文書のH5クラスター分析におけるラベリング #858
Unanswered
Saitoh1225
asked this question in
1.2 Q&A (in Japanese)
Replies: 2 comments
-
お使いのKH Coderは最新版(または最近の版)でしょうか? 最近のバージョンでは,文書のデンドログラムの画面で「調整」ボタンをクリックすれば,ラベルとして表示する列をお選びいただけるかと思います。 |
Beta Was this translation helpful? Give feedback.
0 replies
-
早速にありがとうございます。 相変わらず精力的にバージョンアップをされていて恐れ入りました。 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
以前の掲示板で関連質問とそれに対する回答が以下のとおりありましたが、それを見てもまだ諦めきれない点があったので質問させていただきます。
>■ [No.3414] 文と文の類似度について 投稿者:たつこ 投稿日:2017/12/29(Fri) 15:48:42
>あるいは、現在の分析対象ファイルとなっている、「ファイル名_txt0.txt」
>のようなファイルを開き、「---cell---」となっている部分を分かりやすい
>記号や番号に直すかでしょうか。
koichi.nihon.to/cgi-bin/bbs_khn/khcf.cgi?list=&no=3414&mode=allread&page=0
文書単位H5による文書のクラスター分析で、分析対象のA列のセルをラベリング(識別)したいのですが、上記の対応をとるしかないとのことでした。
しかし、この作業は非常に煩雑なので、どうにか避けられないものかと思案中です。
対応分析であれば、
◆ツール→抽出語→対応分析→抽出語×外部変数
により、分析対象A列の隣、B列にラベル情報を入れておけば、これを外部変数としてラベリングできます。
同様に文書のクラスター分析でもラベリングできないものかと期待してしまうところですが、難しいのでしょうか?
よろしくお願いします。
Beta Was this translation helpful? Give feedback.
All reactions