Your "Referee Locator Results from the Web of Science" is a joke. Typically, it shows results that have absolutely nothing to do with the paper for which I am searching referees. The papers are from completely different sciences. I suspect that the algorithm is based only on similarity of keywords, not taking into account that in different disciplines (and even different areas of the same discipline) the same words have different meaning. I do not know how to do it technically, but the algorithm should take into account which science it is and what area of that science.
