(1)
函數關系：functional relation
正相關：positive correlation
負相關：negative correlation
相關系數：correlation efficient
一元線性回歸：simple linear regression
多元線性回歸：multiple linear regression
參數：parameter
參數估計：parameter estimation
截距：intercept
斜率：slope
誤差：error
殘差：residual
擬合：fit
最小二乘法：method of least squares
殘差平方和：residual sum of squares (RSS)
創建向量：create a vector
回歸系數：regression coefficient
建立線性模型：create a linear model
內推插值：interpolate
外推歸納：extrapolate
回歸診斷：regression diagnostics
離群值：outlier
多重共線性：multicollinearity
廣義線性回歸模型：generalized linear regression model
啞變量（虛擬變量）：dummy variable
logistic回歸：logistic regression
非線性回歸：nonlinear regression
(2)
頻繁模式挖掘：frequent pattern mining
無偏估計：unbiased estimation
最小二乘法：least square methods (LSM)
矩陣：matrix
系數：coefficient
截距項：intercept
隨機誤差：random error
廣義逆：generalized inverse
有偏估計：biased estimation
嶺回歸：ridge regression
等價模型：equivalence model
懲罰模型：penalty function
估計族：class of estimators
嶺參數：ridge parameter
嶺跡圖：ridge trace
方差擴大因子：variance inflation factor (VIF)
LASSO：least absolute shrinkage and selection operator
最小角回歸：least angle regression (LAR)
(3)
主成分分析：principal component analysis
降維：dimension reduction
特征選擇：feature selection
特征提取：feature extraction
協方差矩陣：covariance matrix
對角化：diagonalization
因子分析：factor analysis
因子載荷矩陣：factor loading matrix
特殊方差矩陣：special variance matrix
極大似然法：maximum likelihood method
正交旋轉：orthorgonal rotation
因子得分：factor score
(4)
分類模型：classification model
線性判別法：linear discriminant analysis
線性分類器：Linear Classifier
距離判別法：distance discriminant method
k-NN algorithm（k-最近鄰算法）：k-Nearest Neighbors algorithm (or k-NN for short)
朴素貝葉斯分類器：Naïve Bayes Classifier
決策樹：decision tree
支持向量機：support vector machines (SVM)
神經網絡：neural network
文本挖掘：text mining
貝葉斯信念網絡：Bayesian belief network
條件概率：conditional probability
先驗概率：prior probability
后驗概率：posterior probability
條件概率表：conditional probability table (CPT)
貝葉斯推理：Bayesian reasoning
(5)
決策樹：decision tree
ID3算法（迭代二叉樹3代）：iterative dichotomiser 3
分類與回歸樹：classification and regression tree(CART)
信息增益：information gain (IG)
分裂屬性：splitting attribute
剪枝：pruning
代價復雜度：cost-complexity
組合算法：combinatorial algorithm
裝袋算法：bagging algorithm
提升算法：boosting algorithm
AdaBoost（Adaptive Boosting：
自適應增強）算法：Adaboost algorithm
隨機森林算法：random forest algorithm
(6)
支持向量機：support vector machines (SVM)
線性可分：linearly-separable
最優分離平面：optimal separating plane
決策邊界：decision boundary
最大邊緣超平面：maximum marginal hyperplane (MMH)
凸優化問題：convex optimization problem
拉格朗日乘子法：lagrange multiplier method
KKT條件：Karush-Kuhn-Tucker conditions
對偶：duality
松弛變量：slack variable
懲罰函數：penalty function
SMO算法（序列最小優化算法）：sequential minimal optimization
高維空間：high dimension space
維度災難：curse of dimensionality
核函數：kernel function
高斯徑向基函數：Gaussian radial basis function
Mercer定理：Mercer's theorem
(7)
人工神經網絡：artificial neural network
神經元，神經細胞：neuron
樹突：dendrite
軸突：axon
細胞體：cell body
突觸：synapsis
