Scaling personalized web search

By Glen Jeh, Jennifer Widom. 2003

1,414 citations recorded by the source corpus.

Subfields: Information Systems, Signal Processing

How researchers see this paper

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

In Information Systems

In Signal Processing

Citations in the corpus

Position-aware Graph Neural Networks (2019, background): “Position-based similarities can be computed via the shortest path distance, or, for example, personalized PageRank (Jeh & Widom, 2003).”
Deep Graph Infomax (2018, background): “While powerful—and related to traditional metrics such as the personalized PageRank score (Jeh & Widom, 2003)—random walk methods suffer from known limitations.”
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.”
Exposing the probabilistic causal structure of discrimination (2015, methodology): “, vn , we perform a personalized PageRank [16] computation with respect to v1, .”
FIRM: An Intelligent Fine-Grained Resource Management Framework for SLO-Oriented Microservices (2020, neutral): “Microscope [61] and MicroRCA [124] are both designed to identify abnormal services by constructing service causal graphs that model anomaly propagation and by inferring causes using graph traversal or ranking algorithms [51].”
Subgraph-augmented Path Embedding for Semantic User Search on Heterogeneous Social Network (2018, methodology): “Earlier graph semantic search work such as Personalized PageRank [17] and SimRank [16] often consider homogeneous networks as input, and they do not differentiate semantic classes.”
Distance-Aware DAG Embedding for Proximity Search on Heterogeneous Graphs (2018, positive): “Pioneer work often tries to measure the semantic proximity by paths connecting two nodes (Jeh and Widom 2003; Backstrom and Leskovec 2011; Shi et al. 2017).”
Deeper Inside PageRank (2004, neutral): “Several groups have suggested incorporating text information into the link analysis [16, 38, 60, 67 , 103].”
Extrapolation methods for accelerating PageRank computations (2003, methodology): “Furth ermore, recent approaches to personalized and topic-sensiti ve PageRank schemes [11, 19, 13] require computing many PageRank vectors, each biased towards certain types of pages.”
Personalizing search via automated analysis of interests and activities (2005, methodology): “For example, in the same way that we personalize content-based matching, we could personalize link-based computations as proposed by Jeh and Widom [10].”
Random walks and diffusion on networks (2016, background): “An alternative choice is a “personalized PageRank” [23, 288–291, 301–304], in which the preference vector is localized around one node or a small number of nodes (which can be helpful for applications to community detection [68]).”
Using ODP metadata to personalize search (2005, methodology): “[12] improves this somewhat, allowing the algorithm to bias on any subset of a given set of pages (H).”
FAST-PPR: scaling personalized pagerank estimation for large graphs (2014, background): “Also, Jeh and Widom [2] proposed a scheme for PageRank estimation us-ing intermediate estimates from a ﬁxed skeleton of target nodes.”
On the Embeddability of Random Walk Distances (2013, methodology): “In contrast, hitting time and PPR are asymmetric [26, 18], i.”
ObjectRank: Authority-Based Keyword Search in Databases (2004, methodology): “Jeh and Widom [22] present a method to efficiently calculate the PageRank values for multiple base sets, by precomputing a set of partial vectors which are used in runtime to calculate the PageRanks.”