Copyscape - What percentage uniqueness

Am eager to find out about that too...

the various sites use different ways to determine 'uniqueness'. Copyscape, as I understand it, uses a 3 word (or longer) shingle algo (and the google API). Dupecop uses something that seems more akin to a levenstein/sim text system. If all you want to do is 'pass copyscare', you can do it for free by screwing up every third word. There's even a free tool to do this (but it REALLY isn't recommended, it's just to show how easy it is to 'pass copyscape'). It's here - Pass Copyscape

detecting uniqueness isn't as easy as it seems. For example, you can 'mash up' stuff from multiple sources. Some systems will then say it's unique, some won't. IMHO, it's best to try and use the same kind of detection criteria the search engines use.

for example, the php similar text function thinks this:-

IMHO, it's best to try and use the same kind of detection criteria the search engines use. detecting uniqueness isn't as easy as it seems. Some systems will then say it's unique, some won't. For example, you can 'mash up' stuff from multiple sources.

is only about 34% the same as the paragraph it is based on. If reordering a few sentences that way can fool the standard algos, what do you think the se's use? That's right. Something else.

Anyway, I digress, yes, there are people who use tools and software to increase their productivity while writing. And there are those who claim not to (which begs the question of how they manage to upload their 'hand-written' epic onto the net, and how they spellcheck it, etc - last time I looked, 'notepad' was a software tool that helps you write).

I think this is the correct thread to ask my doubt...and of course my post in this forum..
when i go for using CopyScape...there is a limit of words...2000 words...

so should I test against a 500 word article to get more accurate result..?? or it is fine to combine 4 articles and test them in a single run...??

I know the thread started is quite against CS itself...but for me as a writer..it's still necessary...

What about ArticleChecker? Does anyone use that? As an aspiring ghostwriter, I always run my own stuff through there and it always comes back very clean. It's a case of "I know I'm not plagiarizing and this article is unique, but I want my clients to know it too."

When I write an article, I do my research but I NEVER just take something else off the net and change a few words here and there. I don't think I could live with myself if I tried to pass someone else's work off as my own. Just don't have it in me.

So the only way I know how to prove it is by using these copyscape-type sites. I wish there was a better way.

Have a great day, all.
Debbie

I would like to ask you how much percentage of uniqueness do you consider ok to pass the copyscape?

Copyscape is truly not as good as most of us think.
I have had ghostwritten articles rejected by EZA for
duplicate content but they went through Copyscape
premium fine. But that was in the past. Right now
I use a better system than Copyscape and for $0

I will work on writing a very comprehensive thread
about how anyone can also use this $0 combination
of services to get better results than using Copyscape.

Kingsley

well - not quite, but almost. No-one comes up with something new by reaching into a cloud. So yes - some people cut and paste from all kinds of different directions. Ezine articles says it uses humans to check, but they also use software. I have The Best Spinner - the most powerful article spinner software., a content spinner and I want more than 80% originality, or I'm not happy. Do your writers snag article pieces dfrom here and there? probably. They get paid 3 bucks for 500 words? You expect originality..? If Copyscape accepts 30% then copyscape is not as good as I thought.

Copyscape tags articles as duplicates only if they have sequences of four or more words that match. That is, Copyscape would see the following two sentences as duplicates:

Notice that "The quick brown fox" and "over the lazy dog" are each four word long segments, and are unchanged. Copyscape sees that as a duplicate. However, Copyscape will not see these two sentences as duplicates:

Notice that the segments which remain the same in these two sentences ("The quick brown" and "jumped over the" and "dog") are all less than four words long. Copyscape is unable to spot duplicates of less than four words, therefore these two sentences would pass as unique.

Copyscape uses the four word sequence as a measure of duplication because a three or two word sequence is far too likely to appear in articles which are not duplicates. However, a four word sequence is much less statistically likely to occur in two unrelated articles.

What this means it that you need to modify every fourth word in your article in order to pass Copyscape. Technically that's 25% unique, but it's not so much a matter of the percentage as it is the sequence of words.

Any time I hire someone to write content, I expect it to show up as 100% unique in Copyscape. I know that may sound strict to some--but that's always white I shoot for when I rewrite PLR, so I definitely expect it when I hire out for original content.

I just use Copyscape because most customers want me to. Honestly, I don't think to much of it either, because all I got yet were two false positives.

I mean if I write an article about credit cards, the following matches are only natural, aren't they?

Personally I write and do my own research all by myself, but I do know some people that change a few phrases around and post it.