1.

K-Means Clustering

 Model Name: K-Means Clustering

 Required Package(s): cluster

 Function and Arguments:

kmeans(data, centers, nstart)

o data: Dataset (e.g., iris_1)

o centers: Number of clusters (k) (e.g., 3)

o nstart: Number of random initializations (e.g., 20)

 Evaluation Metrics/Arguments:

o Cluster Assignments: kmeans.re$cluster

o Cluster Centers: kmeans.re$centers

o Visualization: plot() for visualizing the clusters

 Model Name: Naive Bayes Classifier

 Required Package(s): e1071 or caret

 Function and Arguments:

naiveBayes(formula, data, laplace)

o formula: Target variable and predictors (e.g., Species ~ .)

o data: Dataset (e.g., train_data)

o laplace: Additive smoothing (e.g., 1)

 Evaluation Metrics/Arguments:

o Predictions: predict() for class predictions

o Confusion Matrix: table() to compare predicted vs actual values

 Model Name: Decision Tree

 Required Package(s): rpart

 Function and Arguments:

rpart(formula, data, method)

o formula: Target variable and predictors (e.g., Species ~ .)

o data: Dataset (e.g., train_data)

o method: Type of model ("class" for classification or "anova" for regression)

 Evaluation Metrics/Arguments:

o Predictions: predict() for class predictions

o Confusion Matrix: table() to compare predicted vs actual values

o Model Visualization: rpart.plot() to plot the decision tree

 Model Name: Linear Regression

 Required Package(s): stats

 Function and Arguments:

lm(formula, data)

o formula: Target variable and predictors (e.g., Sepal.Length ~ Sepal.Width + Petal.Length)

o data: Dataset (e.g., train_data)

 Evaluation Metrics/Arguments:

o Model Summary: summary() to check coefficients, R-squared, and p-values

o Predictions: predict() for predicted values

o Residuals: residuals() to examine the residuals

 Model Name: Logistic Regression

 Required Package(s): stats

 Function and Arguments:

glm(formula, data, family)

o formula: Target variable and predictors (e.g., Species ~ Sepal.Length + Sepal.Width)

o data: Dataset (e.g., train_data)

o family: binomial for logistic regression

 Evaluation Metrics/Arguments:

o Predictions: predict() for class probabilities or outcomes

o Confusion Matrix: table() for comparing predicted vs actual values

o Model Summary: summary() to inspect coefficients and significance levels

 Model Name: Support Vector Machines

 Required Package(s): e1071

 Function and Arguments:

svm(formula, data, kernel)

o formula: Target variable and predictors (e.g., Species ~ Sepal.Length + Sepal.Width)

o data: Dataset (e.g., train_data)

o kernel: Kernel type ("linear", "radial", etc.)

 Evaluation Metrics/Arguments:

o Predictions: predict() for class predictions

o Confusion Matrix: table() to compare predicted vs actual values

 Model Name: K-Nearest Neighbors

 Required Package(s): class

 Function and Arguments:

knn(train, test, cl, k)

o train: Training data (e.g., iris_train)

o test: Testing data (e.g., iris_test)

o cl: Class labels (e.g., train_data$Species)

o k: Number of neighbors (e.g., 3)

 Evaluation Metrics/Arguments:

o Predictions: predict() for class predictions

o Confusion Matrix: table() to compare predicted vs actual values

 Model Name: Random Forest

 Required Package(s): randomForest

 Function and Arguments:

randomForest(formula, data, ntree)

o formula: Target variable and predictors (e.g., Species ~ .)

o data: Dataset (e.g., train_data)

o ntree: Number of trees (e.g., 500)

 Evaluation Metrics/Arguments:

o Predictions: predict() for class predictions

o Confusion Matrix: table() to compare predicted vs actual values

o Variable Importance: randomForest::importance() to see feature importance

 Model Name: K-Means Clustering

 Required Package(s): cluster

 Function and Arguments:

kmeans(data, centers, iter.max, nstart)

o data: Dataset (e.g., iris_1)

o centers: Number of clusters (k)

o iter.max: Maximum number of iterations (e.g., 100)

o nstart: Number of random initializations (e.g., 20)

 Evaluation Metrics/Arguments:

o Cluster Assignments: kmeans.re$cluster

o Cluster Centers: kmeans.re$centers

 Model Name: Hierarchical Clustering

 Required Package(s): stats

 Function and Arguments:

hclust(d, method)

o d: Distance matrix (e.g., dist(data))

o method: Linkage method ("complete", "single", "average")

 Evaluation Metrics/Arguments:

o Dendrogram: plot() to visualize the hierarchical tree

o Cluster Assignments: cutree() to cut the tree and assign clusters

 Model Name: Association Rules (Apriori)

 Required Package(s): arules

 Function and Arguments:

apriori(data, parameter)

o data: Transaction data (e.g., transactions)

o parameter: Minimum support and confidence (e.g., support = 0.1, confidence = 0.8)

 Evaluation Metrics/Arguments:

o Rules: inspect() to view the generated association rules

o Support: The frequency of itemset occurrence

o Confidence: The likelihood that a rule holds true

 Definition: Involves two or more independent variables (predictors) and one dependent variable
(target).

 Model Name: Multiple Linear Regression

 Required Package(s): stats

 Function and Arguments:

lm(formula, data)

o formula: Target variable and multiple predictors (e.g., target ~ predictor1 + predictor2 +
predictor3)

o data: Dataset (e.g., train_data)

 Example:

lm(Sepal.Length ~ Sepal.Width + Petal.Length, data = iris_train)

 Evaluation Metrics/Arguments:

o Model Summary: summary() for coefficients, R-squared, and p-values

o Predictions: predict() for predicted values

o Residuals: residuals() to analyze the residuals for checking assumptions

o Adjusted R-squared: To assess how well the model fits

 Definition: A type of linear regression where the relationship between the independent and
dependent variable is modeled as an nth degree polynomial.

 Model Name: Polynomial Regression

 Required Package(s): stats

 Function and Arguments:

lm(formula, data)

o formula: Polynomial form (e.g., target ~ poly(predictor, degree = 2))

o data: Dataset (e.g., train_data)

 Example:

lm(Sepal.Length ~ poly(Sepal.Width, 2), data = iris_train)

 Evaluation Metrics/Arguments:

o Model Summary: summary() to inspect the fit

o Predictions: predict() for predicted values

o Residuals: residuals() to check for overfitting