101

1. What is MT?

Determine if the following statements are true or false

2. What is KantanMT?

Where should I go to...

Сombine elements from the left column with the elements on the right
...create groups of data?
 
Training Data
...manage and activate my profiles?
 
Client Files
... build an engine?
 
Templates
... link your engine to your CAT tools?
 
Clients Profiles
... translate documents
 
API

3. Creating a MT engine in KantanMT

How to build an engine and put it to work

Upload Training Data
Click Translate
Go to Client Files
Go to my Jobs to check the task
Go to your Training Data Tab
Click Build
Activate profile
Upload Files to translate
Go to my Jobs to check the task and download the output

4. Types of Training Data

Which of the following scenarios would result in a properly built engine?

5. Steps to Build an Engine

Please fill in the gaps

I want to create a new engine from scratch. First I will need to add my and

measurebuildlearn
. Once it's finished I will  
buildmeasurelearn
 the scores and  the needs of my engine. Now I will add my 
stock databilingual dataglossaries
 and  
bilingual datamonolingual dataKantanLibraryexcel files
, and train the engine again. I will follow the same process and when it's done, I will check my
training datauntranslated filesword counts
. If they are low, I can add  and  again to boost the scores and have better results when I .

With these three iterations, following our three steps, we will manage to create an engine that will be ready for !

6. Machine Translation Scores

Which ideal figure corresponds to which score?

Сombine elements from the left column with the elements on the right
F-Measure Score
 
> 70%
TER Score
 
< 40%
BLEU Score
 
2 million
Monolingual Word Count
 
1 million
Unique Word Count
 
> 60%
Bilingual Source Word Count
 
> 5 million

7. Build Analytics

Where can I find the following reports?

Сombine elements from the left column with the elements on the right
Untranslated content from my data
 
Gap Analysis
Invalid segments from my bilingual data
 
Rejects Report
Backlog of my engine
 
BLEU Score Report
The precision of my data
 
Timeline
Quality of the target language from my data
 
F-Measure Score Report
Post-editing effort
 
TER Score Report