Need help for Machine Translation Questionnaire
Thread poster: Agasta
Agasta
English
Apr 14, 2006

At present I’m writing my MA thesis on Machine Translation. Right now I’m working on its practical part. In order to complete it I need some people to fill in the questionnaire.
If you find the topic of my thesis interesting, answer the following questions and send it on my e-mail address as soon as possible.
Thank you
Agata

agasta@wp.pl

Questionnaire on Machine Translation
The following questionnaire remains an attempt to obtain some information about users’ opinion on the concept of machine translation and aims at the identification and examination of certain problems stemming from the use of computerised translation.
In some of the questions more than one answer is possible.
1. Gender:……………
2. Age:………………..
3. Which of the following statements, in your opinion, describes the notion of machine translation best?
A. machine translation remains a useful facility that substitutes humans in the task of translating
B. machine translation remains an effective facility only with the assistance of human translators
C. machine translation offers practical help for the translator but does not complete the task of translation successfully
4. Which of the following MT aids do you use in your everyday life?
A. on-line dictionaries
B. term banks
C. glossaries
D. word processors
E. translation workstations
F. others:……………
5. Which of the following MT systems have you ever operated?
A. Systran
B. Logos
C. Susy
D. others:……………
6. Which of the following factors encouraged you to use the MT systems?
A. production of translation in a short period of time
B. production of translation at a lower cost
C. production of good deal of translation
D. the production of translation covering various subject domains
7. What are the purposes for which you use the MT systems?
A. machine translation for professional translators
B. machine translation for non-translators
C. machine translation for foreign language learners
8. What is the way you use machine translation?
A. machine translation as a translation aid for full-quality human translation
B. raw translation for information scanning
C. raw translation as a drafting aid, i.e. use of MT for translating rough versions of a text from the target language into the source language
9. Which of the following techniques do you use for revision of translation?
A. pre-editing, i.e. checking the input text and adjusting it to satisfy the expectations of its receiver
B. post-editing, i.e. revising the output text and error correction
C. none
10. What is the greatest advantage of machine translation?
A. production of translation in a limited period of time
B. translation of large amounts of material
C. ability to maintain consistency while translating large volumes of text
D. possibility of translating documentation into more than one target language
E. production of draft versions of scientific and technical documents for finding out the general content
11. Are the systems effective in the process of translating?
A. yes
B. no
12. Which of the following factors make the MT systems ineffective?
A. poor quality of translated versions
B. need for revision
C. no cultural implication
D. narrow variety of available language pairs
E. need for regular and costly dictionary maintenance and updating
F. inability to produce reliable but low-cost translation without engagement of expert bilingual translators
13. By means of which techniques do you evaluate the quality of translation?
A. accuracy, i.e. whether the information from the final version is compatible with that of the original
B. clarity, i.e. whether the translated text can be easily understandable
C. style, i.e. whether the language used in the translation is suitable for its content and purpose
D. error counting and detailed evaluation of mistakes
14. Which of the following errors frequently appear during the system performance?
A. grammatical errors
B. lexical errors
C. stylistic errors
D. punctuation errors
E. omissions
F. incompleteness
15. Which of them affect the quality of translation most?
A. grammatical errors
B. lexical errors
C. stylistic errors
D. punctuation errors
E. omissions
F. incompleteness


[Subject edited by staff or moderator 2006-04-14 17:00]


Direct link Reply with quote
 
xxxmediamatrix
Local time: 10:54
Spanish to English
+ ...
Some pointers .... Apr 14, 2006

Hi Agasta,

I don't wish to be discourteous, or discourage worthwhile research, but I do think you would be well-advised to:

A Define your interpretation of 'machine translation' which means all things to all men (and women, of course...). Your questionnaire seems to be mixing many topics, some of which have little to do with MT as I understand it.

B Have your questionnaire verified by an English speaker, preferably with experience in the formulation of questionaires, to ensure that your questions are understandable (some are not) and unambiguous (some are not); in its present form many of your answers will not be comparable because we the guinea-pigs will be making different assumptions about the intended meaning of your questions.

