FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W95662009> ?p ?o ?g. }

• W95662009 endingPage "1842" @default.
• W95662009 startingPage "1841" @default.
• W95662009 abstract "Relational data clustering is a form of relational learning that clusters data using the relational structure of data sets to guide the clustering. Many approaches for relational clustering have been proposed recently. The common assumption in much of this research is that relations have a binding tendency (Neville, Adler, & Jensen 2003; Bhattacharya & Getoor 2007; Taskar, Segal, & Koller 2001; Neville & Jensen 2006). That is, edges are assumed to appear more frequently within clusters than between clusters. This binding quality may be too strong an assumption. Bhattacharya & Getoor (2007) acknowledge that it is possible for a relation to provide “negative evidence,” where the presence of a relation between two objects implies that the objects belong in different clusters. If most of the edges in a relational set provide negative evidence, then this set should be considered to have a separating, rather than a binding, tendency. Consider a social network of university students containing both men and women. If the social network is partitioned by gender, edges for a dating relation will appear most frequently between the clusters. This can be visualized as an approximately bipartite structure between the two clusters that “pushes” them apart. Edges for a roommate relation, on the other hand, will appear most frequently within the clusters. This binding relation can be visualized as a net that “pulls” the objects in a cluster closer together, hence the name “Relational Push-Pull Model.” The prevailing approach to clustering when the relational tendency is not known a priori is stochastic block modeling. Nowicki & Snijders (2001) presented a method for automated learning of stochastic block models in which only object-object relations are observed. Recently, researchers (Handcock, Rafferty, & Tantrum 2007; Tallberg 2005; Anthony & desJardins 2007) have acknowledged that Nowicki & Snijders’ approach ignores a valuable source of information: the attributes associated with the data objects in the relational data set. I propose an extension to Nowicki & Snijders’s model in which relations exist probabilistically as a function of the connected objects’ features. My research has taken a generative approach to the clustering problem: I assume that each object and each relation" @default.
• W95662009 created "2016-06-24" @default.
• W95662009 creator A5087343389 @default.
• W95662009 date "2008-07-13" @default.
• W95662009 modified "2023-09-24" @default.
• W95662009 title "The relational push-pull model: a generative model for relational data clustering" @default.
• W95662009 cites W1800493452 @default.
• W95662009 cites W1986592861 @default.
• W95662009 cites W2096091969 @default.
• W95662009 cites W2109526276 @default.
• W95662009 cites W2144799688 @default.
• W95662009 cites W2148019918 @default.
• W95662009 cites W2988590051 @default.
• W95662009 hasPublicationYear "2008" @default.
• W95662009 type Work @default.
• W95662009 sameAs 95662009 @default.
• W95662009 citedByCount "1" @default.
• W95662009 countsByYear W956620092013 @default.
• W95662009 crossrefType "proceedings-article" @default.
• W95662009 hasAuthorship W95662009A5087343389 @default.
• W95662009 hasConcept C124101348 @default.
• W95662009 hasConcept C154945302 @default.
• W95662009 hasConcept C177264268 @default.
• W95662009 hasConcept C177877439 @default.
• W95662009 hasConcept C199360897 @default.
• W95662009 hasConcept C25343380 @default.
• W95662009 hasConcept C33923547 @default.
• W95662009 hasConcept C40207289 @default.
• W95662009 hasConcept C41008148 @default.
• W95662009 hasConcept C5655090 @default.
• W95662009 hasConcept C73555534 @default.
• W95662009 hasConcept C80444323 @default.
• W95662009 hasConceptScore W95662009C124101348 @default.
• W95662009 hasConceptScore W95662009C154945302 @default.
• W95662009 hasConceptScore W95662009C177264268 @default.
• W95662009 hasConceptScore W95662009C177877439 @default.
• W95662009 hasConceptScore W95662009C199360897 @default.
• W95662009 hasConceptScore W95662009C25343380 @default.
• W95662009 hasConceptScore W95662009C33923547 @default.
• W95662009 hasConceptScore W95662009C40207289 @default.
• W95662009 hasConceptScore W95662009C41008148 @default.
• W95662009 hasConceptScore W95662009C5655090 @default.
• W95662009 hasConceptScore W95662009C73555534 @default.
• W95662009 hasConceptScore W95662009C80444323 @default.
• W95662009 hasLocation W956620091 @default.
• W95662009 hasOpenAccess W95662009 @default.
• W95662009 hasPrimaryLocation W956620091 @default.
• W95662009 hasRelatedWork W1480810887 @default.
• W95662009 hasRelatedWork W1520138473 @default.
• W95662009 hasRelatedWork W1527141977 @default.
• W95662009 hasRelatedWork W1552293489 @default.
• W95662009 hasRelatedWork W1598108103 @default.
• W95662009 hasRelatedWork W2038425310 @default.
• W95662009 hasRelatedWork W2103611624 @default.
• W95662009 hasRelatedWork W2133612618 @default.
• W95662009 hasRelatedWork W2156241741 @default.
• W95662009 hasRelatedWork W2181414291 @default.
• W95662009 hasRelatedWork W2253944538 @default.
• W95662009 hasRelatedWork W2324853908 @default.
• W95662009 hasRelatedWork W2518738319 @default.
• W95662009 hasRelatedWork W2578330248 @default.
• W95662009 hasRelatedWork W2744384050 @default.
• W95662009 hasRelatedWork W2969835243 @default.
• W95662009 hasRelatedWork W3019865022 @default.
• W95662009 hasRelatedWork W83489895 @default.
• W95662009 hasRelatedWork W2181116559 @default.
• W95662009 hasRelatedWork W2188933034 @default.
• W95662009 isParatext "false" @default.
• W95662009 isRetracted "false" @default.
• W95662009 magId "95662009" @default.
• W95662009 workType "article" @default.