반응형

목차 및 찾기 기능(Ctrl + F)을 이용해서 용어를 찾아보세요!

다시 참고하고 싶으신 경우 즐겨찾기에 추가해 두시면 편리합니다.

 

 

 

 

 

A

 

absolute value 절대값  
accuracy 정확도  
acquisition function 획득 함수 베이지안 최적화
AdaBoost 에이다부스트 부스팅의 일종
arithmetic mean 산술 평균  
average precision 평균 정밀도  
average precision at k (AP@K) AP@K
K개 평균 정밀도
 
averaging 평균화(하기)  
adversarial validation 적대적 검증  
adaptive overfitting 적응적 오버피팅
적응적 과대적합
 
activation function 활성화 함수  
attention layer 어텐션층  
auto-correlation 자기상관  

 

 

 

B

 

bandit approach 밴딧 접근법  
base learner 기본 모델 의사결정 트리 등과 같이 기본 분류기를 구축하는데 사용된 알고리즘
binary classification 이진 분류  
bivariate analysis 이변량 분석  
binary mask 바이너리 마스크  
binary indicator 이진 지시자  
bounding box 경계 상자  
Bayesian optimization 베이시안 최적화  
bagging 배깅 Bootstrap Aggregation의 약자

통계적 분류와 회귀 분석에서 사용되는 기계 학습 알고리즘의 안정성과 정확도를 향상시키기 위해 고안된 일종의 앙상블 학습법
big data 빅데이터 거대한 규모, 빠른 속도, 높은 다양성을 특징으로 하는데이터
bias 편향  
bootstrap 부트스트랩 샘플링 전략
bootstrapping 부트스트래핑 부트스트랩 방식으로 샘플링 하는 것
Bernoulli Distribution 베르누이 분포  
binomial distribution 이항 분포  
Bayes' theorem 베이즈 정리  
Bayesian Updating 베이즈 갱신  
Bayesian inference 베이즈 추론  

 

 

 

 

C

 

categorical variable 범주형 변수  
cell  
cell frequency 칸 빈도, 칸 도수  
coefficient of determination 결정 계수  
class 클래스, 부류  
classification 분류  
confusion matrix 정오분류표, 혼동 행렬  
conditional probability 조건부 확률 주어진 사건이 일어났다는 가정 하에 다른 한 사건이 일어날 확률
convolution 합성곱  
cross-entropy 교차 엔트로피  
cross-validation 교차 검증
교차 유효성 검사
 
Cohen Kappa score 코헨 카파 점수  
Computer Vision 컴퓨터 비전  
coordinate 좌표  
cost function 비용 함수  
concept drift 개념 표류  
covariate shift 공변량 변화  
calibration function 교정 함수  
clustering 군집화  
cluster analysis 군집 분석  
combinatorial explosion 조합 폭발  
common difference 공차  
credible interval 신용구간  
column vector 열 벡터  

 

 

 

 

D

 

data exploration 데이터 탐색  
data leakage 데이터 누수  
dice coefficient 다이스 계수  
dimensionality reduction 차원 축소  
denominator (분수의) 분모  
decision tree 의사 결정 나무  
Decision-Making
의사 결정  
decision (contour) curve 결정 (윤곽) 곡선  
descriptive statistics 기술 통계학  
dense matrix 밀집 행렬 sparse matrix
희소 행렬

 

 

 

 

 

 

 

E

 

empirical Bayesian approach 경험적 베이지안 방법  
estimation 추정 관측 표변(표본 통계량)에 기초해서 미지량(모집단의 모수)를 추정하는 것
estimate 추정량 모수를 추정하는 통계량
estimator 추정기 관측 표본으로부터 추정량을 계산하는 절차/규칙/알고리즘/장치
estimated value 추정값 모수를 추종하여 나온 특정값
exponential function 지수 함수  
exponential distribution 지수 분포  
estimation 추정  
explanatory variable 설명 변수  
expected value 기댓값  
extremely randomized tree 극단적 랜덤 트리 =extra tree
Extra Trees 엑스트라 트리 =extremely randomized tree
error function 오차함수  

