I recently came across this reddit post:
https://www.reddit.com/r/Anki/comments/1ge2aui/note_types_to_avoid_pattern_matching/
Read the ‘Randomized Cloze’ section.
Grammar with Anki has limitations:
If you have the context in front, you can deduce the meaning of the grammar point from the surroundings without even looking at the grammar point. However, if you have no context at all, you will constantly get the card wrong because of ambiguity, etc., and won’t learn the abstraction.
This is why grammar is best learnt through reading. But, if there was a way for the context to constantly change/cycle through in Anki, that would solve the problem and grammar mined could be integrated more efficiently into Anki. The limitation with Yomitan is that every new sentence addition creates a new card. But if it were possible that an addition of an existent grammar point becomes part of the randomized cloze instead of creating a new card, that would be perfect for grammar.
With randomized cloze, separated by: |
you can manually create the cards, manually add every sentence you create, etc., but I’m wondering if there’s a possibility to change the code of Yomitan in the Anki section such that it automatically checks if the vocab exists (I think this was an inbuilt feature), and if it does, adds the new sentence onto the existent one, only separated by | and formatted to be just 1 card (randomized cloze).
Obviously, Yomitan can still add a note for a vocab you already created, but if I mine 5 sentences and still don’t remember the usage of that grammar point, that’s 5 new notes rather than (the superior) randomized cloze, where you have 5 sentences and only 1 is chosen to be displayed.
I’m not good at coding, so I have to rely on ChatGPT to help with most of the process, which I’ll probably try to do in a month once my break starts. If you use Yomitan and are good at coding, and decide to implement this for your own mining, please share your method.
Here’s an example to demonstrate what I mean:
First sentence mined: “The foxed {jumped} over the house” (the mined word is in brackets)
Vocabulary field: jumped
Sentence field: The foxed jumped over the house
Definition field: push oneself off a surface and into the air by using the muscles in one’s legs and feet.
Second sentence mined: “The cow {jumped} over the moon”
Vocabulary field: jumped
Sentence field: The foxed jumped over the house | The cow jumped over the moon
Definition field: push oneself off a surface and into the air by using the muscles in one’s legs and feet.
Every time you mine the same word, the sentence gets added to the previous sentence field, separated by | rather than creating a whole new note. This is useful for hiragana vocab, which is too difficult outside of context, but are too easy with context. It’s also very useful for other languages too. Please share if you know how to do it.