Hierarchical object-driven action rules

Hajja, Ayman; Raś, Zbigniew W.; Wieczorkowska, Alicja

doi:10.1007/s10844-013-0291-2

Cited by 20 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…More than a decade there has been a lot of research on diverse methods on generating action rules. So far, action rule mining is based on two approaches: rule-based approach [5] [6] and object-based approach [2] [3] [4].…”

Section: Action Rule Miningmentioning

confidence: 99%

“…Object-based approaches as proposed in [2], [3] and [4], extract Action Rules directly from the information system without additional classification rules extraction like in rule-based approaches. We discuss few of the object-based approaches for discovering Action Rules below:…”

Section: Object-based Approachmentioning

confidence: 99%

“…For example, with an information system of n instances, the maximum support of each action rule can be (n/2) 2 , that is when half of the instances satisfy precondition of the action rule and the other half instances satisfy postcondition of the action rule. When these n instances are divided into m subsystems where each subsystem contains p instances which satisfy the condition p < m < n. Thus, it is obvious that the maximum support (p/2) 2 of the system after division reduces when the value of m (number of subsystem) continues to become higher.…”

Section: Object-based Approachmentioning

confidence: 99%

“…Action Rules can help many applications like in online shopping sites, Action Rules help suggesting how to improve customer satisfaction for a product or in medical field, Action Rules can help suggesting how to improve a patient's health. In [2] [3] [4] [5] [6] authors propose a variety of algorithms to extract Action Rules from a given dataset. Since data is growing at a rapid pace, the proposed action rule mining approaches are stumbling to produce Action Rules in an expected time limit.…”

Section: Introductionmentioning

confidence: 99%

“…This paper reviews some of the methods proposed previously to extract knowledge with classical rule based algorithms such as classification rules [7] [8] [10] [13] and association rules [9] [11] [12] in cloud computing frameworks, using MapReduce [14] and Spark [15], as well as commercial Cloud platforms such as Microsoft Azure [16], Amazon Web Services (AWS) [17], and GoogleCloud [18]. This paper also reviews some of the Action Rule extraction algorithms [2] [6] [5]. Finally, we discuss the possibility to extract Action Rules using selected parallel processing frameworks.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Rule Based Systems in a Distributed Environment: Survey

Bagavathi¹,

Tzacheva²

2017

World Congress on Electrical Engineering and Computer Systems and Science

View full text Add to dashboard Cite

-Over the past years, the internet has become faster, computer storage has become larger and the data from internet users and sensors is piling up to larger amount and is also spread around the globe. This requires more time and space for a single computer to cope with them. Ecosystems like Hadoop helps to store, process and retrieve back the data efficiently in a distributed fashion. Data mining finds substantial improvements over such distributed frameworks to process a large volume of data in a lesser time. Currently, there are many approaches to do data mining tasks such as classification and clustering in a distributed setup using Hadoop MapReduce, Spark, and other Cloud platforms. Actionable pattern mining is a rule based data mining approach for discovering knowledge from information systems in a form of Action Rules. An emphasis of traditional classification rules from a supervised Machine Learning is to predict class label of a data object. Whereas Action Rules produce actionable knowledge in the form of suggestions on how an object can change from one class value to another more desirable class value. This paper gives a brief survey of previous works on association and classification rule mining algorithms in a distributed environment, as well as action rule mining algorithms, and discusses Action Rule Mining in a distributed environment.

show abstract

Section: Action Rule Miningmentioning

confidence: 99%

Section: Object-based Approachmentioning

confidence: 99%

Section: Object-based Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%