The Missing Link - A Probabilistic Model of Document Content and Hypertext Connectivity

By David A. Cohn, Thomas Hofmann. Neural Information Processing Systems, 2000

525 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/4ba18b2f35515f7f3ad3bc38100730c5808a52af

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

Link-Based Classification (2003, background): “Cohn and Hofmann (2001) propose probabilistic model for hypertext content and links.”
Learning author-topic models from text corpora (2010, background): “, Cutting et al. [1992]; McCallum et al.”
Mining Text Data (2012, methodology): “Readers who are interested in wrapper induction may refer to [31, 20] for in-depth surveys.”
The Intelligent surfer: Probabilistic Combination of Link and Content Information in PageRank (2001, background): “It can also be viewed as an analog for PageRank of Cohn and Hofman’s [7] var iation of HITS.”
Small-World Phenomena and the Dynamics of Information (2001, background): “This is clearly what was taking place in Milgram’s experiments: participants, using the information available to them, were estimating which of their acquaintances would lead to the shortest path through the full social network.”
Combining content and link for classification using matrix factorization (2007, methodology): “The major difference between the approach of [6] (PLSI+PHITS) and our approach is in the part of link analysis.”
Higher-order Web link analysis using multilinear algebra (2005, methodology): “In an approach that is very similar in spirit to ours, though different in the mathematical implementation, Cohn and Hofmann [11] combine probabilistic LSI (PLSI) and probabilistic HITS (PHITS) so that terms and links rely on a common set of underlying factors.”
A Unified View of Matrix Factorization Models (2008, methodology): “An extension of pLSI to two related matrices, pLSI-pHITS, consists of two pLSI models that share latent variables [39].”
Relational Topic Models for Document Networks (2009, background): “Approaches which generate document links through topic models (Nallapati and Cohen 2008; Cohn and Hofmann 2001; Sinkkonen et al. 2008; Gruber et al. 2008) treat links as discrete " terms " from a separate vocabulary.”
A Survey of Multi-View Representation Learning (2016, unknown): “F”
Topic modeling with network regularization (2008, background): “Cohn and Hofmann proposed a model which combines PLSA and PHITS on the web graph [7].”
Probabilistic Classification and Clustering in Relational Data (2001, background): “In clustering, the emphasis so far has been on dyadic data, such as word-document co-occurrence [Hofmann and Puzicha, 1999], document citations [Cohn and Chang, 2000], web links [Cohn and Hofmann, 2001; Kleinberg, 1998], and gene expression data.”
Data Mining for Web Personalization (2007, methodology): “Due to its great ﬂexibility, the PLSA model has been successfully used in a variety of application domains, including information retrieval [51], text learning [18, 60], and co-citation analysis [29, 30].”