DataMining and Warehousing - chapter1
DataMining and Warehousing - chapter1
Overview
Brief description
of Data Mining
Key
processes Prediction: Using
discovered patterns
to make future
Pattern Discovery:
Identifying patterns,
associations, or
in Data predictions. trends in the data.
Mining
Evaluation:
Assessing the
quality and
relevance of the
discovered patterns.
Data
Mining
Architectu
re
Market Basket Analysis
Fraud Detection
Use cases
of Data Customer Segmentation
Mining
Predictive Maintenance
Network Intrusion
Detection
Descriptive data mining: involves summarizing
and describing the characteristics of a data set.
Mining
Prescriptive data mining: involves using data
and models to make recommendations or
suggestions about actions or decisions. This type
of data mining is often used to optimize processes,
allocate resources, or make other decisions that
can help organizations achieve their goals.
Data Mining vs Statistics
• Statistics is a field of mathematics that focuses on data collection,
analysis, interpretation, and presentation using established
theories and mathematical models.
Data Size Works best with small to medium Works well with very large
datasets datasets
Process Starts with a hypothesis, then tests Finds patterns automatically
it using data without prior hypotheses
Techniques Regression, hypothesis testing, Clustering, classification,
sampling neural networks
Data Complexity
Interpretability
Scalability
Ethical Concerns about Data
Mining
Applications of data mining: Healthcare and
Medicine
Risk
Fraud Management Stock Market
Detection and Credit Prediction
Scoring
Applications of data mining: Retail and E-commerce
Predictive Maintenance
Student Performance
Prediction
Adaptive Learning Systems
Exam Cheating Detection
Applications of data mining: Social Media and
Entertainment
Sentiment Analysis
Recommendation Systems
Clustering
Data
mining Regression