Real-world data analysis with physical model approach물리적 모델을 통한 실제 세상에서의 데이터 분석 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 437
  • Download : 0
DC FieldValueLanguage
dc.contributor.advisorJeong, Hawoong-
dc.contributor.advisor정하웅-
dc.contributor.authorSoh, Hyungjoon-
dc.date.accessioned2019-08-22T02:41:38Z-
dc.date.available2019-08-22T02:41:38Z-
dc.date.issued2018-
dc.identifier.urihttp://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=734285&flag=dissertationen_US
dc.identifier.urihttp://hdl.handle.net/10203/264666-
dc.description학위논문(박사) - 한국과학기술원 : 물리학과, 2018.2,[xi, 67 p. :]-
dc.description.abstractReal-world data often behaves different as its collection of partial information, which is called emergence. Due to long-range correlation, the local fluctuation drives whole system and produces distinct features. Such systems are referred as Complex Systems, where long-range correlation is shown by several aspects in temporal, spatial and internal degrees of the randomness. This thesis discusses a few systems which exhibits complex behavior. Chapter 2 discusses emergence of long-term memory in popularity dynamics, using ranking chart of western popular music, the Billboard Hot 100 chart. Using simple spreading model with long-term memory kernel, we found the rise and fall of popularity is influenced under long-term memory. Also, the memory effect becomes greater in recent, which means different type of popularity dynamics is emerging. Chapter 3 discusses quantitative evidence of industry convergence. Interaction between companies is quantified by co-mention frequency, and the convergence can be represented by weakening community structure between community of companies. As a result, mesoscopic scale dynamics of community composition showed industry convergence. Chapter 4 discusses totally asymmetric simple exclusion process (TASEP) on one-dimensional system with a defect. The discrepancy between numerical and analytical solution have different opinion on localization of a defect. Scaling analysis of the finite-size effect of a defect shows there is a crossover from localized phase to bulk phase. Chapter 5 discusses spreading impact of local defect in one-dimensional body-centered solid-on-solid (BCSOS) surface growth. The universal class of growing surface is shown by fluctuation structure, known as Kardar-Parisi-Zhang (KPZ) class. Numerical simulation shows that local defect breaks surfaces' fluctuation universality class with the impact spreading. Chapter 6 discusses generalization of TASEP dynamics which can change localization of the defect. Using zero-range process (ZRP) type attraction/repulsion to the particles, condensation due to attraction and jamming caused by local defect competes and produces rich phase structure. Using extension of defect mean-field, we obtained marginal critical defect strength, as well as phase diagram. Also, condensation of the vacancies affected by local defect is discussed.-
dc.languageeng-
dc.publisher한국과학기술원-
dc.subjectcomplex systems▼areal-world data▼astatistical model▼anonequilibrium systems▼along-range order-
dc.subject복잡계▼a데이터 분석▼a통계 모델▼a비평형 모형▼a장범위 규칙-
dc.titleReal-world data analysis with physical model approach-
dc.title.alternative물리적 모델을 통한 실제 세상에서의 데이터 분석 연구-
dc.typeThesis(Ph.D)-
dc.identifier.CNRN325007-
dc.description.department한국과학기술원 :물리학과,-
dc.contributor.alternativeauthor소형준-
Appears in Collection
PH-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