This document describes using eigenvector centrality to rank authors and documents for a search engine on Amazon food reviews. It involves building a graph of authors based on shared products reviewed and calculating transition probabilities between authors. Eigenvector centrality is then calculated iteratively on this graph to rank authors. For a query, the top 25 documents by cosine similarity are ranked by multiplying their similarity score by the author's eigenvector centrality score to produce the final ranked lists. Correlations between the different ranking methods are calculated using Spearman's rank correlation coefficient.