 

 

 

F

 

factorization machine 인수분해 머신  
F1 score F1 점수, F1 스코어  
fraud detection 이상 거래 탐지  
feature 특징  
feature engineering 특징 공학  
federated learning 연합 학습 다수의 로컬 클라이언트와 중앙 서버가 협력해서 테이터가 탈중앙화 된 상황에서 글로벌 모델을 학습하는 기술
focal loss 초점 손실  
frame problem 사고범위 문제  

 

 

G

 

generative function 생성함수  
geometric mean 기하 평균  
grid search 그리드 서치  
genetic algorithm 유전 알고리즘  

 

 

 

 

H

 

harmonic mean 조화 평균  
hyperparameter 하이퍼파라미터  
hyperband optimization 하이퍼밴드 최적화  
holdout test 홀드아웃 테스트 최종 검증을 위해서 훈련 등에 사용되지 않은 데이터로 하는 테스트
high-cardinality feature 높은 카디널리티 특징 카디널리티는 특정 데이터 집합의 유니크한 값의 개수. 예를 들어 성별의 경우 '남'과 '여'가 있어 카디널리티가 2이다

 

 

 

I

 

inter-annotation agreement 어노테이션 일치도  
Intersection over union (IoU) IoU


Intersection => 영역의 공통 부분, 교집합
Union => 영역의 합집합, 합집합
over => 나누다
Instance Segmentation 인스턴스 분할
인스턴스 세그멘테이션
객체 분할
 
localization 위치 식별  
Intersection 교집합  
isotonic regression 등장 회귀  
i.i.d.
(iid,independent and identically distributed)
독립 항등 분포 데이터가 독립적이로 동일한 확률 분포를 가짐
Internet of Things 사물인터넷  
leave-one-out(LOO) LOO
하나 빼기
 
independent variable 독립 변수  
Image Segmentation 영상 분할  
image classification 이미지 분류  

 

 

 

 

J

 

Jaccard index 자카드 지수  
Jaccard coefficient 자카드 계수  
Jaccard similarity 자카드 유사도  

 

 

 

 

K

 

Keras 케라스 라이브러리
Kappa coefficient 카파 계수  
k-nearest neighbors k-최접근 이웃  
k-fold cross-validation k-분할 교차 검증  
kurtosis 첨도  
kernel 커널 두 벡터 x와 y의 내적을 (대부분 매우 고차원의) 특징 공간에서 계산하는 방법

 

 

 

 

 

L

 

label prediction 레이블 예측  
lambda factor 람다 인자  
lasso selection 올가미 선택
라쏘 선택
 
logarithmic transformation 로그 변환  
localization 위치 식별  
linear transformation 선형 변환  
linear combination 선형 결합  
L2 norm L2 노름  
Law of Large Numbers 큰 수의 법칙
대수의 법칙
 
List conprehension 리스트 내포  
loss function 손실함수  
lossless compression 무손실 압축  

 

 

 

M

 

marginal 한계  
Matthews correlation coefficient (MCC) 매튜스 상관 계수  
macro averaging 매크로 평균화  
Majority voting 다수결 투표 (알고리즘)
최다 득표자 찾기 (알고리즘)
평균화 방법
a way to average
mean absolute error(MAE) 평균 절대 오차  
mean squared error(MSE) 평균 오차 제곱  
micro averaging 마이크로 평균화  
minimization 최소화  
multi-class classification 다중 분류, 다중 클래스 분류  
multi-label classification 다중 레이블 분류  
multivariate 다변량  
MPNN (Message Passing Neural Network) 메시지 전달 신경망  
model building 모델 구축  
Mean Average Precision at K(MAP@{K} ) MAP@{K}
K개의 평균 정밀도 평균값
 
manipulated variable 조작 변수  
maximum likelihood method 최대 가능도 방법  
matrix 행렬  
meta-features 메타 특징  
meta-prediction 메타 예측  
meta-predictor 메타 예측 변수  
multivariate distribution 다변량 분포  
missing value 결측값  
missing value imputation 결측값 대체  

