A North Carolina State University study found significant shortcomings in Facebook's user interest discovery process.
The process of determining the interests of social network users is the basis for the effective display of advertisements. According to a study by North Carolina State University, even a platform as big as Facebook makes significant mistakes here. And this, in turn, leads to the fact that about 30% of advertisements do not reach their goal and turn out to be ineffective.
Analysts called the most important shortcoming in the process of determining the interests of users on the Facebook platform that it does not take into account the context in which users perform a particular action. The paper gives an example when a user speaks negatively in his post about a particular product, and the Facebook algorithm does not take into account the context, but highlights only the interest in the product itself. As a result, a user who directly stated that he did not like product "A" begins to see more ads with this product, which, of course, is completely ineffective.
Also, based on specially created test pages, the researchers saw that simply scrolling through the page without reading it already allowed Facebook to determine the interests of the user. Thus, the approximate share of incorrectly defined interests was, according to the study, about 33.22%.