Database Design: Special Databases, Data Warehouses, Data Mining and Information Retrieval, Study notes of Principles of Database Management

Download Database Design: Special Databases, Data Warehouses, Data Mining and Information Retrieval and more Study notes Principles of Database Management in PDF only on Docsity! CMSC 424 – Database design Lecture 25 Special databases Data warehouses Data mining/Information retrieval Mihai Pop Admin • Course evaluation: http://www.CourseEvalUM.umd.edu • Review sessions: Thursday & Monday – e-mail me topics to cover, questions, problems, etc. R-tree (chap. 24) • Binary search tree on Y-coordinate • Each internal node contains search structure on X-coordinate for all points with Y coordinates in the corresponding subtree OLAP (chap. 18) ■ On-line Analytical Processing ■ Why ? Exploratory analysis • Interactive • Different queries than typical SQL queries Data CUBE A summary structure used for this purpose –E.g. give me total sales by zipcode; now show me total sales by customer employment category Much much faster than using SQL queries against the raw data –The tables are huge ■ Applications: – Sales reporting, Marketing, Forecasting etc etc Cross Tabulation of sales by item-name and color ■ The table above is an example of a cross-tabulation (cross-tab), also referred to as a pivot-table.  Values for one of the dimension attributes form the row headers  Values for another dimension attribute form the column headers  Other dimension attributes are listed on top  Values in individual cells are (aggregates of) the values of the dimension attributes that specify the cell. Data warehouses • Brute-force solution to federation: – download all databases – convert them to a common schema – provide a common interface • Problems: – data storage & duplication – hard to keep up to date – performance (single point of entry/ failure) • Examples: – GenBank (US biological data repository) – Ensembl (EU biological data repository) Data Mining • Searching for patterns in data – Typically done in data warehouses ■ Association Rules: When a customer buys X, she also typically buys Y Use ? • Move X and Y together in supermarkets – A customer buys a lot of shirts Send him a catalogue of shirts Patterns are not always obvious • Classic example: It was observed that men tend to buy beer and diapers together (may be an urban legend) • Other types of mining Classification Decision Trees Information retrieval (chap. 19) • Extracting meaning from data • Examples: – Google (document indexing/ranking) – Image search – Automatic annotation of documents, e.g. extracting information from bio-medical literature