisted of some of the first users of one of the first machine translation systems: the Georgetown Automated Translation system, developed in the 1950s and early 1960s at Georgetown University. Although its development was discontinued in 1964 due to lack of financial support, the Georgetown system continued to be used, virtually in its 1964 state, for a decade or so by 2 groups of scientists in Oak Ridge, Tennessee and Ispra, Italy.

The persona in the story was compiled in a straightforward way. I used 23 of the survey’s 51 questions. For each question, I took the most popular answer, or the mean or average of the figures given in answers, to describe Dr. Koby. The name ‘Koby’ came from an internet name generator and does not reflect any one true person.

The story is based entirely on information from Bozena Henisz-Dostert’s article ‘Users’ evaluation of machine translation’ in the book Machine Translation by Henisz-Dostert et al. 1979.

 

Max, manager at a large multinational company

Around 2010 my company acquired a factory in Brazil. Although there were a few native Spanish speakers on my team, no one knew Portuguese. Our Brazilian colleagues did not speak English. I decided I would start studying Portuguese because I had some responsibilities in the acquisition project.

While I was still learning the language, I had a lot of documents I had to review which were in Portuguese. I heard from colleagues about Google Translate and I thought I would try using it to help me with those documents, just to be able to get some idea of what is covered in them. I am by no means a techie kind of person, I’m not that interested. To me, applications are just tools you use to get things done. Google Translate was one of those.

I realized pretty much right away that there are advantages and disadvantages to Google Translate. Use of a machine translation tool is not the same thing as proficiency in a language. But if you understand that and how to use it as a tool, then it is helpful. I ended up using Google Translate for several years while learning Portuguese.

One way I used it was to translate documents from Portuguese to English, to get an idea what they were about. I translated articles in trade magazines, economic reports, excerpts from longer things like books. Things that were not that demanding technically. I would copy/paste an excerpts or whole articles for translation. I never translated anything like legal documents. I seriously doubt Google Translate would produce anything other than harm in those cases.

I usually understood the topic of the things I translated, what they were about. Something about the substance. The thing is that if you understand the context, you can take what the machine translation gives you and connect the dots and bridge the gaps. You get some building blocks and put together the whole picture yourself.

If I didn’t understand something, I could always call a colleague and ask them for clarification. We had very good cooperation with our Brazilian colleagues and they were very appreciative that I was trying to learn their language. They were helpful.

Another way I used the tool was in learning Portuguese and building my vocabulary. I would check the translations of single words in Google. Why didn’t I use a dictionary? Because Google Translate is faster. And when you look up a single word, the translation is almost always correct.

I also used it to check forms in Portuguese, to see if they were correct. I used it along with other language learning materials. When I was trying to produce Portuguese, I would check the form in things I was writing to see if it made sense. That was for informal things like e-mails. For more important things I asked colleagues about things I was unsure about.

I never made decisions based on the information I used Google to translate. Goodness no. Not for big strategic decisions or even smaller ones.

Thoughts on using machine translation

Machine translation does not equal proficiency in the other language. Nor is it a shortcut to proficiency. People are so used to getting things immediately. But it’s not that easy.

You have to have some critical thinking and understanding of the limitations of the tool. When you understand the theory and why things happen the way they do, then use of a tool like Google Translate is nice, neat and handy. I think of it like this: if you’ve studied medicine for 5 years and use information on the internet to help you diagnose symptoms you’ve been having, that is a totally different thing than a person with no background who googles their symptoms.

It is also helpful if you’ve studied the other language involved (the one you’re translating from) and know something about what it’s about. For example, you can spot wrong synonyms that the machine produces. If I were translating from a language I have no idea of, that would be a very different thing. For me it was important to be able to reflect on my own language capabilities – there’s always some kind of reflection on the language.

It should never be used blindly, with the user just accepting what machine translation gives them.