Matches in SemOpenAlex for { <https://semopenalex.org/work/W2044687533> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W2044687533 endingPage "93" @default.
- W2044687533 startingPage "73" @default.
- W2044687533 abstract "Aggregating multiple correlated data sources may help improve the learning performance of a given task. For example, in recommendation problems, one can aggregate (1) user profile database (e.g., genders, age, etc.), (2) users’ log data (e.g., clickthrough data, purchasing records, etc), and (3) users’ social network (useful in social targeting) to build a recommendation model. All these data sources provide informative but heterogeneous features. For instance, user profile database usually has nominal features reflecting users’ background, log data provides term-based features about users’ historical behaviors, and social network database has graph relational features. Given multiple heterogeneous data sources, one important challenge is to find a unified embedding feature subspace that captures the knowledge from all sources. To this aim, we propose a principle of collective component analysis (CoCA), in order to find the optimal embedding across a mixture of vector-based features and graph relational features. The CoCA principle is to find a feature subspace with maximal variance under two constraints. First, there should be consensus among the projections from different feature spaces. Second, the similarity between connected data (in any of the network databases) should be maximized. The optimal solution is obtained by solving an eigenvalue problem. Moreover, we discuss how to use prior knowledge to distinguish informative data sources, and optimally weight them in CoCA. Since there is no previous model that can be directly applied to solve the problem, we devised a straightforward comparison method by performing dimensionality reduction on the concatenation of the data sources. Three sets of experiments show that CoCA substantially outperforms the comparison method." @default.
- W2044687533 created "2016-06-24" @default.
- W2044687533 creator A5011510969 @default.
- W2044687533 creator A5036357902 @default.
- W2044687533 date "2014-03-01" @default.
- W2044687533 modified "2023-09-25" @default.
- W2044687533 title "Heterogeneous Embedding via Aggregating Multiple Sources" @default.
- W2044687533 cites W1496866547 @default.
- W2044687533 cites W2037603696 @default.
- W2044687533 cites W2048679005 @default.
- W2044687533 cites W2100235303 @default.
- W2044687533 cites W2137678375 @default.
- W2044687533 cites W2153959628 @default.
- W2044687533 cites W2158933803 @default.
- W2044687533 cites W2172013605 @default.
- W2044687533 cites W87822204 @default.
- W2044687533 doi "https://doi.org/10.1007/s40745-014-0006-8" @default.
- W2044687533 hasPublicationYear "2014" @default.
- W2044687533 type Work @default.
- W2044687533 sameAs 2044687533 @default.
- W2044687533 citedByCount "2" @default.
- W2044687533 countsByYear W20446875332018 @default.
- W2044687533 countsByYear W20446875332020 @default.
- W2044687533 crossrefType "journal-article" @default.
- W2044687533 hasAuthorship W2044687533A5011510969 @default.
- W2044687533 hasAuthorship W2044687533A5036357902 @default.
- W2044687533 hasBestOaLocation W20446875331 @default.
- W2044687533 hasConcept C114614502 @default.
- W2044687533 hasConcept C124101348 @default.
- W2044687533 hasConcept C132525143 @default.
- W2044687533 hasConcept C138885662 @default.
- W2044687533 hasConcept C154945302 @default.
- W2044687533 hasConcept C2776401178 @default.
- W2044687533 hasConcept C32834561 @default.
- W2044687533 hasConcept C33923547 @default.
- W2044687533 hasConcept C41008148 @default.
- W2044687533 hasConcept C41608201 @default.
- W2044687533 hasConcept C41895202 @default.
- W2044687533 hasConcept C70518039 @default.
- W2044687533 hasConcept C80444323 @default.
- W2044687533 hasConcept C83665646 @default.
- W2044687533 hasConcept C87619178 @default.
- W2044687533 hasConceptScore W2044687533C114614502 @default.
- W2044687533 hasConceptScore W2044687533C124101348 @default.
- W2044687533 hasConceptScore W2044687533C132525143 @default.
- W2044687533 hasConceptScore W2044687533C138885662 @default.
- W2044687533 hasConceptScore W2044687533C154945302 @default.
- W2044687533 hasConceptScore W2044687533C2776401178 @default.
- W2044687533 hasConceptScore W2044687533C32834561 @default.
- W2044687533 hasConceptScore W2044687533C33923547 @default.
- W2044687533 hasConceptScore W2044687533C41008148 @default.
- W2044687533 hasConceptScore W2044687533C41608201 @default.
- W2044687533 hasConceptScore W2044687533C41895202 @default.
- W2044687533 hasConceptScore W2044687533C70518039 @default.
- W2044687533 hasConceptScore W2044687533C80444323 @default.
- W2044687533 hasConceptScore W2044687533C83665646 @default.
- W2044687533 hasConceptScore W2044687533C87619178 @default.
- W2044687533 hasIssue "1" @default.
- W2044687533 hasLocation W20446875331 @default.
- W2044687533 hasOpenAccess W2044687533 @default.
- W2044687533 hasPrimaryLocation W20446875331 @default.
- W2044687533 hasRelatedWork W1492779159 @default.
- W2044687533 hasRelatedWork W2023181450 @default.
- W2044687533 hasRelatedWork W2774254162 @default.
- W2044687533 hasRelatedWork W2950918317 @default.
- W2044687533 hasRelatedWork W4283014405 @default.
- W2044687533 hasRelatedWork W4300803113 @default.
- W2044687533 hasRelatedWork W4306745664 @default.
- W2044687533 hasRelatedWork W4320070445 @default.
- W2044687533 hasRelatedWork W2107103981 @default.
- W2044687533 hasRelatedWork W2886616187 @default.
- W2044687533 hasVolume "1" @default.
- W2044687533 isParatext "false" @default.
- W2044687533 isRetracted "false" @default.
- W2044687533 magId "2044687533" @default.
- W2044687533 workType "article" @default.