Paul Hsiung

Graduate Student

My Website

Research Interests

Datamining, Decision Trees, and Statistics.

Tags

Link Analysis

Papers

• Alias Detection in Link Data Sets (2005)
  Combining string similarity with contextual similarity when searching for aliases using active learning.
• Alias Detection in Link Data Sets (2004)
  An active learning approach to deciding whether two names correspond to the same entity, combining string similarity information and context similarity.

Software

• Fast EM Clustering
  Rapid Learning of Gaussian Mixture Models from large datasets.
• Many Names One Person
  This program will identify the most likely aliases for a given query name, using a semi-supervised learning approach. The program will then ask the user to confirm the validity of these most likely aliases.