Setting Memo Q spellcheck to avoid proper nouns Thread poster: Paula Darwish
|
|---|
Paula Darwish 
United Kingdom
Local time: 19:27
Member (2013)
Turkish to English
+ ...
Has anyone had any success with setting the MemoQ spellcheck to avoid proper nouns? (Hunspell marks Turkish names and towns etc in the English as errors).The MemoQ support team told me it wasn't possible to do this using the Hunspell dictionary but it might be possible to do it with some kind of custom Word dictionary. Has anyone else had the same issue and found a solution?
| | | |
Adieu
Ukrainian to English
+ ...
| You'd have to add a dictionary of Turkish names | Oct 29, 2021 |
These things can't tell a noun from any other word, afaik.
Although... maybe there is some setting somewhere to skip spellcheck on capitalized words?
| | | |
How should software be able to determine that a word is a proper noun and not a typo?
There's a German town called Gotha. Maybe you wanted to type 'gotcha'. How can the software determine if it's one or the other? And what about Forst? Did the user intend to type 'forest'?
| | | |
Paula Darwish
United Kingdom
Local time: 19:27
Member (2013)
Turkish to English
+ ...
TOPIC STARTER | Ignoring capitalised words | Oct 29, 2021 |
I was hoping there was some way to ignore capitalised words for texts with a lot of proper nouns in although obviously there might be times when you wouldn't want to ignore capitalised words.
| | |
|
|
|
Paula Darwish wrote:
I was hoping there was some way to ignore capitalised words for texts with a lot of proper nouns in although obviously there might be times when you wouldn't want to ignore capitalised words.
To do that, someone would need to formulate an algorithm that a computer program can follow.
How would it determine which of the following words to spellcheck? Could you formulate a method to detect what to skip?
'Our new Sumer Hollidey Kattalogg for Brno is ready to download.'
| | | |
| Avoid proper names | Oct 30, 2021 |
A related problem that can occur, in particular, when translating software manuals is with identifiers that are often written in CamelCase. To prevent such identifiers from being flagged as spelling errors, they can be converted temporarily to tags. You could use the same technique to tag uppercased words within a sentence, and so exclude them from being spell-checked.
| | | |
Paula Darwish
United Kingdom
Local time: 19:27
Member (2013)
Turkish to English
+ ...
TOPIC STARTER | CamelCase/Tags | Oct 30, 2021 |
Thanks Anthony that is really helpful. I will try that.
Anthony Rudd wrote:
A related problem that can occur, in particular, when translating software manuals is with identifiers that are often written in CamelCase. To prevent such identifiers from being flagged as spelling errors, they can be converted temporarily to tags. You could use the same technique to tag uppercased words within a sentence, and so exclude them from being spell-checked.
| | | |
Stepan Konev
Russian Federation
Local time: 21:27
English to Russian
| How to distinguish proper names and typos? | Oct 30, 2021 |
Anthony Rudd wrote:
You could use the same technique to tag uppercased words But how do you tell memoQ which capitalized word is a proper name and which is a typo? (See the question asked above by Thomas T. Frost). If you set a regex for Regex Tagger to tag all Upper Case Words, it will capture 5 words in this example: Our new Sumer Hollidey Kattalogg for Brno is ready to download, while you need just one.
| | |
|
|
|
Paula Darwish
United Kingdom
Local time: 19:27
Member (2013)
Turkish to English
+ ...
TOPIC STARTER | issue of typo or proper noun | Oct 30, 2021 |
Stepan Konev wrote: Anthony Rudd wrote:
You could use the same technique to tag uppercased words But how do you tell memoQ which capitalized word is a proper name and which is a typo? (See the question asked above by Thomas T. Frost). If you set a regex for Regex Tagger to tag all Upper Case Words, it will capture 5 words in this example: Our new Sumer Hollidey Kattalogg for Brno is ready to download, while you need just one.
Thanks for your comment. As you and Thomas have said, identifying this difference would be a problem. But I am thinking of developing something for use with specific texts that have many place names and few other capitalised words. As another user has suggested, if I had a database of Turkish proper nouns I could add them to my English dictionary and that would resolve the problem. When I do a spellcheck on Turkish texts, it does not pick up the personal names, place names etc. Therefore, there should be a way to extract these proper nouns from the Turkish dictionary, or obatin a separate list, and add them to an ignore list. If I come up with any solutions, I will certainly share them. No spellcheck is foolproof of course but I think such a system would be useful with specific texts.
Thanks to all for their input so far.
| | | |
Stepan Konev
Russian Federation
Local time: 21:27
English to Russian
[Edited at 2021-10-30 17:31 GMT]
| | | |
Login to reply/comment 
