일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
7 | 8 | 9 | 10 | 11 | 12 | 13 |
14 | 15 | 16 | 17 | 18 | 19 | 20 |
21 | 22 | 23 | 24 | 25 | 26 | 27 |
28 | 29 | 30 | 31 |
- preprocessing
- X윈도우
- Unity
- systemd
- 셀레니움
- Polymolphism
- Inheritance
- selenium
- BFS
- dbms
- X.org
- 리눅스
- 자바
- 리눅스 기초
- Java
- Operator
- Class
- literal
- Mac
- Physical Scheme
- Entity Set
- 리눅스 마스터 1급
- zsh
- 백준
- Entity
- python
- Binary Search
- Reference Type
- OOP
- External Scheme
- Today
- Total
목록2024/07 (37)
Byeol Lo
다수의 기업에서는 부서 간에 데이터를 공유하는 것이 필수이다. 이 데이터는 IS(Information System)를 통해서 전부 얻게 되는데, 여기서 기존 단위 업무 위주의 IS 개발 시에는 그때그때 필요에 따라 IS들간의 point-to-point 연결이 필요할 것이고, 유지보수성이 극도로 저하되고, 데이터 통합과 표준화를 불가능하게 한다. 이때 필요한 것이 EAI인데, EAI는 기업의 IS 들의 데이터를 동기화시키는 소프트웨어 및 IS 아키텍처 프레임워크이다. 이질적인 정보들을 서로 같게 해서 데이터 엔티티들에 대한 폭넓고 통합적인 뷰를 제공한다. 또한 비즈니스 프로세스를 자동화하고 실시간으로 통합 연계할 수 있다. EAI를 디자인하는 방식은 다양한데, point-to-point 방식으로 정보시스템..
데이터베이스는 하나의 시점의 데이터가 아니라 계속 변화되는 데이터다. 따라서 데이터 내의 변경이 있을 때는 다른 사람들이 데이터베이스 내부의 데이터가 변경됐는지 모를 것이다. CDC는 이를 알려주기 위해 필요한 후속 처리(데이터 전송/공유) 를 자동화하는 기술 또는 설계 기법이자 구조다. 또한, CDC는 실시간 또는 근접 실시간 데이터 통합을 기반으로 하는 데이터 웨어하우스 및 기타 데이터 저장소 구축에 폭 넓게 활용된다. 이때 CDC는 물리 계층부터 어플리케이션 계층까지 다양한 기술을 통해 구현될 수 있다.Time Stamp on Rows: 마지막 변경 시점을 기록하기 위한 time stamp 필드를 두고 최근의 타임스탬프 값을 갖는 레코드로 변경Version Numbers on Rows: 버전을 기록..
ETL은 데이터 이동과 변환 절차와 관련된 업계 표준 용어임. ETL은 데이터 웨어 하우스(DW, Data Warehouse), 운영 데이터 스토어(ODS, Operational Data Store), 데이터 마트(DM, Data Mart)에 대한 데이터 적재 작업의 핵심 구성 요소임. 데이터 통합(Data Integration), 데이터 이동(Data Migration), 마스터 데이터 관리(MDM, Master Data Management) 에 폭넓게 활용됨. 먼저 관련 시스템(DW, ODS, DM) 이 무엇인지 간단하게만 보자. DW, Data Warehouse 여러 운영 시스템과 외부 데이터 소스에서 수집된 데이터를 통합하고, 분석가능한 형태로 저장하는 중앙 저장소임. 보통 BI 를 지원하고 있어..
모든 문자는 정수입니다. Common Divisorc is common divisor of a and b $\equiv c | a \land c | b$ GCD, Greatest Common Divisor$\gcd(a,b) \\ = \max(\{c\; |\; c|a \} \bigcap \{c\; |\; c|b \}) \\ = \max \{ c\; |\; c|a \land c|b \}$ Common Multiplec is common multiple of a and b $\equiv a | c \;\land\; b | c$ LCM, Least Common Multiple$\operatorname{lcm}(a, b) = \min\{ c \;|\; \forall a, b \in \mathbb{Z}, \; c ..
이 증명은 다른 증명을 위해서 많이 쓰이는 증명이기에 짚고 넘어간다. Well-Ordering Principle(정렬 정리)$ \forall S ( \subsetneq ℕ) \neq ∅ , ∃ min S $𝑝𝑓) S가 최소원을 갖지 않는다고 하자. P(n) ≔ ∀ S ⊂ ℕ, n ∈ S ⇒ ∃ min Sn=1 일때 1을 원소로 가지는 S 에 대해 S는 ℕ 의 부분집합이므로 1보다 작은 원소를 가지지 않는다. 따라서 S는 최소원 1을 가지기 때문에 S가 1을 원소로 가질 때 S는 최소원을 가져야 할 것이다.n=1, 2, 3, ..., k 일때 k를 원소로 가지는 S 에 대해 min S 가 존재함을 가정하고,n=k+1 인 경우에서 S의 최소원이 존재함을 보이자. S가 1, ..., k을 가진다면 P(1)..
