Interactive Summarization and Exploration of Top Aggregate Query Answers

Area of Science:

• Data Mining
• Information Visualization
• Human-Computer Interaction

Background:

• Large datasets often yield numerous aggregate query answers, overwhelming users.
• Existing systems lack effective methods for summarizing and interactively exploring these answers.
• Making high-valued query answers more informative is crucial for data analysis.

Purpose of the Study:

• To develop a system for summarization and interactive exploration of high-valued aggregate query answers.
• To enhance user understanding of large sets of query results.
• To provide tools for parameter selection, cluster inspection, and visualization of clustering effects.

Main Methods:

• Clustering high-valued query answers based on common properties.
• Ensuring cluster diversity to avoid information redundancy.
• Developing optimization problems and efficient algorithms for clustering.
• Implementing a prototype with a graphical user interface (GUI).

Main Results:

• The system generates diverse, informative clusters of query answers.
• Efficient algorithms and optimizations were proposed and evaluated.
• The GUI facilitates interactive exploration and parameter adjustment.
• A user study confirmed the usability of the approach.

Conclusions:

• The proposed system effectively summarizes and enables interactive exploration of aggregate query answers.
• The approach enhances data analysis by providing diverse and informative clusters.
• The interactive features and GUI improve user comprehension and control over data exploration.