MultiTrans 4.4 TermBase Agent
Tutorial, Level I
|Other MultiTrans tutorials:||Other active terminology recognition system tutorials:|
|MultiTrans TextBase Builder Tutorial, Level I|
|MultiTrans TermBase Manager Tutorial, Level I|
|MultiTrans Term Extractor Tutorial, Level I|
|MultiTrans Analysis Agent Tutorial, Level I|
|MultiTrans Translation Agent Tutorial, Level I|
|MultiTrans TextBase Agent Tutorial, Level I|
(Generic description coming soon.)
MultiTrans is a translation environment. Together with powerful alignment and translation memory functions, MultiTrans is comprised of the TermBase Agent which identifies terms and can insert them directly into a text to be translated (or one that is already partially translated, for example using the Translation Agent). This feature is largely intended to complement the traditional segment-level translation memory functions.
You can find out more about MultiTrans by consulting the MultiCorpora website at http://www.multicorpora.com. When you open MultiTrans, you can also click on the Aide MultiTrans icon to read the help files providing information on MultiTrans’ different functions.
II. Getting ready
III. Using the TermBase Agent to insert terms in a text to be translated
The TermBase Agent allows you to compare the document you wish to translate with the contents of the TermBase. If a match for a term in the TermBase is found, its equivalent as it appears in the term record is automatically inserted into your text. If more than one match is found, it is up to you to decide which equivalent to use.
The TermBase Agent has automatically replaced the terms found in the TermBase with their translation equivalents, which appear in blue. Take a look at the text to see what terms have been inserted, and where.
IV. Completing and re-reading the text
V. Wrapping up
NOTE 1: If you have not completed these tutorials, download the ready-made database MultiTrans Bases — TextBase and TermBase Agents. For this tutorial, extract the contents of the file to your sub-directory. (For instructions, see Extracting files from a compressed folder in Windows.)
NOTE 2: If you do not have this text, you can use the file WHO – obesity Text_translation agent.doc, included in the compressed folder MultiTrans Bases — TextBase and TermBase Agents. This file contains a version of the text which has been partially translated using the MultiTrans Translation Agent. If you would like to process the whole text you can do so by going to the folder you created at the beginning of this tutorial and opening the WHO obesity – Text to translate document.
WARNING: If you are using your only copy of the source-text document, be sure to follow the steps above to make another copy of it before proceeding. The text in the document will automatically be replaced with the translation, and if you do not make a copy of the document before beginning you will lose your source text.
NOTE 3: If you have not yet created a TextBase and TermBase you can do so before continuing, by using the MultiTrans TextBase Builder Tutorial, Level I. Alternatively, you can use the ready-made TextBase and TermBase and the WHO-obesity Text_translation agent.doc file you may have saved to your folder and extracted from the compressed folder MultiTransBases - Agents at the beginning of this tutorial.
NOTE 4: If you wish, you can also add some additional terms, as described in the MultiTrans TermBase Manager Tutorial, Level I.
VI. Questions for reflection
- What are your first impressions about the functions and functioning of MultiTrans' TermBase Agent?
- What could theTermBase Agent help you to do? In what kind of situation?
- How does MultiTrans’ TermBase Agent compare to other tools you have tried? Have you found anything similar in other tools?
- What are some of the advantages and disadvantages of using theTermBase Agent? How does this compare to a manual approach? To the Translation Agent in MultiTrans?
- How did the replacements made by the TermBase Agent compare to what you expected to see when you compared the text and the TermBase? Were all of the replacements you expected made?
- Did you notice any challenges in inserting terms from your TermBase into the text? If so, what were they? Were they linked to the form of the source language term, or its equivalent? Or to the meaning of the source text term or its equivalent? To something else? Did they affect whether or not equivalents were inserted and where (e.g., missing some occurrences, or inserting equivalents where they should not have been)? Or which equivalents were used? How might they affect the usefulness of this kind of tool?
- Do you think it is important for a translator to be able to choose among a number of potential equivalents for a term? Why or why not? What kinds of information can help you to decide on an equivalent to use? Did MultiTrans give you easy access to this kind of information?
- Do you think that there may be differences in the advantages and challenges of inserting single-word and multi-word items into a text automatically with a tool like the TermBase Agent? If so, what might they be, and how might they affect the choices you might make in building a TermBase?
- What should you as a translator be aware of when using such a tool? What kinds of steps should you take after using the tool? (You can think about the things you observed above, or other issues that you think could be observed in other cases.)
Did you find that you had a preference for the automatic type of insertion done by the TermBase Agent, or the manual insertion using the TextBase Agent? What are the advantages and disadvantages of each kind of approach?
Tutorial created and updated by the CERTT team. (2011-04-13)