How to Build a Termbase From Scratch
How to Build a Termbase From Scratch

Your company has recognized the importance of terminology management and is aware of the added value that well-kept terminology has for your business. You might be the terminologist or terminology manager in charge and want to start by setting up a termbase. This may be daunting if you don’t know how and where to start. Let me help you out with this brief 3-step guide.

1.   Think about the purpose of your termbase.

If your termbase is to be solely translation-oriented, then you don’t necessarily need too many data categories. You should be aware, however, that you will not be tapping a termbase’s full potential. If you are striving for “bigger” things, you will have to give more thought to your termbase definition and the data categories you choose to use. You might want to build interfaces from your terminology management system to other tools, like content management or PIM systems, or you might be planning on using your termbase as a basis for knowledge management, like building an ontology, at some point. And remember: a termbase can even serve all of the mentioned purposes at the same time.

2.   Choose your Terminology Management System

If you’re lucky, you will get to choose which terminology management system you want to use. Get in touch with different companies and have them present their tool to you. Take into consideration the results of step 1 and see which TMS caters to all your needs. However, if the tool has already been acquired, you will have to work with what you’ve got, including any possible limitations in how freely you can build your termbase. Of course, there is no “right” or “wrong” TMS, but generally speaking, the more flexible a tool is, the better.

3.   Define your data categories

When it comes to creating the termbase, you can either stick to the TBX data categories or look for inspiration in the DatCat database. When setting up the termbase definition, you have to adhere to the principles of concept orientation, term autonomy, and granularity. This also means you will have three levels in your termbase definition: concept, language, and term level.

Concept level:

  • Definition: This might be the most important data category, which is unfortunately often skipped when creating a termbase entry (especially in translation-oriented termbases). It defines the concept (not a term!) and paves the way for finding the corresponding term in each language. (If you don’t know how to write a proper definition, have a look at chapter 4.1.4 of COTSOES.)
  • Source of definition: Only use reliable sources for your definitions.
  • Subject: List of subjects to choose from. (If you’re struggling with coming up with subjects, it might help to have a look at different classifications and taxonomies.)

Language level:

Indicates what language the respective terms are in.

Term level:

  • Term: Word or string of words that represent the concept. Each entry can have several terms in one language.
  • Term source: Indicate where you found the term.
  • Term type: Here, you select whether a term is a full form, a short form, an acronym, a synonym, etc.
  • Usage status: This might be the second most important data category. The values are “preferred” and “forbidden,” and sometimes it might be useful to also use “allowed.” Bear in mind that while there must exactly be one preferred term per language level, no more and no less, there can be more than one forbidden (or allowed) term.
  • Part of Speech: Useful because in some languages, like English, it cannot be differentiated between a substantive and a verb.
  • Grammatical gender: not applicable to English but important for translators or non-native speakers of other languages. The gender category also comes in handy to distinguish between homonyms.

These are the basic data categories that I would very much advise you to use. On top of that, a note field on concept and/or term level, where you can put any information that does not fit into the other data categories (remember: granularity!), might come in handy. Of course, it might be necessary to have even more data categories in your specific use case. But always try to keep in mind not to go overboard with your termbase definition, as you don’t want to make creating a termbase entry too complicated and time-consuming. You should also remember that many terminology management systems don’t allow you to change the termbase definition once you’ve made your first entries, so don’t rush through the process of creating it.

After familiarizing yourself with these three steps, you now have a rough idea of what is important when setting up a termbase and are ready to get to work and set up your own from scratch!

Submit a Comment

Your email address will not be published. Required fields are marked *