單層感知器：single-layer perceptron
線性神經網絡：linear neural network
BP神經網絡：BP Neural Network
輸入節點：input node
輸出節點：output node
權向量：weight vector
偏置因子（偏因）：bias factor
激活函數：activation function
自學習算法：self-learning algorithm
學習率：learning rate
權重：weight
偏移：bias
偏置值：bias value
平均絕對誤差：mean absolute error
均方誤差：mean square error
誤差平方和：square error sum
拓撲：topology
前饋型網絡：feedforward networks
反饋型網絡：feedback network
學習規則：learning rule
線性神經網絡：linear neural network
梯度下降法：gradient descent algorithm
(8)
深度學習：deep learning
有監督學習：supervised learning
無監督學習：unsupervised learning
半監督學習：semi-supervised learning
BP神經網絡：BP Neural Network
誤差反向傳播算法：error back-propagation algorithm
多層前饋神經網絡：multilayer feed-forward neural network
最速下降法：method of steepest descent
圖像壓縮：image compression
Hopfield神經網絡：Hopfield neural network
光學字符識別：Optical Character Recognition (OCR)
PCA（Principal Component Analysis）神經網絡：PCA Neural Network
神經網絡芯片：neural network chip
通用逼近器：universal approximator
徑向基函數神經網絡：radial basis function neural network (RBFNN)
正則化RBF神經網絡：normalized RBF neural network (NRBFNN)）
廣義RBF神經網絡：Generalized RBF neural network
概率神經網絡：probabilistic neural network
貝葉斯信念網絡：Bayesian belief network
(9)
貝葉斯信念網絡：Bayesian belief network
梯度計算：gradient computation
權重更新：weight updating
聚類：clustering
孤立點（離群值、異常值）：outlier
距離：distance
絕對值距離：absolute value distance
歐氏（歐幾里得）距離：Euclidean distance
閔可夫斯基：Minkowski distance
切比雪夫距離：Chebyshev distance
馬氏（馬哈拉諾比斯）距離：Mahalanobis distance
蘭氏距離：Lance and Williams distance / Canberra Distance
離散變量：discrete variable
數據中心化與標准化：data centralization and standardization
極差（全距）：range
相似系數：similarity coefficient
夾角余弦：included angle cosine
相關系數：correlation coefficient
凝聚聚類算法：aggregate clustering algorithm
層次聚類法：hierarchical clustering
動態聚類：dynamic clustering
K-means 算法：K-means algorithm
K中心聚類法：k-medoids clustering
圍繞中心點的划分算法：Partitioning Around Medoids (PAM)
CLARA 算法：CLARA algorithm (Clustering for LARge Applications)
基於密度的聚類：density-based clustering
DBSCAN 聚類算法：DBSCAN (A Density-Based Spatial Clustering of Application) clustering algorithm
基於網格的聚類算法：Grid-Based Clustering Algorithm
CLIQUE算法：CLIQUE (CLustering In QUEst) Algorithm
稠密單元：dense unit
簇的裝配：assembling of cluster
最小覆蓋：minimum covering
貪心算法：greedy algorithm
(10)
霍普金斯統計量：Hopkins statistic
簇數制定：determining the number of clusters
肘方法：elbow method
偽F統計量：pseudo f-statistics (PSF)
偽T平方統計量：Pseudo-T2 Statistic (PST2)：
B-cubed聚類評分：B-cubed cluster scoring
輪廓系數：silhouette coefficient
模糊聚類：fuzzy clustering
誤差平方和：square sum of error (SSE)
基於概率模型的聚類：probabilistic-model-based clustering
概率聚類：probabilistic clustering
最大似然估計：maximum likelihood estimation
最大期望算法：expectation-maximization (EM) algorithm
混合模型：hybrid model
離群值檢測：outlier detection
基於直方圖的離群值檢測：histogram-based outlier detection
基於鄰域的離群值檢測：neighborhood-based outlier detection
基於網格的離群值檢測：grid-based outlier detection
基於聚類的離群值檢測：clustering-based outlier detection