Topic-Sensitive PageRank (commonly referred to as TSPR) is a context-sensitive ranking algorithm for web search developed by Taher Haveliwala while at Stanford University, and thought to be used by Google for the purpose of indexing and ranking search results in the search engine results pages, although no evidence has been shown of it in practice.
Algorithm
Topic-Sensitive PageRank is based on the PageRank algorithm, and provides a scalable approach for personalizing search rankings using Link analysis.
References
Topic-Sensitive PageRank Wikipedia(Text) CC BY-SA