11.2.1 Data mining techniques

• What is the primary goal of data mining?

  Discover patterns in data
• Data mining can be used to forecast future trends.
• Data mining is the process of extracting patterns and insights from large datasets
• Match the data mining technique with its purpose:

  Association Rule Learning ↔️ Discovers relationships between variables

  Classification ↔️ Assigns data to predefined categories

  Clustering ↔️ Groups similar data points together

  Anomaly Detection ↔️ Identifies unusual data points
• Steps in Association Rule Learning

  1️⃣ Discover frequent itemsets

  2️⃣ Generate association rules

  3️⃣ Evaluate rule metrics
• Association Rule Learning identifies relationships between variables in a dataset
• What type of data is required for classification?

  Labeled data
• Clustering is sensitive to feature selection.
• What is a common weakness of anomaly detection?

  High false positive rates
• In Association Rule Mining, support measures the frequency of the rule's occurrence
• Confidence in Association Rule Mining measures the probability that the rule is correct.
• What does lift measure in Association Rule Mining?

  The rule's effectiveness
• Classification categorizes data into predefined classes
• Classifying emails as spam or not spam is an example of regression.

  False
• Clustering groups similar data points together based on shared characteristics
• DBSCAN is capable of finding clusters of arbitrary shapes.
• K-Means partitions data into clusters based on minimizing the distance to centroids
• Hierarchical clustering requires specifying the number of clusters beforehand.

  False
• What does DBSCAN use to identify clusters?

  Density
• Data mining discovers patterns, trends, and useful information from datasets
• Trends in data mining refer to recurring sequences or relationships.

  False
• What is the purpose of decision-making in data mining?

  Selecting best action
• Association Rule Learning identifies relationships between variables
• Clustering is sensitive to feature selection.
• Which data mining technique predicts continuous numerical values?

  Regression
• Match the metric in Association Rule Mining with its description:

  Support ↔️ Frequency of the rule's occurrence

  Confidence ↔️ Probability that the rule is correct

  Lift ↔️ Measures the rule's effectiveness
• In market basket analysis, a rule might be "if a customer buys bread, they are likely to buy butter
• What is the primary difference between classification and regression?

  Output type
• Regression outputs discrete categories.

  False
• Give an example of a classification problem.

  Email spam detection
• K-Means clustering partitions data into K clusters based on minimizing distance to centroids
• Hierarchical clustering is computationally intensive for large datasets.
• What is a key weakness of DBSCAN clustering?

  Sensitive to density parameters
• Anomaly detection identifies data points that deviate significantly from the norm
• Machine learning models can be used for anomaly detection.