빅데이터 분석과 전략 인사이트경영자 측에서의 문제점빅데이터의 회의론으로 투자효과를 거두지 못했던 부정적 학습효과로 인해 빅데이터 분석에 대한 투자를 의심의 눈초리로 바라보는 것데이터 분석 기반 경영 문화의 부재기존 관행을 따를 뿐 중요한 시도를 하지 않음정확성이나 공정한 분석을 필요로 하지 않을 때, 오히려 정반대로 직관적 결정을 귀한 재능으로 칭송할 때아이디어 자체보다는 아이디어를 낸 사람이 누구인지에 관심을 두는 경향 빅데이터에서 빅이라고 말할 수 있는 것은 양이 아니라 다양성과 관련이 있으며, 분석적 방법과 성과에 대한 이해부족이 위와 같은 결과를 초래한다. 전략 인사이트 도출을 위한 필요 역량데이터 사이언스: IT, Analytics, Business Analysis 분야에서 일을 하며, 데이터 ..
우선 범주를 정확히 나누어서 전체적으로 숲을 보는게 중요하고, 용어들을 간단하게만 살펴보자. 자세히 다루는 것은 나중에 해도 된다. AI는 지능적인 행동을 시뮬레이션하는 시스템과도 같다. machine learning은 AI의 부분집합으로 볼 수 있다. Supervised Learning 보통 수식으로 이루어져 있고, 어떤 입력에 대해 출력으로 나오는 것을 함수라고 배웠을 것이다. 만약 특정 입력과 출력이 수로 나타내어질 때, 함수 내의 어떤 수식을 통해 결과값을 얻을 수 있었다. 삼각형의 공식 또한 똑같다. 밑변과 높이를 넣으면 넓이가 나오게 된다. 이를 실생활에서 더 나아가 통계 지식을 넣어서 다양한 문맥에 대해, 문제에 대해 함수를 적용시키고 이런 함수를 model 이라고 하게 된다.입력과 출력을 ..
빅데이터의 정의데이터 자체의 특성 변화 관점: 3V(데이터의 양에서 Volume, 유형과 소스 출처에서의 Variety, 수집과 처리에서의 Velocity)를 가지는 데이터처리, 분석, 기술적 변화를 위한 정의: 새로운 데이터를 처리, 저장, 분석하는 기술 및 아키텍처 혹은 클라우드 컴퓨팅보다 넓은 정의: 대용량 데이터를 활용해 작은 용량에서는 얻을 수 없었던 새로운 통찰, 가치를 추출하는 일 출현 배경 - 패러다임의 전환(paradigm shift)산업계의 변화 - 양질 전환 법칙으로 설명할 수 있음. 정보가 지속적으로 축적되면서 가치 창출이 가능할 만큼 충분한 규모에 도달학계의 변화 - 빅데이터를 활용한 학문 분야가 늘어나면서 필요한 기술 아키텍처 및 통계 도구들이 지속적으로 발전기술발전의 변화아날로..
Problem 첫줄에 정수 $\overline{a_{1}a_{2}] \dots a_{n}}$ 가 있고, 둘째줄에 정수 $\overline{p_{1}p_{2}] \dots p_{m}}$ 이 있다. 다음 조건을 만족하는 맨처음 나오는 소수열 암호 하나를 찾고 출력해라. (단, 각 $ ∀i, j ∈ ℕ, a_{i}, p_{j} $ 는 한자리 자연수)$p_{i} is prime number ⇒ a_{i} is prime number$(단, p_i 가 소수가 아니면 항상 참이다)Constraints1 ≤ m ≤ n ≤ 100,000,0004초Inputs928371003737293720273937912027396210372848624973753717531234567123456123456770000000000007O..
1.4.1 Overview of Delay in Packet-Switched Network 패킷 스위치로 된 네트워크에서는 패킷들은 일련의 라우터들을 거쳐서 호스트로 간다는 것을 봤다. 이때 발생할 수 있는 전체적인 딜레이는 $$(nodal processing delay) + (queueing delay) + (transmission delay) + (propagation delay)$$가 있다. 이 딜레이를 최소화 한다는 것은 네트워크 내의 전송을 최적화한다는 것과 같다. Types of Delay delay 가 어느 부분에서 발생하는지 보자. 위의 그림에서는 A라우터가 B라우터로 가기 위한 outbound link를 지니고 있기 때문에 end system에서의 B로 패킷을 보냈을 때, 패킷이 sour..