Cube Interestingness: Novelty, Relevance, Peculiarity and Surprise

TL;DR

This paper introduces a multidimensional approach to measure the interestingness of data cube queries by defining scores for novelty, relevance, surprise, and peculiarity, supported by algorithms and a user study.

Contribution

It proposes a comprehensive framework for quantifying interestingness in data cubes, including new measures, algorithms, and an empirical user study analysis.

Findings

Significance of different interestingness dimensions identified

Interestingness scores evolve over time during user interaction

User behavior correlates with specific interestingness dimensions

Abstract

In this paper, we discuss methods to assess the interestingness of a query in an environment of data cubes. We assume a hierarchical multidimensional database, storing data cubes and level hierarchies. We start with a comprehensive review of related work in the fields of studies of human behavior and computer science. We define the interestingness of a query as a vector of scores along difference dimensions, like novelty, relevance, surprise and peculiarity and complement this definition with a taxonomy of the information that can be used to assess each of these dimensions of interestingness. We provide both syntactic (result-independent) checks and extensional (result-dependent) measures and algorithms for assessing the different dimensions of interestingness in a quantitative fashion. We also report our findings on a user study that we conducted, analyzing the significance of each dimension, its evolution over time and the behavior of the study's participants.