Fragment recall and Fuzzy match repair - upLIFT

Hi everyone,

I have just downloaded STUDIO 2019 trial and I have got a problem with the upLIFT feature. It just does not work.

I have created a new project, file-based TM with over 9000 TUs and changed upLIFT settings as per training video:

I didnt have to upgrade my TM, as the "upgrade" function was inactive, so I presumed it didnt need upgrading.

The screen below is the editor view of my project:

As you can see "puerperal fever" appears in TU1 and TU4, but no fragment match found for "puerperal fever" in TU6. There are no relevant fragments suggestions when I start typing the translation of eg. "puerperal fever".

I looked at similar queries on the forum, but could'n't find anything that would solve my problem.

Any suggestions would be greatly appreciated Slight smile

Top Replies

  • There are a number of features that help you with to keep your translations consistent.

    Concordance: If you mark some source text and press F3, Studio will do a concordance search and call…

Parents
  • There are a number of features that help you with to keep your translations consistent.

    Concordance: If you mark some source text and press F3, Studio will do a concordance search and call up (and highlight) former uses of a term or phrase. So if there's a term and you know you translated that before but don't remember exactly how, that is one way to help you.

    Termbase: Maybe you are a medical translator and come across terms like "puerperal fever" regularly, then you start entering them into a Termbase. Termbases are extremely helpful to keep your translations consistent, and SDL has quite a few webinars on them, so I won't say much more. If you translate a term that took a bit of research or that you know can be translated in several ways, just hit CTRL+F2 to quick-add it to the termbase. If it re-appears, it will be marked up in the source text and the translations, annotations, illustrations etc. whatever you entered into the termbase will be shown to you.

    Fragment matches: This function gains the larger your TMs are. At some point you will see lots of fragment matches, but it takes a bit to gain traction. In your example, you are expecting Studio to know which words in your target text correspond with which words in the source text. Over time, Studio can conclude that in some cases, but not from just one former instance.

    Daniel

  • Hi Daniel,

    thank you for your reply and useful tips!

    I still have a few training videos to watch and haven't seen the Termbase one yet...

    I  want to practice everything I see in the videos, so it takes me longer to get through the training, but I am getting there... hopefully... Slight smile

Reply Children
  • Studio is a fairly complex program. My approach was to learn enough initially to get me going, so that I could work with it - bare bones so to say. Then I started to branch out to see what there is to make my work more efficient. This is the approach I would recommend. When we were looking for a new CAT program (we had clearly outgrown our previous solution), I went for the industry standard - which is Studio - trusting that we would not be too far off the mark. Looking back, I am very happy with that decision. It's too complex to assess everything right away. You have to trust. I trusted in the market leadership of this product, and I don't regret it.

    Daniel