C Present your questionnaire in a form which minimises the effort needed by your guinea-pigs; you would do well to have a look at how other students have presented questionnaires here to get some clues on this, e.g.: www.proz.com/topic/42324 and www.proz.com/topic/44050

D Finally, please say 'please', at least once in your posting.

MediaMatrix


Direct link Reply with quote
 
xxxIanW
Local time: 16:54
German to English
+ ...
Two points Apr 14, 2006

I'd like to make two points, if I may.

Firstly, rather than copying the unformatted contents into a forum, why not ask people who are interested in taking part to contact you, and then send a nice, formatted Word file or similar?

Secondly, I think you'll find that language professionals such as ourselves are not the target group that use machine translation, for the same reason as taxi drivers tend not to drive Ladas.


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
MT questionnaire topics are all relevant Apr 14, 2006

mediamatrix wrote:

I don't wish to be discourteous, or discourage worthwhile research, but I do think you would be well-advised to:

A Define your interpretation of 'machine translation' which means all things to all men (and women, of course...). Your questionnaire seems to be mixing many topics, some of which have little to do with MT as I understand it.


Hi MediaMatrix,

I have read through the MT questionnaire carefully and have found all points to be fully relevant to the topic of Machine Translation (MT) and translation assessment measures of MT output.
The references on defining MT would be at websites of the European Association for MT (http://www.eamt.org) and the Association for MT in the Americas (http://www.amtaweb.org) and the Asia-Pacific Association for Machine Translation (http://www.aamt.info/) which are the 3 regional subdivisions of the International Association for MT (IAMT).

I don't see what points seem to you to be inconsistent with the topic of MT.

Jeff Allen, PhD
Certified PROMT MT dictionary builder
Certified Reverso (v3-5) MT dictionary builder
Expert SYSTRAN MT user
Tester of other MT and TM tools
http://www.geocities.com/jeffallenpubs/


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
language professionals do use MT Apr 14, 2006

Ian Winick wrote:

Secondly, I think you'll find that language professionals such as ourselves are not the target group that use machine translation, for the same reason as taxi drivers tend not to drive Ladas.


Ian,

I have a list of many professional translators who use MT tools + dictionary building + MT postediting on a regular basis for their translation projects. You just have to dig a little bit to find out who does. And I know a number of them who do not advertise that they use it. This is to avoid being tagged by others of possibly producing poor work, but they do produce the same high-quality output in the end and get paid for it.

Jeff

======
Jeff Allen, PhD
http://www.geocities.com/mtpostediting/


Direct link Reply with quote
 
xxxmediamatrix
Local time: 10:54
Spanish to English
+ ...
Some of my 'quibbles' ... Apr 15, 2006

Jeff Allen wrote:

I have read through the MT questionnaire carefully and have found all points to be fully relevant to the topic of Machine Translation (MT) and translation assessment measures of MT output.

I don't see what points seem to you to be inconsistent with the topic of MT.


Thanks for your comments, Jeff. I hesitate to differ with an acknowleged expert in MT but, for what it's worth, here are just a few of the 'inconsistencies' I found on a first reading of the questionnaire, and now confirmed on a second and third reading, and having looked (again) at www.eamt.org:

Q3 Options A, B and C refer to CAT, not to MT as such.

Q4 Option D is a 'writing aid' and not, in itself, an MT or even a CAT tool.
The other options may be used in CAT or MT, with entirely different results in each case.

Q8 Options A and B are MT (by EAMT's definitions) and Option C is CAT.

Q9 Option A - The definition of pre-editing is incorrect (pre-editing is the adaptation of the source text to increase the likelihood of it being undertood correctly by the MT system).

Q10 greatest advantage? - compared to what?

Q11 effective? - in terms of what? (cost, time, accuracy, ROI, ....)

Q14/Q15 what about 'flagrant errors of comprehension' ?

