Current Projects

DFIS Project: Dynamic Frequent Itemset Mining and Hiding                   

This project focuses on the development of a test bed which deals with the important challenges of association rule hiding which is the most popular task of data mining field. These challenges are size of databases, dynamicity of databases, multiple support thresholds of itemsets and hiding sensitive knowledge before sharing databases. Project Leader is Ahmet Cumhur ÖZTÜRK.

Current tasks carried out within the project:

  • Frequent Itemsets under Multiple Support Thresholds/Utility Mining
  • Dynamic Frequent Itemset Mining under Multiple Support Thresholds
  • Frequent Itemset Hiding under Multiple Support Thresholds
  • Dynamic Frequent Itemset Hiding under Multiple Support Thresholds

 

FQP-LD Project: Federated Query Processing on Linked Data                                 

A large number of data providers publish and connect their structured data on the web as linked data. By this means, the web of data becomes a global data space. In this project, the main goal is to propose novel techniques that can be used in query processing and query execution parts of a typical federated query processing engine for linked data. Project Leader is Damla OĞUZ.

Current tasks carried out within the project:

  • Query Optimization based on Caching Results
  • Parallel Query Execution Strategies
  • Metadata Management
  • Dynamic Query Execution