Ranking USPTO Patent Documents by Importance Using Random Surfer Method (PageRank)

This study attempted to construct an alternative method for ranking patents by their importance in order to improve the existing techniques of determining citation weighted patent stocks. We apply the random surfer PageRank method to the problem of ranking the importance of the patent documents issued to inventors by the United States Patent and Trademark Office. PageRank patent importance weights turn out to differ systematically from the weights based on the number of backward or forward citations. Yet PageRanks perform similarly to the knowledge transfer indicator, which is the ratio of forward to backward citations