The paper proposes a new metric called PaperRank that can be used to evaluate the trustworthiness of academic articles. PaperRank is based on the PageRank algorithm, which is commonly used to rank node reputation in networks. By analyzing the citation network of academic articles, PaperRank can determine the extent to which a scientific community trusts the claims made in a given paper.

To calculate PaperRank scores, the authors use a probabilistic model called the Gamma Mixture Model. This model allows them to apply a fitting algorithm to a set of PageRank scores and infer a probability distribution representing the trust that communities place in their research publications. In many knowledge graph applications, it is important to have a scalable way of computing trust for the evidence cited within those graphs. Since many knowledge sources cite academic articles, PaperRank can provide a useful method of computing confidence scores in probabilistic knowledge graphs.

**Note: This work was never submitted for review by Jamie McCusker**