数据挖掘-Pattern_Discovery_Lecture_1.1.pdf

Lecture 1: A Brief Introduction to Data Mining Lecture 1. A Brief Introduction to Data Mining  Why Data Mining?  What Is Data Mining?  Data Mining: A Knowledge Discovery (KDD) Process  Data Mining: A Multi-Dimensional View  Data Mining from Data Point of View  Data Mining from Knowledge Point of View  Data Mining from Methodology Point of View  Data Mining from Application Point of View 2 Why Data Mining?  Explosive Growth of Data  Data explosion: Our capability of generating, collecting, storing, and managing data has grown tremendously in the last 50 years  Major sources of abundant data  Web and computerized society: News, digital media, social networks, e-commerce, transactions, stocks, …  Science and engineering: Remote sensing, bioinformatics, scientific simulation, …  We are drowning in data but starving for knowledge!  “Necessity is the mother of invention”—Data mining—Automated and scalable analysis of massive data sets 3 What Is Data Mining?  Data mining: Knowledge discovery from data (KDD)  Extraction of interesting (non-trivial, implicit, previously unknown and potentially useful) patterns or knowledge from massive data  Data mining: A misnomer?  Knowledge mining from data!  Data mining: Its relationship with other disciplines  Machine learning, pattern recognition, statistics, databases, business intelligence, big data, …… 4 Data Mining: A Knowledge Discovery (KDD) Process Data cleaning Multi-dimensional data Pattern evaluation summary Data integration Pattern selection