-
Notifications
You must be signed in to change notification settings - Fork 2
Description
It should be possible to lump aspectual pairs together (e.g. начинать|начать) when checking target vocabulary against the story text. In other words, if начать (perfective) is listed in the target vocabulary, but a form of начинать (imperfective) is used in the story text, then it should be marked as used.
The database includes aspect and aspect_counterpart for lemmas, so it should be possible to make the connection.
Question: Should this be an optional ability that can be toggled? Is there a scenario where you wouldn't want this behavior?
Scenario: Imperfective used in story, perfective listed in vocabulary
Input Story:
Я начинаю урок завтра.
Input Vocabulary:
начать
завтра
| Actual Output | Desired Output |
|---|---|
| Я начинаю урок завтра. | Я начинаю урок завтра. |
The word начинаю is used in the story, and has lemma начинать (imperfective). But the vocabulary only lists начать (perfective). It should be recognized that начинать|начать form an aspectual pair, and therefore начинаю should be marked in the story as used.
Scenario: Perfective used in story, imperfective listed in vocabulary
Input Story:
Я начал учиться вчера.
Input Vocabulary:
начинать
вчера
| Actual Output | Desired Output |
|---|---|
| Я начал учиться вчера. | Я начал учиться вчера. |
The word начал is used in the story, and has lemma начать (perfective). But the vocabulary only lists начинать (imperfective). It should be recognized that начинать|начать form an aspectual pair, and therefore начал should be marked in the story as used.