Data Mining Unit 1 : Intro to DM

Data

Raw facts gathered through observation, experience, or experiments

Information

Meaningful data that has been processed or interpreted

Knowledge

Verified, structured information integrated into a process for decision-making

Data Mining (DM)

The process of extracting useful and previously unknown patterns from large datasets

Data Mining Purpose

Discover patterns to support decisions in fields such as business, healthcare, science, and government

Supervised Learning

A data mining approach using labeled data to train models (e.g., classification, regression)

Unsupervised Learning

A data mining approach that explores patterns in unlabeled data (e.g., clustering, association)

Classification

Predicts categorical outcomes like Yes/No or types of diseases

Regression

Predicts continuous outcomes such as temperature or price

Clustering

Groups data based on similarity; used in market segmentation and document analysis

Association Rule Mining

Discovers relationships between items in transactions, like {Milk} → {Bread}

Anomaly Detection

Identifies data that deviates significantly from expected patterns (e.g., fraud detection)

KDD Process (Knowledge Discovery in Databases)

• Selection

• Preprocessing

• Transformation

• Data Mining

• Evaluation

Preprocessing Tasks

• Data cleaning

• Data integration

• Data transformation

• Data reduction

Normalization Methods

• Min-max scaling

• Z-score

• Decimal scaling

• Log transformation