MediaMatrix


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
replies concerning issues with the MT questionnaire questions Apr 15, 2006

Thanks MediaMatrix for the comments. I've provided my replies below:

mediamatrix wrote:
Thanks for your comments, Jeff. I hesitate to differ with an acknowleged expert in MT but, for what it's worth, here are just a few of the 'inconsistencies' I found on a first reading of the questionnaire, and now confirmed on a second and third reading, and having looked (again) at www.eamt.org:[/quote]

mediamatrix wrote:
Q3 Options A, B and C refer to CAT, not to MT as such.


agata wrote in the questionnaire:
3. Which of the following statements, in your opinion, describes the notion of machine translation best?
A. machine translation remains a useful facility that substitutes humans in the task of translating
B. machine translation remains an effective facility only with the assistance of human translators
C. machine translation offers practical help for the translator but does not complete the task of translation


I recommend reading the following posts that I have previously made on this topic:

Definition of CAT
http://www.proz.com/post/184880#184880

&

MT is part of CAT & MT postediting is worthwhile
http://www.proz.com/post/189873#189873

mediamatrix wrote:
Q4 Option D is a 'writing aid' and not, in itself, an MT or even a CAT tool.
The other options may be used in CAT or MT, with entirely different results in each case.


agata wrote in the questionnaire:
4. Which of the following MT aids do you use in your everyday life?
A. on-line dictionaries
B. term banks
C. glossaries
D. word processors
E. translation workstations
F. others:……………



This question does not seem to ask if these are MT tools or CAT tools, but rather if they are used as aids to MT.
On-line dictionaries are used for building MT dictionaries. Same with term banks and glossaries. Word processors are typically used for MT, and I specifically stated in my book review (http://www.multilingual.com/allen46.htm) concerning a book on the topic of MT postediting published in 2001 that it referred to MT processing and postediting techniques based on paper evaluation processes that were used in the late 1980s and early 1990s.
Even in the mid-1990s at Caterpillar, many of the translators using the MT systems did their assessment on print-out copies of the documents.
Translation Workstations have been part of the discussion of MT usage for many years. See:
http://cslu.cse.ogi.edu/HLTsurvey/ch8node6.html
http://ourworld.compuserve.com/homepages/WJHutchins/Conchist.htm
http://www.aymara.org/biblio/mtranslation.pdf
http://acl.ldc.upenn.edu/X/X96/X96-1024.pdf
http://ethesis.helsinki.fi/julkaisut/hum/yleis/vk/hoge/towardsa.pdf


mediamatrix wrote:
Q8 Options A and B are MT (by EAMT's definitions) and Option C is CAT.


agata wrote in the questionnaire:
8. What is the way you use machine translation?
A. machine translation as a translation aid for full-quality human translation
B. raw translation for information scanning
C. raw translation as a drafting aid, i.e. use of MT for translating rough versions of a text from the target language into the source language


I specifically described the difference between Content gisting (inbound translation, also referred to as assimilation in MT circles) and translation for publication (outbound Translation, also referred to as dissemination in MT circles) in the following:

Inbound vs Outbound MT systems and software
http://www.proz.com/post/190126#190126

And a description of each level of working with the MT output (known as MT postediting) is in my book chapter on the topic:
Post-editing. In Computers and Translation: A Translators Guide. Benjamins Translation Library, Number 35.
http://www.geocities.com/mtpostediting/PE-book-chapter-jeff-allen.pdf


mediamatrix wrote:
Q9 Option A - The definition of pre-editing is incorrect (pre-editing is the adaptation of the source text to increase the likelihood of it being undertood correctly by the MT system).


agata wrote in the questionnaire:
9. Which of the following techniques do you use for revision of translation?
A. pre-editing, i.e. checking the input text and adjusting it to satisfy the expectations of its receiver


On this point, your definition is closer to the one normally used. Since 2001, I have also started to refer to MT dictionary building as a pre-editing (or rather an MT pre-processing) step which can be done before MT processing or in parallel with it.


mediamatrix wrote:
Q10 greatest advantage? - compared to what?


On this point, the answers show that she means "what are the key diffenciators, or selling points, for using MT in your context?"

mediamatrix wrote:
Q11 effective? - in terms of what? (cost, time, accuracy, ROI, ....)


