Data Mining and Exploration Quiz - CSC213

Course Information

• Course: Data Mining and Exploration (CSC213)
• Credits: 3
• Instructor: Dr. Samia M.Abd-Alhalem
• Email: [email protected]
• Prerequisites: Database Management Systems (CSC125)

Course Description

• Introduction to data mining and hands-on experience with all phases of the data mining process using real data and modern tools.
• Topics include:
  • Data formats and cleaning
  • Prediction using supervised and unsupervised learning using Python and other tools
  • Sound evaluation methods
  • Data/knowledge visualization

Data Mining Functions

• Classification

  • Construct predictive models based on training examples
  • Describe and distinguish classes or concepts for future predictions
  • Examples: Classifying countries based on climate or classifying cars based on gas mileage
  • Typical methods: Decision trees, naïve Bayesian classification, support vector machines, neural networks, rule-based classification, pattern-based classification, logistic regression
  • Typical applications: Credit card fraud detection, direct marketing, classifying stars, diseases, web-pages

• Association and Correlation Analysis

  • Identify frequently purchased items together
  • Understand association, correlation, and causality
  • Examples: "Diaper → Beer [0.5%, 75%]"
  • Support and confidence are used to evaluate associations
  • Mine patterns and rules efficiently in large datasets
  • Use these patterns for classification, clustering, and other applications

Why Data Mining?

• Discover interesting patterns and knowledge from massive amounts of data
• A natural evolution of database technology with wide applications
• A KDD (Knowledge Discovery in Databases) process includes data cleaning, integration, selection transformation, mining, pattern evaluation, and knowledge presentation
• Mining can be performed in a variety of data formats

What Is Data Mining?

• Also known as Knowledge Discovery from Data (KDD)
• Extraction of interesting (non-trivial, implicit, previously unknown and potentially useful) patterns or knowledge from massive amounts of data

Data Mining Tasks

• Association
• Classification
• Clustering
• Outlier and trend analysis

Data Mining Applications

• Credit card fraud detection
• Direct marketing
• Classifying stars, diseases, web-pages
• Understanding customer behavior
• Identifying trends in sales data
• Detecting anomalies in network traffic

Major Issues in Data Mining

• Data quality
• Scalability
• Efficiency
• Interpretation
• Visualization

Data Mining Technologies and Applications

• From major dedicated data mining systems/tools (e.g., SAS, MS SQL-Server Analysis Manager, Oracle Data Mining Tools) to invisible data mining

Examples of Classification Task

• Classifying credit card transactions as legitimate or fraudulent
• Classifying land covers (water bodies, urban areas, forests) using satellite data
• Categorizing news stories as finance, weather, entertainment, sports
• Identifying intruders in cyberspace
• Predicting tumor cells as benign or malignant
• Classifying secondary structures of protein

Regression

• Predict a value of a given continuous-valued variable based on the values of other variables
• Linear or nonlinear models
• Examples:
  • Predicting sales amounts of a new product based on advertising expenditure
  • Predicting wind velocities as a function of temperature, humidity, air pressure
  • Time series prediction of stock market indices

Clustering

• Finding groups of objects
• Objects in a group are "similar" to each other but "different" from objects in other groups
• Intra-cluster distances are minimized and inter-cluster distances are maximized

Applications of Cluster Analysis

• Understanding
  • Customer profiling for targeted marketing
  • Group related documents for browsing
  • Group genes and proteins with similar functionality
  • Group stocks with similar price fluctuations
• Summarization
  • Reduce the size of large datasets

Recommended Reference Books

• "Mining the Web: Statistical Analysis of Hypertex and Semi-Structured Data" by S. Chakrabarti
• "Pattern Classification" by R.O. Duda, P.E. Hart, and D.G. Stork
• "Exploratory Data Mining and Data Cleaning" by T. Dasu and T. Johnson
• "Advances in Knowledge Discovery and Data Mining" by U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy
• "Information Visualization in Data Mining and Knowledge Discovery" by U. Fayyad, G. Grinstein, and A. Wierse
• "Data Mining: Concepts and Techniques" by J. Han and M. Kamber
• "Principles of Data Mining" by D.J. Hand, H. Mannila, and P. Smyth
• "The Elements of Statistical Learning: Data Mining, Inference, and Prediction" by T. Hastie, R. Tibshirani, and J. Friedman
• "Web Data Mining" by B. Liu
• "Machine Learning" by T.M. Mitchell
• "Knowledge Discovery in Databases" by G. Piatetsky-Shapiro and W.J. Frawley
• "Introduction to Data Mining" by P.-N. Tan, M. Steinbach, and V. Kumar
• "Predictive Data Mining" by S.M. Weiss and N. Indurkhya
• "Data Mining: Practical Machine Learning Tools and Techniques" by I.H. Witten and E. Frank

The Evolution of Data Science

• 1950s-1990s: Computational Science
  • Most disciplines developed a third branch – computational
  • Traditionally focused on simulation
• 1990-now: Data Science
  • Flood of data from new scientific instruments and simulations
  • Cost-effective storage and management of petabytes of data
  • The Internet and computing Grid made archives universally accessible
  • Data mining is a major new challenge!