Web Information Systems Engineering – WISE 2007
DOI: 10.1007/978-3-540-76993-4_10
|View full text |Cite
|
Sign up to set email alerts
|

BEIRA: A Geo-semantic Clustering Method for Area Summary

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(1 citation statement)
references
References 14 publications
0
1
0
Order By: Relevance
“…To address the high dimensionality of the vectors and improve performance, we apply the LSI (Latent Semantic Indexing) dimensionality reduction algorithm [Deerwester et al, 1990]. Having initially calculated the pair-wise photo geographic distance (taken as their spherical distance based on the Haversine formula) and their mapping onto a number of the most significant semantic dimensions (LSI results), photos are clustered via an algorithm that combines both the geographic and semantic types of similarity [Masutani and Iwasaki, 2007]. For each cluster, the platform generates and stores a model containing: a list of information about the photos assigned to the cluster (their numeric identifiers, URLs, geolocation), a summarized description containing terms and their estimated "significance" for the cluster (based on their frequency), and a cluster popularity score (for ranking purposes) in terms of the number of photos which belong to them and the number of unique users-photographers.…”
Section: Backend Processes and Functionalitymentioning
confidence: 99%
“…To address the high dimensionality of the vectors and improve performance, we apply the LSI (Latent Semantic Indexing) dimensionality reduction algorithm [Deerwester et al, 1990]. Having initially calculated the pair-wise photo geographic distance (taken as their spherical distance based on the Haversine formula) and their mapping onto a number of the most significant semantic dimensions (LSI results), photos are clustered via an algorithm that combines both the geographic and semantic types of similarity [Masutani and Iwasaki, 2007]. For each cluster, the platform generates and stores a model containing: a list of information about the photos assigned to the cluster (their numeric identifiers, URLs, geolocation), a summarized description containing terms and their estimated "significance" for the cluster (based on their frequency), and a cluster popularity score (for ranking purposes) in terms of the number of photos which belong to them and the number of unique users-photographers.…”
Section: Backend Processes and Functionalitymentioning
confidence: 99%