Yes, objectives, targets, and measurements need to be set. In many cases, as I have written in a couple of articles, these are not set. Two of my current articles in progress and nearly ready to be submitted will deal with these issues in more detail.

mediamatrix wrote:
Q14/Q15 what about 'flagrant errors of comprehension' ?


It appears that she was taking the assessment criteria from standards like J2450 or Blackjack. I agree that additional criteria, like the one you mentioned, could be added. However, the criteria she listed are objective, measurable items.

agata wrote in the questionnaire:
15. Which of them affect the quality of translation most?
A. grammatical errors
B. lexical errors
C. stylistic errors
D. punctuation errors
E. omissions
F. incompleteness


"Flagrant errors of comprehension" is in itself a subjective evaluation-based statement, which can be determined based on assessing a combination of the measurable points she listed in that question. I understand that what you are trying to say is "this is junk output that I don't understand". But how do we evaluate "excellent translation" versus "very good translation" versus "good translation" versus "acceptable attempt" versus "half-way understandable" versus "total junk"? It is difficult to establish a value level for subjective criteria and use it consistently, unless such an item can be calculated from other measurable sub-criteria.

Thanks for your feedback and discussion on this topic. It is encouraging to see that you have taken the time to look at the questionnaire, and carefully consider it compared with other information available elsewhere on this general topic.

This also shows that surveys need to be carefully created and presented. My experience in creating and administering about 10 different language technology surveys confirms that they more time and effort invested up front to making the survey understandable and easy-to-fill-in will make the survey/questionnaire more valuable for gathering accurate and useful responses.

I hope that Agata can benefit from your comments.

Jeff


[Edited at 2006-04-16 22:49]

Direct link Reply with quote
 
xxxmediamatrix
Local time: 10:54
Spanish to English
+ ...
MT, word-processors and all that ... Apr 16, 2006

Thank you again, Jeff.

In as much as your comments are based essentially on the writings of acknowleged experts in the field - including several of your own publications - I'm not sure they are entirely compatible with the context of Agasta's proposed questionnaire in which she is, by all accounts, seeking the opinions and experiences of non experts. As suggested by Ian, if Agasta wants expert opinions she would do better to submit her questionnaire elsewhere.

Whilst you, as an expert, may like to consider a word-processor as an 'MT aid', simply because Caterpillar used it 10 years ago to produce paper print-outs for verification, I - and I'm sure many others in the translation/publishing business - regard a word-processor as a fancy typewriter and very little more. Including 'word-processor' in a list of MT tools/aids, in a questionnaire intended for non specialists in MT, is asking for trouble and will inevitably lead to results inconsistent with those that might be obtained from a study among MT experts.

What is more - as if to deny the validity of your own argument - one of the references you provided ( http://ourworld.compuserve.com/homepages/WJHutchins/Conchist.htm ) states, on the very first page:

"The term 'machine translation' (MT) refers to computerized systems responsible for the production of translations with or without human assistance. It EXCLUDES computer-based translation tools which support translators by providing access to on-line dictionaries, remote terminology databanks, transmission and reception of texts, etc. The boundaries between machine-aided human translation (MAHT) and human-aided machine translation (HAMT) are often uncertain and the term computer-aided translation (CAT) can cover both, but the central core of MT itself is the automation of the full translation process."

Another of your references ( http://cslu.cse.ogi.edu/HLTsurvey/ch8node6.html ) is specifically concerned with Machine-Assisted Human Translation (MAHT), which it equates with CAT, so it is again not very relevant with regard to the point you seem to be trying to reinforce which is that a word-processor is an MT aid. No! - at best a word-processor can be used as an MAHT aid.

In much of the remainder of your response I observe that you have chosen, likewise, to exploit my comments primarily as hooks on which to hang references to your own publications and web-sites; bravo! - you have spotted an opportunity to display your wisdom, experience and credentials in this specialist field and you have exploited it well, albeit inappropriately in my opinion. The fact remains that a questionnaire about MT must include a definition of the scope of the survey (MT, MAHT, HAMT, CAT, etc.), it must stay within the bounds of the declared scope, and above it must all avoid confusing the issue with things like dumb word-processors.