 

 

 

N

 

non-parametric model 비모수적 모델 parametric model
모수적 모델
normalize 정규화하다  
normal distribution 정규분포  
number of cases 경우의 수  
numerator (분수의) 분자  
numeric variable 수치형 변수  
Nelder-Mead algorithm 넬더-미드 알고리즘  
non-parametric regression 비모수 회귀  
non-i.i.d.
(non-iid,
non independent and identically distributed)
non-i.i.d
비독립 항등분포
데이터가 비독립적이고 동일하게 분산되지 않음
Nested cross-validation 중첩 교차 검증  
no information rate 무작위성의 비율  
no-information error rate 무작위성의 오류 비율  
normalization 정규화  

 

 

 

O

 

outlier 이상값  
object detection 객체 탐지  
optimizer 최적화기
옵티마이저
가중치, 학습률 등의 파라미터를 최적화하는 함수나 알고리즘
ordinal value 순서 값  
ordinal prediction 순서 예측  
oversampling 오버샘플링  
overfitting 과대적합
과적합
underfitting 과소적합
out-of-sample data 표본 외 샘플 데이터  

 

 

 

P

 

Pasting 페이스팅 샘플링 방식
parametric model 모수적 모델 non-parametric model
비모수적 모델
probability 확률  
principal diagona 주대각선 = Main diagonal  
precision 정밀도  
precision/recall curve 정밀도/재현율 커브  
precision/recall trade-off 정밀도/재현율 트레이드 오프  
PyTorch 파이토치 라이브러리
pixel accuracy 픽셀 정확도  
perceptron 퍼셉트론  
probabilistic approach 확률적 접근 방식  
Probabilistic evaluation 확률적 평가  
Probabilistic estimator 확률적 추정기  
predictor variable 예측 변수  
PGA
(principal component analysis)
주성분분석  
parameter 파라미터  
Poisson distribution 푸아송 푼포  
posterior probability 사후 확률  
priori probability 사전 확률  
pruning 가지치기  

 

 

 

Q

 

Quadratic Weighted Kappa 이차 가중치 카파  
quantile transformation 분위수 변환
데이터 분포도 변환
 

 

 

 

R

 

RAdam
Rectified Adam
RAdam  
reporting function 보고 함수  
root mean squared error(RMSE) 평균 제곱근 오차  
root mean squared log error(RMSLE) 평균 제곱근 로그 오차  
R squared R스퀘어 = 결정 계수  
recurrent layer 순환층  
regression 회귀  
regressor 회귀 변수
회귀자
 
recall 재현률  
random search 랜덤 탐색
랜덤 서치
 
ROC(receiver operating characteristic) ROC, 수신자 조작 특성  
ROC curve ROC 곡선  
receiver operating characteristic curve(ROC curve) 수신자 조작 특성 곡선  
Reinforcement Learning 강화 학습  
regressand 반응 변수 = response variable
response variable 반응 변수 = regressand
ReLU(Rectified Linear Unit) ReLU 발음이 아래의 3가지로 나뉘는데 rel-you가 경험자들 사이에서는 많이 쓰이는 듯 하다.
ray-loo
reh-loo
rel-you
row vector 행 벡터  
random forest 랜덤 포레스트  
Random patch 랜덤 패치 앙상블 샘플링 전략
random seed 랜덤 시드
무작위 시드
임의 시드
 
Random subspace 랜덤 서브스페이스 앙상블 샘플링 전략
reproducibility 재현성  
run-length encoding(RLE) 런 렝스 인코딩
런 렝스 부호화
공간 절약하는 부호화 방식. 같은 값이 연속해서 나타나는 것을 그 개수와 반복되는 값만으로 표현

