# How to remove overlaps using R

So I have some data which talks about unique visits for each site. Now, these have overlaps since people visit multiple websites. SO a user who visited `site1` could have also visited `site3` and `site4`. So for the unique visits in site1 might be `20M` instead of `23M`. So in order to remove the overlaps, we did get a percentage overlap between each pair of websites. Based on this how can I calculate the actual unique visits for each website .

``````tibble::tribble(
~X1, ~unique_visits, ~site1, ~site2, ~site3, ~site4, ~site5, ~site6, ~site7,
"site1",       23873274,    100,   96.1,   95.6,     95,   91.6,   96.6,   92.9,
"site2",        4249486,   54.1,    100,   46.2,   46.5,   56.9,   77.5,   43.2,
"site3",         887786,   47.2,   40.5,    100,   41.5,   38.5,   55.3,     85,
"site4",        3727497,   41.7,   36.3,   36.9,    100,   38.6,   56.3,   51.5,
"site5",        1833995,   23.7,   26.2,   20.2,   22.8,    100,     28,   87.5,
"site6",        1617476,    5.1,    7.3,    5.9,    6.8,    5.7,    100,    2.9,
"site7",         760829,    0.2,    0.2,    0.4,    0.3,    0.8,    0.1,    100
)

``````

What do you mean by unique visits? From the data it looks like there were 23873274 people who visited site1, and 96.1% of them also visited site2, 95.6% of them also visited site3, etc. Do you want to know how many ONLY visited site1? Hm, tricky.