Rough Set Theory and Its Applications to Data Analysis

Pawlak, Zdzisław

doi:10.1080/019697298125470

Cited by 816 publications

(401 citation statements)

References 4 publications

Supporting

Mentioning

396

Contrasting

Unclassified

Order By: Relevance

“…After checking all the attributes, elements of the calculated subset are added to reducts and then the main condition (4) is checked. When it is false-algorithm exits and results on output with calculated set of RED(C), being a minimum set of attributes for the decision table DT [11]. The final stage is to determine the classification decision rules.…”

Section: Investigation Resultsmentioning

confidence: 99%

“…To obtain the results of the content search closer to the user expectations one can use methods of an artificial intelligence: the method of latent semantic indexing [9], rough sets [10][11][12] and association rules learning [13,14]. These methods have one thing in common-to define the substantive content of the documents use patterns acquired through exploration of training set of documents.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Formalization of Technological Knowledge in the Field of Metallurgy using Document Classification Tools Supported with Semantic Techniques

Regulski

2017

Archives of Metallurgy and Materials

View full text Add to dashboard Cite

The process of knowledge formalization is an essential part of decision support systems development. Creating a technological knowledge base in the field of metallurgy encountered problems in acquisition and codifying reusable computer artifacts based on text documents. The aim of the work was to adapt the algorithms for classification of documents and to develop a method of semantic integration of a created repository. Author used artificial intelligence tools: latent semantic indexing, rough sets, association rules learning and ontologies as a tool for integration. The developed methodology allowed for the creation of semantic knowledge base on the basis of documents in natural language in the field of metallurgy.

show abstract

Section: Investigation Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Formalization of Technological Knowledge in the Field of Metallurgy using Document Classification Tools Supported with Semantic Techniques

Regulski

2017

Archives of Metallurgy and Materials

View full text Add to dashboard Cite

show abstract

“…It is not enough to accurately identify which property is most important only by attribute dependency. So we also need to look at dependencies between it and other properties, that is, the attribute set dependence must be considered [4,5].…”

Section: Attribute Set Dependencementioning

confidence: 99%

Bayesian Spam Filtering Mechanism Based on Decision Tree of Attribute Set Dependence in the MapReduce Framework

Guo¹,

Zhou²,

He³

et al. 2014

TOCSJ

View full text Add to dashboard Cite

Bayesian spam filtering is a classification method based on the theory of probability and statistics, and the Bayesian spam filtering based on Mapreduce can solve the defect of the traditional Bayesian spam filtering that consumes large amounts of system resources and network resources when the mail set is pre-training. It needs to classify mails manually in the pre-training phase of mail set, which consumes a lot of human and financial resources and affects the efficiency of the system. Bayesian spam filtering mechanism based on decision tree of the attribute sets dependence in the MapReduce framework which is presented in this paper. And the decision tree of attribute sets dependence is used in the training stage of the mail set, which improves execution efficiency of the system by lowering the time complexity.

show abstract

“…However, this trend of enormity on both size and dimensionality also poses severe challenges to reduct generation algorithms too. Rough Set Theory (RST) [1,2], a new mathematical approach to imperfect knowledge, is popularly employed to evaluate significance of attributes and helps to find the reduct.…”

Section: Introductionmentioning

confidence: 99%