Subgroup Discovery Algorithms： A Survey and Empirical Evaluation

Subgroup discovery is a data mining technique that discovers interesting associations among different variables with respect to a property of interest. Existing subgroup discovery methods employ different strategies for searching, pruning and ranking subgroups. It is very crucial to learn which features of a subgroup discovery algorithm should be considered for generating quality subgroups. In this regard, a number of reviews have been conducted on subgroup discovery. Although they provide a broad overview on some popular subgroup discovery methods, they employ few datasets and measures for subgroup evaluation. In the light of the existing measures, the subgroups cannot be appraised from all perspectives. Our work performs an extensive analysis on some popular subgroup discovery methods by using a wide range of datasets and by defining new measures for subgroup evaluation. The analysis result will help with understanding the major subgroup discovery methods, uncovering the gaps for further improvement and selecting the suitable category of algorithms for specific application domains.

On the Expressive Power of Logics on Constraint Databases with Complex Objects

We extend the constraint data model to allow complex objects and study the expressive power of various query languages over this sort of constraint databases.The tools we use come in the form of collapse results which are well established in the context of first-order logic.We show that the natural-active collapse with a condition and the activegeneric collapse carry over to the second-order logic for structures with o-minimality property and any signature in the complex value relations.The expressiveness results for more powerful logics including monadic second-order logic,monadic second-order logic with fix-point operators,and fragments of second-order logic are investigated in the paper.We discuss the data complexity for second-order logics over constraint databases.The main results are that the complexity upper bounds for three theories,MSO+(LIN),MSO+(POLY),and Inflationary DATALOGact^cv(SC,M)without powerset operator are∪iΣi^NC1,NCH=∪iΣi^NC,and AC^0/poly,respectively.We also consider the problem of query closure property in the context of embedded finite models and constraint databases with complex objects and the issue of how to determine safe constraint queries.

Examining the use of narrative constructs in data videos

Data videos are a highly impactful method of communication and are becoming a prevalent medium for communicating information.While the majority of current research focuses on the cinematic aspects of data videos,very little is known about the narrative methodologies involved.This paper presents our insights derived from an initial exploration of this area.We present a taxonomy based on the analysis of 70 existing data videos examining their narrative and visual approaches.We propose that our taxonomy can be used to explain the characteristics or design of data videos.Applying this taxonomy,we present our observations,including the trend of popular technologies applied in current data videos,the under-utilization of promising methods,and highlight research opportunities in the field.

Optimization Methods for Mixed Integer Weakly Concave Programming Problems

In this paper,we consider a class of mixed integer weakly concave programming problems(MIWCPP)consisting of minimizing a difference of a quadratic function and a convex function.A new necessary global optimality conditions for MIWCPP is presented in this paper.A new local optimization method for MIWCPP is designed based on the necessary global optimality conditions,which is different from the traditional local optimization method.A global optimization method is proposed by combining some auxiliary functions and the new local optimization method.Furthermore,numerical examples are also presented to show that the proposed global optimization method for MIWCPP is efficient.