Pearson's correlation coefficient

This measure is a classic coefficient, which is quite applicable when comparing vectors. Its primary disadvantage is that when the intersection is estimated as low, then the correlation can be high by accident. To combat accidental high correlation, you can multiply by a factor of 50/min (50, rating intersection) or any other damping factor, the effect of which decreases with an increasing number of estimates. An example is shown here:

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.188.151.107