 

 

S

 

scoring function 점수 함수  
Stochastic Weighted Averaging
(SWA)
확률적 가중치 평균  
standard deviation 표준편차  
standardization 표준화  
segmentation 세그멘테이션, 분할  
Sentiment analysis 감정 분석  
Semantic Segmentation 시멘틱 세그멘테이션, 의미론적 분할  
sum of squared errors(SSE) 오차 제곱합  
sum of squares total(SST) 총 변동  
sensitivity 민감도  
Sparse matrix 희소 행렬 dense matrix
밀집 행렬
Sparsity-awareness 희소성 인식  
split point 분할 지점 GBM 알고리즘 관련
standardization 표준화  
stacked prediction 누적 예측 여러 모델을 훈련 모델을 훈련 데이터를 분할한 데이터로 구축해서 표본 외 데이터 예측을 하고 과대적합을 예방할수 있게 해주는 방식
stability selection 안정선 선택  
symmetric Mean Absolute Percentage Error(sMAPE) 대칭 평균 절대 백분률 오차  
support-vector machine 서포트 벡터 머신  
systematic experimentation 체계적인 실험  
supervised learning 지도 학습  
Successive Halving Algorithm(SHA) 연속 분할 알고리즘  
simple random sampling 단순 무작위 샘플링
단순 무작위 추출법
단순 무작위 표집
 
stratification 계층화  
stratified sampling 계층화 샘플링
층화 샘플링
계층화 추출법
계층화 표집
계층적 샘플링
 
stratified k-fold 계층화 k-분할
층화 k-분할
 
stratification variable 계층화 변수
층화 변수
층별 번수
 
sampling by groups 그룹별 샘플링
그룹별 추출법
그룹별 표집
 
time sampling 시간 샘플링
시간 추출법
시간 표집
 
sweet point 정점 과대적합(overfitting)이 시작되는 지점
subsampling 서브샘플링  
surrogate function 대리 함수 베이지안 최적화 관련
standard error 표준 오차  
scatterplot
scatter plot

산점도 =scatter graph
=scatter chart
=scattergram
=scatter diagram
skewness dhoeh  

 

 

T

 

tolerance 허용 오차
공차(公差)
 
two-way table 이원분류표  
true positive rate(TPR) 참 양성 비율  
Transformers 트랜스포머 라이브러리
transformation 변환  
time sampling 시간 샘플링
시간 추출법
시간 표집
샘플링 전략
time series forecasting 시계열 예측 순차적으로 수행되는 일련의 관찰로 구성된 시계열 정보로 해당 시리즈의 미래 값을 예측
training loss 훈련 손실  
test-based randomization 테스트 기반 무작위화  
theorem 정리  
threshold 문턱값
임계값
 
target encoding 목표 인코딩 =
likelihood encoding
우도 인코딩

=
impact coding
충격 인코딩

=
mean encoding
평균 인코딩

 

 

 

 

U

 

 

Union 교집합  
univariate 단변량  
univariate analysis 단변량 분석 =
undersampling 언더샘플링  
underfitting 과소적합  
unsupervised learning 비지도 학습  

 

 

 

 

V

 

variance 분산  
validation 검증
유효성 검사
 
validation loss 검증 손실 검증 데이터에 대한 손실
Visual Attention 시각주의  
Variational auto-encoder 변이형 오토 인코더  

 

 

 

W

 

Weighted Root Mean Squared Scaled Error (WRMSSE) 가중 눈금 조절된 평균 제곱근 오차  
weighted average (mean) 가중 평균  

 

 

 

 

숫자

 

 

 

632 bootstrap 632 부트스트랩 검증 전략

 

 

 

반응형
  • 네이버 블러그 공유하기
  • 네이버 밴드에 공유하기
  • 페이스북 공유하기
  • 카카오스토리 공유하기