AC-CS: An Immune-Inspired Associative Classification Algorithm

Elsayed, Samir A. Mohamed; Rajasekaran, Sanguthevar; Ammar, Reda A.

doi:10.1007/978-3-642-33757-4_11

Cited by 2 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This additional knowledge proved to enhance the classi¯cation accuracy of the outputted classi¯er if compared with traditional clas-si¯cation approaches according to several experimental studies, i.e. (Yin and Han, 2003;Yu et al, 2011;Wang et al, 2011;Elsayed et al, 2012). Though, the additional knowledge may contain redundant or con°icting rules in which if no appropriate pruning is invoked can cause larger problem called the exponential growth of rules (Thabtah et al, 2011;Thabtah and Hammoud, 2013).…”

Section: Advantages Of Ac Approachmentioning

confidence: 97%

“…Recently, another AIS based on AC called AC-CS was proposed in (Elsayed et al, 2012). This algorithm follows the same track of the previously described AIS-AC and it uses the same strategies in deriving the rules and classifying test data.…”

Section: Clonal Ratementioning

confidence: 99%

“…Empirical evaluations using a limited number of UCI data sets indicated that the AIS proposed algorithms (Elsayed et al, 2012;Do et al, 2009) are highly competitive in accuracy and execution time to the \Predictive Apriori" algorithm which is a simpli¯ed version of CBA that primarily uses Apriori algorithm for extracting the rules without pruning.…”

Section: Clonal Ratementioning

confidence: 99%

See 2 more Smart Citations

Associative Classification Approaches: Review and Comparison

Abdelhamid

Thabtah

2014

J. Info. Know. Mgmt.

View full text Add to dashboard Cite

Associative classi¯cation (AC) is a promising data mining approach that integrates classi¯cation and association rule discovery to build classi¯cation models (classi¯ers). In the last decade, several AC algorithms have been proposed such as Classi¯cation based Association (CBA), Classi¯cation based on Predicted Association Rule (CPAR), Multi-class Classi¯cation using Association Rule (MCAR), Live and Let Live (L 3 ) and others. These algorithms use di®erent procedures for rule learning, rule sorting, rule pruning, classi¯er building and class allocation for test cases. This paper sheds the light and critically compares common AC algorithms with reference to the abovementioned procedures. Moreover, data representation formats in AC mining are discussed along with potential new research directions.

show abstract

Section: Advantages Of Ac Approachmentioning

confidence: 97%

Section: Clonal Ratementioning

confidence: 99%

Section: Clonal Ratementioning

confidence: 99%

See 1 more Smart Citation

Associative Classification Approaches: Review and Comparison

Abdelhamid

Thabtah

2014

J. Info. Know. Mgmt.

View full text Add to dashboard Cite

show abstract

“…Recently, another AIS based on AC called AC-CS was proposed in [5]. This algorithm follows the same track of the previously described AIS-AC and it uses the same strategies in deriving the rules and classifying test data.…”

Section: New Trends and Future Ac Research Problemsmentioning

confidence: 99%

“…Research studies for instance [5] have shown that AC mining has two distinguishing features over other traditional classification approaches. The first one is that it produces very simple knowledge (rules) that can be easily interpreted and manually updated by the end-user.…”

Section: Introductionmentioning

confidence: 99%

Emerging Trends in Associative Classification Data Mining

Abdelhamid¹,

Ayesh²,

Thabtah³

2014

IJEEE

View full text Add to dashboard Cite

Utilising association rule discovery to learn classifiers in data mining is known as Associative Classification (AC). In the last decade, AC algorithms proved to be effective in devising high accurate classification systems from various types of supervised data sets. Yet, there are new emerging trends and that can further enhance the performance of current AC methods or necessitate the development of new methods. This paper sheds the light on four possible new research trends within AC that could enhance the predictive performance of the classifier or their quality in terms of rules. These possible research directions are considered starting research points for other scholars in rule based classification in data mining.

show abstract

AC-CS: An Immune-Inspired Associative Classification Algorithm

Cited by 2 publications

References 24 publications

Associative Classification Approaches: Review and Comparison

Associative Classification Approaches: Review and Comparison

Emerging Trends in Associative Classification Data Mining

Contact Info

Product

Resources

About