Notwithstanding that, I guess we have clarified for Agasta the need to think very carefully about the presentation, structure and precise formulation of her questionnaire.

MediaMatrix

PS - Somewhat off topic, but far more interesting IMHO: You write:

Jeff Allen wrote:
It is difficult to establish a value level for subjective criteria and use it consistently, unless such an item can be calculated from other measurable sub-criteria.


There is an interesting comparison to be made here, I believe, between the subjective/objective assessment of the quality of translations (human or MT) and the quality of other forms of communication. In my specialist field, one of the main themes of study for many decades has been the search for objective measures for the subjective quality of TV sound and pictures. The result of this research has been the establishment and progressive refinement of some remarkably reliable, consistent and repeatable test methods that are recognised internationally in the broadcasting field, and indeed have been standardised for use world-wide. It seems to me that there might be some scope for the transposition of such methods into MT...


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
the merging of CAT, MT, MAHT, HAMT, etc Apr 17, 2006

mediamatrix wrote:
What is more - as if to deny the validity of your own argument - one of the references you provided ( http://ourworld.compuserve.com/homepages/WJHutchins/Conchist.htm ) states, on the very first page:

"The term 'machine translation' (MT) refers to computerized systems responsible for the production of translations with or without human assistance. It EXCLUDES computer-based translation tools which support translators by providing access to on-line dictionaries, remote terminology databanks, transmission and reception of texts, etc. The boundaries between machine-aided human translation (MAHT) and human-aided machine translation (HAMT) are often uncertain and the term computer-aided translation (CAT) can cover both, but the central core of MT itself is the automation of the full translation process."

Another of your references ( http://cslu.cse.ogi.edu/HLTsurvey/ch8node6.html ) is specifically concerned with Machine-Assisted Human Translation (MAHT), which it equates with CAT, so it is again not very relevant with regard to the point you seem to be trying to reinforce which is that a word-processor is an MT aid. No! - at best a word-processor can be used as an MAHT aid.
....

The fact remains that a questionnaire about MT must include a definition of the scope of the survey (MT, MAHT, HAMT, CAT, etc.), it must stay within the bounds of the declared scope, and above it must all avoid confusing the issue with things like dumb word-processors.


Thanks MediaMatrix for your comments.

I've pointed to a few of the websites that I have compiled on these topics, which do contain not only my publications but also publications of others. Most published books on the topic take at least 3-5 years to get out to the public (as was the case with my last book chapter). Many MT academic choose to take those types of publication channels because books are the only type of publications that help their academic career. Yet, by the time the public can read it, the information is often very obsolete because 1-2 major versions of MT software have already appeared in each product line, and have addressed the issues already written about in articles, books, and lots of incoming bug reports to the customer support helpdesk.

A brand new MT information web site is being compiled at the moment, yet the references available on it have been old and are requiring approval from various publishing houses to put the more recent publications available. This will be announced soon.

One key trend to note over the past 5 years in the overall field of translation tools is the boundary which becomes more fuzzy over time between MT/MAHT/CAT, etc. I cannot agree that it is a "fact" that there is limited scope (MT, MAHT, HAMT, CAT) to translation tools.

For example:
Translation Memory (TM) tools are no longer just processing of parallel databases of aligned segments. Some TM tools now include an "Assemble" function which attempts to produce a translation (with Rule-Based MT concepts) above and beyond the information stored in the parallel databases as aligned sentences. Statistical-based algorithms are now included to produce better fuzzy matches and thus set lower threshold for texts which can be considered as acceptable for human translator post-processing. So now it is no longer an issue of talking about pure TM, which BTW is simply a commercialized branding of what Nagao called Example-based Machine Translation (EBMT) in the early 1980s, but the wave of Statistical-based MT (SBMT) is now being included in TM. So, TM tools (often referred by professional translators as CAT tools) are based on an EBMT methodology and now include SBMT processes and are finding ways to become more RBMT-like over time, yet no one has a problem with that.

