Those characters you quoted here are Han (used by Chinese language), as they matched by the unicode character property \p{Han}.
You can perform a regular expression search in a plugin like so:
<?php
/**
* Plugin Name: Drop comments by chars
* Description: Delete comments which includes unicode characters of Han, Hangul and Cyrillic.
* Version: 2014.02.18
* Author: David Naber
* Licence: MIT
* Licence URI: http://opensource.org/licenses/mit-license.php
* Copyright: (c) 2014 David Naber
* @see http://wordpress.stackexchange.com/q/116973/31323
*/
/**
* check for the occurence of Han, Hangul and Cyrillic characters
*
* @param string $content
* @return bool
*/
function wpse_116973_has_unallowed_char( $content ) {
return (bool) preg_match( '~\p{Hangul}|\p{Han}|\p{Cyrillic}~u', $content );
}
/**
* @wp-hook comment_post
* @param int $comment_ID
* @param array $approved
* @return void
*/
function wpse_116973_trash_unallowed_comments( $comment_ID, $approved ) {
$comment = get_comment( $comment_ID );
if ( ! wpse_116973_has_unallowed_char( $comment->comment_content ) )
return;
wp_trash_comment( $comment_ID );
}
add_action( 'comment_post', 'wpse_116973_trash_unallowed_comments', 10, 2 );
The control function wpse_116973_has_unallowed_char() searches for all characters of Chinese (Han), Korean(Hangul) and Russian (Chyrillic) languages. The plugin move those comments to trash.
If someone just want to mark them as spam, use the filter pre_comment_approved like this:
/**
* @wp-hook pre_comment_approved
* @param bool $approved
* @param array $commentdata
* @return bool|string Returns FALSE, TRUE, or 'spam'
*/
function wpse_116973_allow_comment( $approved, $commentdata ) {
if ( wpse_116973_has_unallowed_char( $commentdata[ 'comment_content' ] ) )
return 'spam';
return $approved;
}
add_filter( 'pre_comment_approved', 'wpse_116973_allow_comment', 10, 2 );