Scalable Event-Based Clustering of Social Media Via Record Linkage Techniques.
Timo ReuterPhilipp CimianoLucas DrumondKrisztian BuzaLars Schmidt-ThiemePublished in: ICWSM (2011)
Keyphrases
- record linkage
- social media
- clustering algorithm
- duplicate detection
- parameter free
- privacy preserving
- k means
- entity resolution
- clustering method
- multiple databases
- data cleaning
- social networks
- census data
- approximate matching
- categorical data
- hierarchical clustering
- linked data
- cluster analysis
- high dimensional data
- artificial intelligence
- outlier detection
- disclosure risk
- data analysis
- information systems
- group membership