On the other hand, MT tools have been moving away from just a push-button perspective of behind-the-scenes batch file processing, and have been taking on a user-oriented translation workstation approach which includes MT software, translator-friendly features, dictionary building techniques, terminology research tools, postediting principles, double and triple postediting and revision cycles. This can be seen in the texts of the calls for proposals for various MT conferences over the past 1-2 years (AMTA, EAMT). But when this approach is announced, the translator community flames up and says "No way, you can't use our CAT or MAHT labels for your tools, stay away from us".

The former so-called lines of distinction between MT, MAHT, HAMT, etc have been disappearing over the years as both sides of the spectrum are integrating new features into their tools and workflow processes. And now several MT tools include direct plug-ins for specific TM tools (PROMT has a PROMT for Trados plugin, and the same is under development at SYSTRAN). Wordfast starting with version 5 has been announced with full innate compatibility with all MT tools which have direct plug-in compatibility with Microsoft Word.

mediamatrix wrote:
Another of your references ( http://cslu.cse.ogi.edu/HLTsurvey/ch8node6.html ) is specifically concerned with Machine-Assisted Human Translation (MAHT), which it equates with CAT, so it is again not very relevant with regard to the point you seem to be trying to reinforce which is that a word-processor is an MT aid. No! - at best a word-processor can be used as an MAHT aid.


That HLT survey book is very old. I recall reading it in 1995 as the only existing overall reference that had previously been published. What I was demonstrating with that reference is that the several terms of MT / MAHT / HAMT / CAT and several others had already been equated at that point in time. Yet, every translation tool supplier that I know, whether they be TM tools or MT tools, spends a major amount of effort trying to be fully compatible with MS Word. And via the plug-in compatibility with MS Office pack applications which have been put into place in just the past couple of years for most of these tools, it is now possible to do the translation tasks inside of MS Word by completely bypassing the need to go into the Editor environment of the TM tool or the MT tool. Just clicking on various installed buttons in the task bar of MS Word allows the user to access the dictionary tools and other features of the TM and MT tools. This full integration with MS Word has made this possible, and now MS Word has become the environment in which MT pre- and post-processing takes place. It is of course still possible to do the work within the dedicated MT interface of a given MT tool, but translators prefer to do it all in MS Word, as is done with TRADOS. So, MS Word is an example where a well-known and well-used word processor has become an MAHT/MT aid because nearly all of the tools are now embedded into it.

Yet, there are other examples of where this is not the case. The KANT MT system which had been deployed not only to Caterpillar but also to General Motors had full system integration with the SGML editor Arbortext Adept Editor. Note: None of the TM tools were compatible with TM tools at that time. I know, because I tested them all and as a result, we developed our own TM tool to integrate with the SGML files. Over the years, this has been extended to work with XML-based editors for the processing of these file types. Yet, it is not necessary to use the XML or SGML based editor to do the translation work. If I was on a Windows workstation, I conducted file troubleshooting support work on 11 languages for 40 human translators at Caterpillar by using MS Notepad to open and fix the files. And when I was on the AIX or Sparc Unix workstations, I opened the file with VI, made the tag changes, and saved and quit the file. So, it is very possible to conduct translation tasks, including inserting text in many languages by using SGML-based codes, by working outside of a common text editor environment. It can be done, it has been done, it is still done today by people who work on large Unix and Linux based networks. There are complex environments where using a typical word processor on an isolated workstation is not possible. this can be a real pain for those who are not used to working off of the command line, and my guess is that maybe 10-15% of translators on this site would be able to make such changes to database files using VI without some additional training and mentoring. Yet, in large corporate environments with merging technologies, and their subcontracted translation suppliers who used exported files, it is sometimes necesssary to have those skills to complete the "translation tasks" on-time.


