Software techniques of data mining and experimental design for high-throughput screening of catalysts and materials촉매 및 재료 고속처리탐색을 위한 데이터마이닝 및 실험계획법 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 414
  • Download : 0
During the last several years, the development of combinatorial technology has enabled synthesis and characterization of a huge amount of chemical compounds in a short time. The rapid development of high-throughput experimental tools requires the parallel development of software techniques of data mining and experimental design for the comprehension and specific modeling of experimental data and the optimization of specific performance of catalysts and materials, respectively. A self-organizing algorithm of multilayer perceptrons to establish the model without human intervention was proposed to effectively predict characteristics of catalysts and materials. And some best practice recommendations are provided to apply Evolutionary Strategy for efficient experimental design. These approaches can accelerate high-throughput screening of catalysts and materials and improve the efficiency of discovering catalysts and materials by saving experimentation cost and manpower.
Advisors
Park, Sun-Wonresearcher박선원researcherWoo, Seong-Ihlresearcher우성일researcher
Description
한국과학기술원 : 생명화학공학과,
Publisher
한국과학기술원
Issue Date
2008
Identifier
303580/325007  / 020045803
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 생명화학공학과, 2008. 8., [ 117 p. ]

Keywords

High-throughput Screening; Design of experiments; Genetic Algorithms; Data Mining; Artificial neural networks; 고속처리탐색; 실험계획법; 유전알고리즘; 데이타마이닝; 인공신경망; High-throughput Screening; Design of experiments; Genetic Algorithms; Data Mining; Artificial neural networks; 고속처리탐색; 실험계획법; 유전알고리즘; 데이타마이닝; 인공신경망

URI
http://hdl.handle.net/10203/29034
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=303580&flag=dissertation
Appears in Collection
CBE-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0