enigma1 - 11:19 am on Dec 3, 2011 (gmt 0)
E.g. the letters "M" "o" "c" "b" etc. display the same in Russian and in English, but they actually have a different code.
Ok I see now, its visual spoofing so even if you copy/paste the strings you may not see the problem, subject to the editor in use.
Since from the articles you would expect ascii characters in the range of 0-127 decimal, 7-bit character set in other words setup a filter - you should in this type of business. Or do what lucy said above. An editor macro or a browser addon will be more useful in this case.
With the lower ascii range, say the original text is converted to utf-8, stored in the database you run the filter against it and see if any differences occur. Or just apply the filter first and use its result for the content search. For instance:
$str = preg_replace('/[^(\x00-\x7F)]*/','',$str);
That's in php, In your case above it will strip all the foreign characters and you will be aware of the fraud right at the beginning. Kinda late but next time you will have something to counter it.