Relevance Versus Diversity for Web Search Results

Greetings from the CSIRO ICT Centre at the Australian National University in Canberra, where I am attending a presentation on web search research by Kar Wai Lim of ANU & NICTA. The object of the exercise is to come up with web documents as the result of a search which are relevant but not all too much the same (diverse). For example, if you get ten copies of essentially the same document, that is not much use. It turns out that there is a mathematical equation which shows the trade-off between relevance and diversity. The details are published in a short paper, which I don't pretend to understand:

Kar Wai Lim, Scott Sanner, and Shengbo Guo. 2012. On the mathematical relationship between expected n-call@k and the relevance vs. diversity trade-off. In Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval (SIGIR '12). ACM, New York, NY, USA, 1117-1118. DOI=10.1145/2348283.2348497 http://doi.acm.org/10.1145/2348283.2348497