Authoritative sources in a hyperlinked environment

By Jon M. Kleinberg. 1999

How researchers see this paper

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

Citations in the corpus

ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software (2014, methodology): “Note that here we improperly use the concepts of Hub and Authority defined by Kleinberg [19].”
CodeT: Code Generation with Generated Tests (2022, methodology): “Inspired by the work for bipartite graph ranking (Kleinberg, 1999; He et al., 2016; Yang et al., 2020), the scores of xi and yj can be calculated as:”
struc2vec: Learning Node Representations from Structural Identity (2017, background): “CCS CONCEPTS •Computing methodologies→Unsupervised learning; Learning latent representations; •Arti cial Intelligence→ Learning; KEYWORDS feature learning; node embeddings; structural identity”
The science of science: from the perspective of complex systems (2017, methodology): “A method based on the PageRank [173] and the Hyperlink-Induced Topics Search (HITS) [248] algorithms is then proposed to rank journals.”
Propagation of trust and distrust (2004, methodology): “However, the value of co-citation has been proven for web pages by the success of the HITS algorithm [15], so we included it and the other basis elements.”
A Bayesian Approach to Discovering Truth from Conflicting Sources for Data Integration (2012, methodology): “The original HubAuthority (HITS) algorithm was developed to compute quality for webpages [9], AvgLog is a variation.”
McRank: Learning to Rank Using Multiple Classification and Gradient Boosting (2007, background): “This is different from the query-independent s atic rankingbased on, for example, “page rank” [3] or “authorities and hubs” [12], which may, at least conceptually, serve as an imp ortant “feature” for dynamic ranking or to guide the generation of a list of URLs fed to the dynamic r anker.”
Recommendations in location-based social networks: a survey (2015, methodology): “, hypertext induced topic search ( HITS ) and PageRank , to identify experienced users and interesting locations ; and 3 ) collaborative filtering ( CF ) recommendation , which infers a user ' s preferences from historical behavior ( such as from a location history ) .”
Collective Opinion Spam Detection: Bridging Review Networks and Metadata (2015, methodology): “These scores are formulated in terms of one another, and are computed by an iterative algorithm similar to HITS [12].”
Web mining research: a survey (2000, methodology): “rganization charts. In our framework, their research could be classiﬁed as a combination of Web structure and content mining. Some algorithms have been proposed to model the Web topology such as HITS [73], PageRank [16] and improvements of HITS by adding content information to the links structure [2”
Iterative Classification in Relational Data (2000, methodology): “" Co-training " is an iterative approach to learning models (Blum and Mitchell 1998, Mitchell 1999) that was applied to the WebKB labeling task.”
Generalized Hirsch h-index for disclosing latent facts in citation networks (2007, methodology): “All the metrics used so far in citation analysis, even those which are based on popular spectral techniques, like HITS [20], PageRank [25] and its variations for bibliometrics, like [7], present one or more of the following drawbacks (see also [16]):”