Research Interest : another undisclosed (and redundant) algorithm by ResearchGate

The scholarly social network ResearchGate (RG) started promoting a new composite indicator: Research Interest. That score is built upon a weighting scheme of citations, recommendations, full-text reads, and other reads by RG members. Here I show that the indicator suffers from, at least, two significant issues, which undermine its utility as an alternative metric. Lack of transparency is the former since there are clues that its underlying algorithm is more complex than what is known according to the information conveyed by RG. Redundancy is another issue because the score is essentially driven by the number of citations and seems to bring negligible additional knowledge in the altmetrics landscape.