Interestingness of Association Rules Using Symmetrical Tau and Logistic Regression

Shaharanee, Izwan Nizal Mohd; Hadzic, Fedja; Dillon, Tharam S.

doi:10.1007/978-3-642-10439-8_43

Cited by 10 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the patterns generated need to be evaluated in order to arrive at significant and useful patterns. A unification framework for evaluating the interestingness of frequent itemsets obtained by the Apriori algorithm was previously developed and reported in Shaharanee, Dillon and Hadzic (2009) and Shaharanee, Hadzic and Dillon (2011). It was found that the rules generated from the Apriori algorithm were large and contaminated with useless patterns.…”

Section: Resultsmentioning

confidence: 99%

“…Both AR values for the testing dataset in Apriori (S,C) and Apriori (S) increased while the CR of the rules was still preserved at 100%. As an extension of our previous work in Shaharanee, Dillon and Hadzic (2009), another method of analysis to discard contradictive rules (Zhang and Zhang, 2001) was included. Contradictive rules exist in Apriori (S) because they are constrained by only a minimum support threshold, because at the set confidence threshold of 60% in Apriori (S,C) , they do not exist.…”

Section: Resultsmentioning

confidence: 99%

“…Although many interestingness and constraint-based measures have been successfully utilized in previous works, there is still a need to understand the roles these parameters play and the way they should be utilized. Thus, in the previous works, the problem addressed by Shaharanee, Hadzic and Dillon (2009) and Shaharanee, Hadzic and Dillon (2011) was developing systematic ways to verify the usefulness of rules obtained from association rule mining. A unified framework, that combines several techniques to assess the quality and remove any redundant and unnecessary rules, has been proposed.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, while confidence measures are often used to reduce the rule set size to only those reflecting highly confident association, no study has been performed on the implication of using different confidence values and the differences of applying this constraint at different stages of the rule verification process. This paper extends the previously developed framework by Shaharanee, Hadzic and Dillon (2009) and Shaharanee Hadzic and Dillon (2011) which seeks to evaluate the impact on classification accuracy, generalization power, and rule coverage rate, when rules are generated using frequent itemset mining algorithms, as well as when different confidence measures are used and applied at different stages of the verification process.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Untitled

2014

APJITM

View full text Add to dashboard Cite

Deriving useful and interesting rules from a data mining system is an essential and important task. Problems such as the discovery of random and coincidental patterns or patterns with no significant values, and the generation of a large volume of rules from a database commonly occur. Works on sustaining the interestingness of rules generated by data mining algorithms are actively and constantly being examined and developed. In this paper, a systematic way to evaluate the association rules discovered from frequent itemset mining algorithms, combining common data mining and statistical interestingness measures, and outline an appropriated sequence of usage is presented. The experiments are performed using a number of real-world datasets that represent diverse characteristics of data/items, and detailed evaluation of rule sets is provided. Empirical results show that with a proper combination of data mining and statistical analysis, the framework is capable of eliminating a large number of non-significant, redundant and contradictive rules while preserving relatively valuable high accuracy and coverage rules when used in the classification problem. Moreover, the results reveal the important characteristics of mining frequent itemsets, and the impact of confidence measure for the classification task

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Untitled

2014

APJITM

View full text Add to dashboard Cite

show abstract

“…To verify the usefulness of the key characteristics obtained from the association rules, this article used the logistic regression method to conduct hypothesis testing and estimate the probability that a particular outcome would occur (Shaharanee, Hadzic, & Dillon, 2009).…”

Section: Methodsmentioning

confidence: 99%

Analyzing the association between emotions and socioeconomic characteristics of census tracts via user‐generated content

Yu¹,

Fei

Li³

2021

Transactions in GIS

View full text Add to dashboard Cite

Environmental and socioeconomic characteristics of places continuously influence human emotions. To understand the quality of human life, human emotions must be effectively evaluated and their connections with socioeconomic characteristics must be determined. Social computing tools can extract emotional information from massive user‐generated content (UGC); thus, we proposed an approach that correlates residents' emotions with the socioeconomic characteristics of place based on census tracts from New York State. First, demographic and economic characteristic data from census tracts were obtained and clustered; then, an emotion recognition technique extracted human emotions quantitatively from the UGC. Applying the Apriori algorithm and logistic regression analysis showed that the socioeconomic characteristics that may significantly influence residents’ emotions include family situation, race, and economic status. Local policymakers can use the proposed approach to compare the quality of human life in different regions over a large scale via UGC and identify socioeconomic characteristics to formulate policies to improve residents’ well‐being and welfare.

show abstract

A Framework for Interestingness Measures for Association Rules with Discrete and Continuous Attributes Based on Statistical Validity

Shaharanee

Jamil

2015

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Assessing rules with interestingness measures is the pillar of successful application of association rules discovery. However, association rules discovered are large in number, some of which are not considered as interesting or significant for the application at hand. In this paper, we present a systematic approach to ascertain the discovered rules, and provide a precise statistical approach supporting this framework. Furthermore, considering that many interestingness measures exist, we propose and compare two established approaches in selecting relevant attributes for the rules prior to rule generation. The proposed strategy combines data mining and statistical measurement techniques, including redundancy analysis, sampling and multivariate statistical analysis, to discard the non-significant rules. In addition to that, we consider real world datasets which are characterized by the uniform and non-uniform data/items distribution with mixture of measurement level throughout the data/items. The proposed unified framework is applied on these datasets to demonstrate its effectiveness in discarding many of the redundant or non-significant rules, while still preserving the high accuracy of the rule set as a whole.

show abstract

Interestingness of Association Rules Using Symmetrical Tau and Logistic Regression

Cited by 10 publications

References 16 publications

Untitled

Untitled

Analyzing the association between emotions and socioeconomic characteristics of census tracts via user‐generated content

A Framework for Interestingness Measures for Association Rules with Discrete and Continuous Attributes Based on Statistical Validity

Contact Info

Product

Resources

About