I'm adapting the Anki "Japanese corePLUS" deck to use just to practice reading / writing kanji, starting with the Core 6000 expressions which include kanji. I already know how to read well over half of them, but I am trying to bring my writing up to snuff (as well as raising my reading level).

The other thing I'm trying to do it to move away from Heisig keywords for all of the RTK1 6e kanji. Therefore, I want to have at least one expression for each of the Heisig kanji (even the rarer ones). For this, I need to know which kanji aren't in any Core 6000 words.

A prior poster listed the kanji which weren't in any Core 6000 /sentences/, but that doesn't really help me with this. And I don't have the computer-fu to figure it out myself. If no one knows, or knows how to figure it our all snappy and computery, then I'll do it by hand and (if anyone's interested) post which ones here.

Thanks in advance for any assistance.

I had an idea about exporting Core 6000 cards to an excel sheet where each row represents one card. Thereupon deleting all sentence cards and cards with a word without kanji.
Do you understand what I am thinking of?
Still, I can't think how one would go from there to assigning one or more words from that list with the same Kanji to the correct Heisig's card...

momokun Wrote:I'm adapting the Anki "Japanese corePLUS" deck to use just to practice reading / writing kanji, starting with the Core 6000 expressions which include kanji. I already know how to read well over half of them, but I am trying to bring my writing up to snuff (as well as raising my reading level).

The other thing I'm trying to do it to move away from Heisig keywords for all of the RTK1 6e kanji. Therefore, I want to have at least one expression for each of the Heisig kanji (even the rarer ones). For this, I need to know which kanji aren't in any Core 6000 words.

A prior poster listed the kanji which weren't in any Core 6000 /sentences/, but that doesn't really help me with this. And I don't have the computer-fu to figure it out myself. If no one knows, or knows how to figure it our all snappy and computery, then I'll do it by hand and (if anyone's interested) post which ones here.

Thanks in advance for any assistance.

I can help you generate this deck if you still need it.

Cards in Core 6000 cover 1599/2200 out of kanjis in Volume 1. That means that 601 of them are not covered.
I also noticed that 47 kanjis from Volume 3 are used.

I'm not exactly sure what kind of deck you need. I could generate a deck smaller version of this deck for you that contains 1291 cards that cover all of these 1598 kanjis.

Or a deck containing 1599 cards, each one with one example from the set.

Let me know what kind of deck you need.

I got a list of 636 kanji (out of 2200) with this shell command:

comm -23 <(awk -F$'\t' 'NR!=1&&$12!=""{print $1}' kanji.txt|sort) <(cut -f2 core-6000.txt|grep -o .|sort)

I uploaded the output to http://lri.me/upload/rtk1-kanji-not-used...-words.txt.

The kanji are from the sixth edition of RTK 1. core-6000.txt is based on a newer version of the Core 6000 data where there are slightly more words that use hiragana in place of kanji.

I got almost the same list of 637 kanjis. I guess one more is due to a bug or something.

What about pronunciation and listening comprehension? Are you using the core sentences for either?

One thing you could do to train your reading comprehension (as far as kanji recognition goes, I don't mean actual reading) is this deck:
https://ankiweb.net/shared/info/70031263
useful background knowledge (I wouldn't pass on it): http://www.valdes.titech.ac.jp/~terry/2kcw.html#jwf

Another to train your listening comprehension:
Front: Core image (or a self-chosen one) + sentence audio
Back: sentence w/ furigana, no english

And writing the sentence out before clicking Answer whenever you're slow to comprehend a sentence.

lauri_ranta Wrote:core-6000.txt is based on a newer version of the Core 6000 data where there are slightly more words that use hiragana in place of kanji.

About that file... How do you rate the new order/selection? Would you recommend it over Kore?

killua Wrote:

lauri_ranta Wrote:core-6000.txt is based on a newer version of the Core 6000 data where there are slightly more words that use hiragana in place of kanji.

About that file... How do you rate the new order/selection? Would you recommend it over Kore?

The order of the words or sentences doesn't really matter in my opinion, but see http://forum.koohii.com/showthread.php?p...#pid185115.

[RawToast]

killua Wrote:

lauri_ranta Wrote:core-6000.txt is based on a newer version of the Core 6000 data where there are slightly more words that use hiragana in place of kanji.

About that file... How do you rate the new order/selection? Would you recommend it over Kore?

If that file is based on the more current iknow ordering then it removes some of the problems people had with the 400-800 range. In that area the old list had a focus on industrial and political terms, which many users disliked.

When iknow reordered the list a few more common words should be in the first 1000 words and the least used/rarer words in the list (like 共産/communist) were pushed into the last few steps of the "Core 6000" so for us your probably looking at words 5500+.

If you're looking at using Nukemarine's guide, then I wouldn't bother with this list. As the guide avoids most of that political/industrial area of the list early on. Whereas if you're just going to go from 1 to 6k, I'd look at this newer list -- especially if you're thinking of doing it in tandem with iknow.

RawToast Wrote:Whereas if you're just going to go from 1 to 6k, I'd look at this newer list

Exactly what I wanted to know, thank you!

RawToast Wrote:If you're looking at using Nukemarine's guide, then I wouldn't bother with this list. As the guide avoids most of that political/industrial area of the list early on.

The TSV file on my website also includes a column for word frequency, so you can for example sort by it in a spreadsheet application and remove the least or most frequent lines. The word frequency list I used is an average of frequency lists based on anime and drama subs, websites, and novels.

Thanks!
But as I'm going to do the whole 6000, I'll stick to the default order.

I guess frequency lists will be useful for studying after Core.

You could make use of cb's Kanji Word Association Tool.

Quote:Kanji Word Association Tool was created for students who want to learn kanji and words at the same time in the most optimal fashion possible. Based on a user-provided list of kanji, this tool will generate a list of words that are associated with each kanji and ensure that each word consists only of kanji that you have already studied up to that point and kana. In addition, words are sorted by frequency and no duplicate words are used.

With that tool you can order whatever vocabulary list you're studying, for example, to study words according to RTK order.

For example, if you have these words in a list:

夏休み、連休、夏、休憩、憩い、休む、連れる

And order them according to RTK, you will get approximately this:

休む、夏、夏休み、連れる、連休、憩い、休憩

It seems really nice if you want to study kanji and vocabulary at the same time.
Personally, I'm fine doing RTK before everything else.

EDIT:
I had a better look and noticed it can actually be used as a tool for more advanced studies too. It makes use of a dictionary to pick up the words, I suppose.

Useful, but it limits you to study words outside context...

killua Wrote:Useful, but it limits you to study words outside context...

Not necessarily.

You can use decks with the word in kanji in a field and and example sentence in a different field, and sort the sentences according to the target word.

You can even use something like Rikaisama to get words and their respective sentences from actual texts you read (blogs, news articles, novels, etc) or Epwing2Anki to extract words with example sentences from Epwing dictionaries (including bilingual and monolingual dictionaries) and sort the sentences according to RTK, JLPT, Kanken, or any other order you follow when studying kanji.

Thanks! Great advice.