A last point about definitions. In 1998, an MT certification committee was formed to help define and promote MT systems as opposed mainly to bilingual dictionaries. There was a significant amount of work done on creating the range of definitions, yet these did not correspond to the MT Compendium by John Huthchins (which you will notice has changed names to become the Compendium for Translation Software because of the mixing of the types of systems).
The MT certification committee did not continue with their work after 2000 (from what I recall).
And now looking back over the past 5 years of how the different "MT" systems and software have evolved in various ways, it would be necessary to redefine all of the categories based on a new set of criteria.

MediaMatrix, you are pointing out a lot of good issues. The answers to these are often very dispersed across different sites, different journals, and in books that will be forthcoming for the next 2-3 years. I have made it a point over the past 7 years to provide a spectrum of fast-turnaround publications on practical aspects of using a range of products (not just one tool), and have made nearly all of them available for free to the community at large. I do cite all who have provided input information to my publications, and the reference sections to some of my articles are complete with links off to works written by others as well. These are often springboard publications to get the nutshell of the issue in a couple of pages and then go read more about it elsewhere. But if you dig a bit, you will see that many of these points are not described elsewhere (like the MT tips page) or are spread out all over the universe. Most people do not have time to go looking for such information. When I saw many such projects conducted with very superficial analyses of existing products and methods because they did not take the time to do necessary background research on relevant information, I set up several information portals to open the channels to the information. Thus the reason for creating the langtecheval site about a year and a half ago.

Jeff


Direct link Reply with quote
 
Teresa Reinhardt  Identity Verified
United States
Local time: 07:54
Member (2002)
English to German
+ ...
Can I have a user-friendly version? Apr 17, 2006

I've worked for CAT and used/maintained/post-edited their tool. I'd be happy to answer your questionnaire, but I don't have the time to re-type what you posted.
If you send me an electronic version, I'll comment on your questions.

Teresa


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
good idea to use TV broadcasting quality for MT quality measures Apr 17, 2006

Jeff Allen wrote:
It is difficult to establish a value level for subjective criteria and use it consistently, unless such an item can be calculated from other measurable sub-criteria.


mediamatrix wrote:

PS - Somewhat off topic, but far more interesting IMHO: You write:

There is an interesting comparison to be made here, I believe, between the subjective/objective assessment of the quality of translations (human or MT) and the quality of other forms of communication. In my specialist field, one of the main themes of study for many decades has been the search for objective measures for the subjective quality of TV sound and pictures. The result of this research has been the establishment and progressive refinement of some remarkably reliable, consistent and repeatable test methods that are recognised internationally in the broadcasting field, and indeed have been standardised for use world-wide. It seems to me that there might be some scope for the transposition of such methods into MT...


Not off-topic at all. It's a very interesting analogy to consider. There is an obvious, underlying type of subjective quality measure in the TV sound and pictures domain given that it is visual and directly tangible. Since I am really into photography and have had manual reflex and now digital cameras over the years, I have tried to describe the degradation or change in quality of images between what is directly perceived by the human eye and that which is converted to a digital image and then perceived again by the human eye.
There seems to be a good parallel with translated text.

I have been discussing this with MT software editors for the past 2 years and I'll contact you off-line to see how it could/would be possible to leverage the quality measure work already done in such a parallel field.

Jeff


Direct link Reply with quote
 

Jeff Allen  Identity Verified
France
Local time: 16:54
Member (2011)
Multiplelanguages
+ ...
Correction in statement above about MT systems at Caterpillar and GM Oct 23, 2010

The GM team did not use the same MT as system (KANT) as Caterpillar, which was deployed by CMU. GM used the system offered by Lantworks.

Jeff


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:

Moderator(s) of this forum
Maria Castro[Call to this topic]

You can also contact site staff by submitting a support request »

Need help for Machine Translation Questionnaire

Advanced search







WordFinder
The words you want Anywhere, Anytime

WordFinder is the market's fastest and easiest way of finding the right word, term, translation or synonym in one or more dictionaries. In our assortment you can choose among more than 120 dictionaries in 15 languages from leading publishers.

More info »
Protemos translation business management system
Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!

The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.

More info »



All of ProZ.com
  • All of ProZ.com
  • Term search
  • Jobs