22

Who claimed that machine translation is as good as a human translator? For me, as a professional translator who makes his living on translation for 35 years now, MT means that my daily production of human quality translation has grown by factor 3 to 5, depending on complexity of the source text. I cannot agree that the quality of MT goes down with the ...


7

Google's translations can be useful, especially if you know that the translations are not perfect and if you just want to have an initial idea of the meaning of the text (whose Google's translations can sometimes be quite misleading or incorrect). I wouldn't recommend Google's translate (or any other non-human translator) to perform a serious translation, ...


6

You have asked quite a lot of questions, some of which cannot be answered definitively . To give an insight of the quality (and its history) of machine translations I like to refer to Christopher Manning his 'one sentence benchmark' as presented in his lecture. It contains one Chinese to English example which is compared with Google Translate output. The ...


5

I will be starting my PhD in natural language processing in a few days and this is very similar to my proposed topic. It's an open problem that ties NLP and AI into philosophy of science and epistemology and is, I think, extremely interesting. I say all this to drive home the point that this is not a simple problem. Two major theoretical concerns come to my ...


5

It's not quite clear what you are asking. So I'll answer in separate parts. Why is the translation different from the official title? It could be simply because machine translation is not perfect, or our human translator took some creative liberties when translating. In this case it seems to be both. Note that 龍爭虎鬥 properly translated doesn't mean ...


4

It really depends on the language pair and the topic of the content. Translating to/from English to any other language usually is the best supported. Translating to and from popular languages works better, for example, translating from English to Romanian is a poorer translation than English to Russian. But translating from English to Russian or Romanian is ...


3

James Ryan has done a lot of 'archaeological' work on this; you can find references to his work on his website. Story generation has been a dream for a long time (in computing terms), and various genres have been explored, with not that much success. There have been episodes of a Western written by a computer (and actually filmed and acted out by human ...


3

Jarvis was built using the suite of tools that facebook developers are constantly updating. The answer to this question is that there's no simple answer; it has a lot of moving parts. Take for example natural language processing. There are a number of sub-topics that are each considered "big" problems, such as part-of-speech recognition, ...


3

One 'easy' way would be to have some sort of conversational memory, where you track what the user has said already. I don't know how complex your patterns are, but if you could recognise names and track references, you could try and build up a mental model of the user's relationships with other people, and perhaps refer to that in your bots responses. The ...


3

This is still a research topic in linguistics. A quick google search brings up a couple of papers that might be useful: Identifying Metaphor Hierarchies in a Corpus Analysis of Finance Articles Metaphor Identification in Large Texts Corpora However, you probably won't get an off-the-shelf tool that recognises metaphors for you. To add more details, the ...


2

As you know google translation works base on statistical methods. In statistical translation, many parameters can be related to the final result. One of these parameters is co-occurrence of words in a sentence. Hence, as this translator learn languages from different utterances by human and pre-written translations, and different parameters in the text are ...


2

The VR and computer speech aspects are entirely corollary. Adding them in would be relatively trivial in comparison to creating an algorithm that can dynamically generate stories of interest to humans. Essentially, aesthetic components not related to story structure (images, sounds, speech) are "window dressings". A story generation algorithm would have ...


2

I suggest you take a look at the syllabus of http://cs224d.stanford.edu/syllabus.html and see what was invented in the last ~20 years. They focus on deep learning methods, and they include there almost everything important. Among the things, which I would include into the review are: Word2Vec models Seq2Seq models Concerning what happened in the last 5 ...


2

There was a lot of work on this topic at UT Austin, which has now migrated to the Alan Institute. There is no off-the-shelf software that will answer your question (if there was, DARPA would stop funding its development!), but you can read about the latest development in a number of recent papers. This paper (Seo et al. EMNLP 2015) discusses the techniques ...


2

Apologies for not writing in English. Please find the adapted translation here: To give interested people an idea of the quality of MT (DeepL) please see this example from a text I was working on this morning (6,300 words, started at 9 am, delivery today around 1 pm and still find time for this post). I was working on this sentence (201 words) when I ...


2

Am I wrong and Google's translations are nevertheless readable, helpful and useful for a majority of users? Yes, they are somewhat helpful and allow you to translate faster. Or does Google have reasons to retain its greatest achievements (and not to show to the users the best they can show)? Maybe, I don't know. If you search for info, Google does ...


2

According to a nice article by Sebastian Ruder https://ruder.io/4-biggest-open-problems-in-nlp/ based on answers from top NLP researchers https://docs.google.com/document/d/18NoNdArdzDLJFQGBMVMsQ-iLOowP1XXDaSVRmYN0IyM/edit Natural language understanding NLP for low-resource scenarios Reasoning about large or multiple documents Datasets, problems, and ...


1

I'd suggest BERT for this. It is essentially a word-embedding model that uses at local context to determine the appropriate embedding for each word. This means it would assign "bat" a different embedding in a sentence containing "hit the ball" vs. a sentence containing "flies and eats bugs". On top of that, Google has released a ...


1

DALL·E is a 12-billion parameter version of GPT-3 trained to generate images from text descriptions should be the same data they used to train the GPT-3


1

This is a difficult problem. First, how do you define 'subject'? Do you have a (closed) lists of labels you want to assign? What about subjects that overlap, or don't occur in your list? What even is a subject? This is a non-trivial issue. Second, and this is even harder, how do you want to recognise subjects? A simple solution could be using a list of ...


1

While your question has some ambiguities, I try to answer. From my understanding you want your model to predict “topic” of a sentence or a description. It’s just a classification problem with huge possible number of output classes. The first initial issue is very short length of documents (sentences). Most of topic modelling algorithms such as LDA have ...


1

They're all important. NLP is an umbrella term that includes the other two; NLG is only concerned with generating language, ie transforming some internal data structure into human language. NLU is about processing information contained in language, and putting it into relation with a knowledge base etc. If you don't know anything about any of these fields, ...


1

I can see several challenges, and the list below is not exhaustive: i. The main problem is how to model a problem of translating a language test into a formal language. It will probably be something like the automatic translators, but with some guarantees that the proof semantics will be preserved. If you are more interested in this path, I recommend ...


1

Surprisingly all the other answers are very vague and try to approach this from the human translator POV. Let's switch over to ML engineer. When creating a translation tool, one of the first questions that we should consider is "How do we measure that our tool works?". Which is essentially what the OP is asking. Now this is not an easy task (some other ...


1

This will be not so much an answer as a commentary. The quality depends on several things, including (as Aaron said above) 1) the language pair and 2) the topic, but also 3) the genera and 4) the style of the original, and 5) the amount of parallel text you have to train the MT system. To set the stage, virtually all MT these days is based off of parallel ...


1

Well this is a relatively new problem very tied to Question Answering. One of the recent systems is EUCLID that can answer those type of question the public Dolphin algebra question set by using a tree transducer cascade approach. This paper details the proposed model Hopkins, M., Petrescu-Prahova, C., Levin, R., Le Bras, R., Herrasti, A., & Joshi, V. (...


1

Google uses user input to improve translation. Some user may have provided an input to the Traditional Chinese characters using English characters only instead of pinyin, which would introduce a mistake into the data used by the translator. Since the model is statistics based, such a mistaken translation can only be assigned a lower but non-zero probability ...


1

Chinese words 'fu' means with different intonation marks either happiness, huspand or tiger. Without correct intonation notation in English it may translate in Chinese as Tiger. Movie title has Chinese 'happiness' character, but Google mixes it as Tiger.


Only top voted, non community-wiki answers of a minimum length are eligible