Did you find this page useful? Give us feedback.
Did you find the information
on this page useful?

  Tell us more about your choice:
Missing information
Incorrect information
Misleading information
I have a question
  Enter your comment in the box below.
If you need assistance, please contact Support instead.

  Type the letters
simple PHP captcha

memoQ includes a linguistic quality assurance (LQA) module that reviewers can use to check many aspects of translations (accuracy, omission, etc.) and to categorize the errors as major, medium or minor. You can also define a pass/fail criteria. This topic describes how you can edit and use an LQA model in your project.

LQA models are light resources in memoQ. In each project, you can use one LQA model. To edit or create an LQA model you wish to use in your project, do the following:

1.Open the project, and in Project home, click the Settings icon. The Settings pane appears.

2.In the Settings pane, click the LQA model icon ().

3.The list of LQA models appears. The list is similar to a list of translation memories. memoQ lists the LQA models available. These will be the LQA models available on your computer, and on the server you are currently connected to, language independent. Check the check box for one of the LQA models that you want to use. You can also clone an existing model. In this example, the LISA model was cloned and is used to adjust an existing LQA model according to your needs. You can also create your own LQA model from scratch.

4.Click the Edit link to open an LQA model for editing. The dialog has three tabs where you can make your configurations.

5.On the QA Model tab, delete a category, or add a new category. Enter the category name in the text field below the last category entry, then type a subcategory. You can select values from 0, 1, 5 or 10 for all three severity levels (minor, major, critical). You can also add a new error severity like medium.

6.On the QA mappings tab, you can select a memoQ QA error type and map it to an LQA error. Open the automatic QA type drop-down list, and choose your QA error type. Then open the LQA error category drop-down list, and choose from the list which error category should match the QA error.

7. On the Pass/Fail criteria tab, you can change the penalty points per word, the normalized score (which will also appear in the LQA reports), and you can check the Count multiple occurrences for the same error as one check box. When the same error occurs several times, it is considered as one error.

Here you can adjust the fail/pass criteria to the criteria your customer may use. You can ensure that your translation will pass the QA process of your customer.

8. Assign your LQA model to your project: Project home > Settings > LQA models.

Note: If you have already imported documents to your project and only afterwards assigning an LQA model, then the LQA model is applied to all imported documents.

Other scenarios:

Documents that already have an LQA model assigned, and you then change the LQA model, or you imported bilingual documents which included an LQA model, then the newly assigned LQA model is not applied, the old one is kept. This works the other way around as well: if you import a document with a different LQA model to the project which has an LQA model, the newly imported document's LQA model will stay.

An LQA model in an online project can only be added from the online PM interfaces. You cannot do this in a checked out copy of your project. The documents in the checked out project will get the new LQA model at the next project synchronization.

Important compatibility note between memoQ versions: When the memoQ desktop application and the memoQ server are both on memoQ 2015 (7.8) or higher, the new LQA model is synchronized. It does not work when the memoQ desktop application has a lower version. A combination of a lower memoQ server version and a higher memoQ desktop version is not affected as the LQA model can only be changed from the online PM interface.

An LQA model can appear in a document anytime during its life time (e.g. in packages, XLIFF based offline/update). It is possible that a document without an LQA model is updated with another one that already contains a model. In this case, the incoming LQA model and the LQA errors in the rows are disregarded. This can happen when a reviewer already started and entered LQA errors.