Evaluating the Correlation Between Objective Rule Interestingness Measures and Real Human Interest

Carvalho, Déborah Ribeiro; Freitas, Alex A.; Ebecken, Nelson F. F.

doi:10.1007/11564126_45

Cited by 32 publications

(38 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In Carvalho et al (2005) and Ohsaki et al (2004) various interestingness measures have been compared with real human interest, and the authors found that in many cases high ranking rules were considered uninteresting by the user. For example in Carvalho et al (2005) there was a positive correlation between an interestingness measure and real human interest only in 35.2% of studied cases.…”

Section: Mining Non-redundant Rulesmentioning

confidence: 99%

See 1 more Smart Citation

Scalable pattern mining with Bayesian networks as background knowledge

Jaroszewicz

Scheffer

Simovici

2008

Data Min Knowl Disc

View full text Add to dashboard Cite

We study a discovery framework in which background knowledge on variables and their relations within a discourse area is available in the form of a graphical model. Starting from an initial, hand-crafted or possibly empty graphical model, the network evolves in an interactive process of discovery. We focus on the central step of this process: given a graphical model and a database, we address the problem of finding the most interesting attribute sets. We formalize the concept of interestingness of attribute sets as the divergence between their behavior as observed in the data, and the behavior that can be explained given the current model. We derive an exact algorithm that finds all attribute sets whose interestingness exceeds a given threshold. We then consider the case of a very large network that renders exact inference unfeasible, and a very large database or data stream. We devise an algorithm that efficiently finds the most interesting attribute sets with prescribed approximation bound and confidence probability, even for very large networks and infinite streams. We study the scalability of the methods in controlled experiments; a case-study sheds light on the practical usefulness of the approach.Responsible editor: M. J. Zaki.

show abstract

Section: Mining Non-redundant Rulesmentioning

confidence: 99%

“…For example in Carvalho et al (2005) there was a positive correlation between an interestingness measure and real human interest only in 35.2% of studied cases. Also, for some datasets almost all measures gave good results and for others almost none.…”

Section: Mining Non-redundant Rulesmentioning

confidence: 99%

Scalable pattern mining with Bayesian networks as background knowledge

Jaroszewicz

Scheffer

Simovici

2008

Data Min Knowl Disc

View full text Add to dashboard Cite

show abstract

“…These measures are usually conflicting, i.e. an accurate rule is not necessarily interesting or easy to read, thus the searching process has to be multicriterial and dozens of such measures have been proposed and investigated [3,13].…”

Section: If "Some Conditions On the Values Of Predicting Attributes Amentioning

confidence: 99%

“…There are more than 40 objective measures for evaluating the interestingness of a rule [13], so choosing appropriate measures for the actual data characteristics is not a trivial task, as Carvalho et al clearly illustrated in their paper, too [3]. They analyzed the correlation between objective interestingness measures and the real human interest evaluated by experts from each domain, and proposed a ranking of objective quality measures.…”

Section: Interestingness Measuresmentioning

confidence: 99%

Interactive search of rules in medical data using multiobjective evolutionary algorithms

Zaharie

Lungeanu²,

Zamfirache

2008

Proceedings of the 10th Annual Conference Companion on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

In this work, we propose an approach for evolving rules from medical data based on an interactive multi-criteria evolutionary search: besides selecting the set of criteria and the sets of potential antecedent and consequent attributes, the user can also intervene in the searching process by marking the uninteresting rules. The marked rules are further used in estimating a supplementary optimization criterion which expresses the user's opinion on the rule quality and is taken into account in the evolutionary process.

show abstract

“…But, in the literature, they seek to find a correlation between real human interest and objective interestingness measures [5,[37][38][39]. BM_IRIL also proposes a new feature weighting technique that takes benefit maximization issues into account.…”

Section: Introductionmentioning

confidence: 99%

Modeling interestingness of streaming association rules as a benefit-maximizing classification problem

Aydın

Güvenir

2009

Knowledge-Based Systems

View full text Add to dashboard Cite

a b s t r a c tIn a typical application of association rule learning from market basket data, a set of transactions for a fixed period of time is used as input to rule learning algorithms. For example, the well-known Apriori algorithm can be applied to learn a set of association rules from such a transaction set. However, learning association rules from a set of transactions is not a one time only process. For example, a market manager may perform the association rule learning process once every month over the set of transactions collected through the last month. For this reason, we will consider the problem where transaction sets are input to the system as a stream of packages. The sets of transactions may come in varying sizes and in varying periods. Once a set of transactions arrive, the association rule learning algorithm is executed on the last set of transactions, resulting in new association rules. Therefore, the set of association rules learned will accumulate and increase in number over time, making the mining of interesting ones out of this enlarging set of association rules impractical for human experts. We refer to this sequence of rules as ''association rule set stream" or ''streaming association rules" and the main motivation behind this research is to develop a technique to overcome the interesting rule selection problem. A successful association rule mining system should select and present only the interesting rules to the domain experts. However, definition of interestingness of association rules on a given domain usually differs from one expert to another and also over time for a given expert. This paper proposes a post-processing method to learn a subjective model for the interestingness concept description of the streaming association rules. The uniqueness of the proposed method is its ability to formulate the interestingness issue of association rules as a benefit-maximizing classification problem and obtain a different interestingness model for each user. In this new classification scheme, the determining features are the selective objective interestingness factors related to the interestingness of the association rules, and the target feature is the interestingness label of those rules. The proposed method works incrementally and employs user interactivity at a certain level. It is evaluated on a real market dataset. The results show that the model can successfully select the interesting ones.

show abstract

Evaluating the Correlation Between Objective Rule Interestingness Measures and Real Human Interest

Cited by 32 publications

References 7 publications

Scalable pattern mining with Bayesian networks as background knowledge

Scalable pattern mining with Bayesian networks as background knowledge

Interactive search of rules in medical data using multiobjective evolutionary algorithms

Modeling interestingness of streaming association rules as a benefit-maximizing classification problem

Contact Info

Product

Resources

About