Topic-sensitive PageRank

By Taher H. Haveliwala. 2002

2,160 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics, Information Retrieval & Search

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Statistical and Nonlinear Physics

In Information Retrieval & Search

Citations in the corpus

Knowledge Guided Text Retrieval and Reading for Open Domain Question Answering (2019, methodology): “Although we considered Personalized PageRank (Haveliwala, 2002) following Sun et al.”
Predict then Propagate: Graph Neural Networks meet Personalized PageRank (2018, background): “Instead of viewing this equation as a combination of a dense fully personalized PageRank matrix with the prediction matrix, we can also view it as a variant of topic-sensitive PageRank, with each class corresponding to one topic (Haveliwala, 2002).”
Supervised random walks: predicting and recommending links in social networks (2010, methodology): “PageRank [25] and variants like Personalized PageRank [17, 15] and Random Walks with Restarts [31] are popular methods for ranking nodes on graphs.”
Abstractive Document Summarization with a Graph-Based Attentional Neural Model (2017, methodology): “Given a topic T , the topic-sensitive PageRank is similar to Eq.”
Challenging the Long Tail Recommendation (2012, background): “• Personalized PageRank: The Personalized PageRank(PPR) [11] algorithm tends to suggest popular items since the stationary distribution for the personalized pagerank w.r.t. starting nodes S is localized around the starting nodes, which combines the factors of similarity and popularity.”
Higher-order Web link analysis using multilinear algebra (2005, background): “Haveliwala [23] introduced a topic-sensitive PageRank that pre-computes several PageRank vectors that are biased towards particular topics.”
SoK: The Evolution of Sybil Defense via Social Networks (2013, background): “Since ACL is based on the normalized stationary distribution of the Personalized PageRank [13] (PPR) random walk, we start”
Large Language Models on Graphs: A Comprehensive Survey (2023, methodology): “In this direction, LTRN [57] proposes a novel GNN architecture with personalized PageRank [94] and attention mechanism for structure encoding while adopting BERT [23] as the language model.”
Web search personalization with ontological user profiles (2007, methodology): “Multiple Topic-Sensitive PageRank values are computed o.-line.”
A large-scale evaluation and analysis of personalized search strategies (2007, methodology): “In [10], multiple Personalized PageRank scores, one for each main topic of ODP, were used to enable “topic sensitive” web search.”
Automatic identification of user goals in Web search (2005, background): “We did not observe any noticeable noise from link spams or mirror sites for our benchmark queries, but existing techniques for spam and mirror detection [18, 19, 20, 21] may be used to avoid any potential issue.”
ClusCite: effective citation recommendation by information network-based clustering (2014, methodology): “(see comparison between PopRank [19] and ClusCite) Haveliwala [9] personalizes the PageRank algorithm by considering query topics to derive query-specific authority score.”
Graph neural networks for ranking Web pages (2005, background): “…Hagenbuchner Ah Chung Tsoi Marco Maggini University of Wollongong Australian Research Council University of Siena Wollongong, Australia Canberra, Australia Siena, Italy markus@uow.edu.au ahchung.tsoi@arc.gov.au maggini@ing.unisi.it Abstract An arti.cial neural network model, capable of…”
The Effect of Recommendations on Network Structure (2016, methodology): “Namely, similar in spirit to personalized PageRank [16], we assume recommendations for user u are generated by performing a two-step random walk starting at u , as stated formally in Deﬁnition 1 below.”
Efficient Global 2D-3D Matching for Camera Localization in a Large-Scale 3D Map (2017, background): “Moreover, there are important differences between our method and PageRank, which make our method particularly relevances for IBL: (1) We use bi-directed graph with two-way weights to capture co-visibility neighborhood relations, in contrast to Google’s undirected “Web graph” with binary (1/0) hyperl”