Our system is able to detect identical and similar content and in some cases paraphrased content. Our algorithm currently can pick up paraphrasing that has specific restructuring along with similar word choices and phrasing.
However, since paraphrased text can be manipulated in a variety of ways, sometimes algorithms are not able to catch all of it. In these instances, to avoid false positives, we will not always show the paraphrased content as a result.
Our algorithm is always being updated and will continue to find more cases of paraphrased text in work.