On Mining Summaries by Objective Measures of Interestingness

Abstract

Abstract. Knowledge discovery in databases is used to discover useful and understandable knowledge from large databases. A process of knowledge discovery consists of two steps, the data mining step and the evaluation step. In this paper, evaluating and ranking the interestingness of summaries generated from databases, which is a part of the second step, is studied using diversity measures. Sixteen previously analyzed diversity measures of interestingness are used along with three not previously considered ones, brought from different well-known areas. The latter three measures are evaluated theoretically according to five principles that a measure must satisfy to be qualified acceptable for ranking summaries. A theoretical correlation study between the eight measures that satisfy all five principles is presented based on mathematical proofs. An empirical evaluation is conducted using three real databases. Then, a classification of the eight measures is deduced. The resulting classification is used to reduce the number of measures to only two, which are the best over all criteria, and that produce non-similar results. This helps the user interpret the most important discovered knowledge in his decision making process.

Discovering diverse association rules from multidimensional schema

“…al in (Zbidi et al, 2006). These measures provide concrete statistical evidence that the diversity of the set of rules produced from our semi-automatically generated multidimensional schema is higher when compared to the rules generated without schema.…”

Measuring Diversity of Associations Rules Extracted from A Data Warehouse

“…In another research study, authors in [13] worked on ranking summaries in data sets. Authors deduced that Rae, International Journal of Knowledge Engineering, Vol.…”

“…The measures proposed by [13] were used by [14] in their research work which extracted patterns from a multi-dimensional schema. The authors applied Rae, Con and Hill measures in order to evaluate the extracted association rules for interestingness.…”

Measuring interestingness of discovered skewed patterns in data cubes

“…Subjective interestingness measures require some form of user input in determining the utility of a mined pattern [18,21,27]. Utilitybased measures have been used for objective-oriented association mining (for example, [26,33]), with user-specified objectives. In addition, numerous interestingness measures for summaries have been proposed in the literature including diversity (e.g., [12,33]), conciseness and generality [5], peculiarity [22][23][24][25], and surprisingness/unexpectedness [9].…”

“…Utilitybased measures have been used for objective-oriented association mining (for example, [26,33]), with user-specified objectives. In addition, numerous interestingness measures for summaries have been proposed in the literature including diversity (e.g., [12,33]), conciseness and generality [5], peculiarity [22][23][24][25], and surprisingness/unexpectedness [9]. Most of these methods with the exception of [9,10,20,30,32] have been applied for identifying patterns that have been mined from a given static dataset as opposed to providing guidance for navigating through multi-dimensional data cubes of graphs, which is the focus of our paper.…”