Analysis of tree-based uncertain frequent pattern mining techniques without pattern losses
| Autor: | Kyung-Min Lee, Unil Yun, Gangin Lee |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Uncertain data
Computer science Data stream mining business.industry Test data generation Concept mining 02 engineering and technology computer.software_genre Theoretical Computer Science Variety (cybernetics) Tree structure Hardware and Architecture 020204 information systems 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing The Internet Data mining business K-optimal pattern discovery computer Software Information Systems |
| Zdroj: | The Journal of Supercomputing. 72:4296-4318 |
| ISSN: | 1573-0484 0920-8542 |
| DOI: | 10.1007/s11227-016-1847-z |
| Popis: | Various large-scale data have been generated in a variety of application fields, since the Internet began to be widely used. Accordingly, researchers have developed various data mining methods for pervasive human-centric computing to deal with the data and discover interesting knowledge. Frequent pattern mining is one of the main issues in data mining, which finds meaningful pattern information from databases. In this area, not only precise data but also uncertain data can be generated depending on environments of data generation. Since the concept of uncertain frequent pattern mining was proposed to overcome the limitations of traditional approaches that cannot deal with uncertain data with existential probabilities of items, several relevant methods have been developed. In this paper, we introduce and analyze state-of-the-art methods based on tree structures, and propose a new uncertain frequent pattern mining approach. We also compare algorithm performance and discuss characteristics of them. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|