Data Mining

data mining - correct answers discovering interesting patterns from large amounts of data KDD stands for - correct answers Knowledge Discovery for Databases KDD Process - correct answers Data Cleaning, Data Integration, Data selection, transformation, data mining, pattern evaluation, knowledge presentation Data Mining Functionalities (7) - correct answers Characterization, Discrimination, Association, Classification, Clustering, Outlier and Trend Analysis Classification Definition - correct answers Assumes there is more than just data; assumes there is a label too. Every data point x has a class y. Goal produce a function such that for f(x) = y Classification Requirements (3) - correct answers High Accuracy Understandable by Humans Fast for very large sets of data Clustering Definition - correct answers Like classification without labels. Outputs k groups called clusters. Goal: given data X, assign every x in X to some cluster. This is unsupervised learning because there are no labels/classes given Patterning Mining Definition - correct answers Look for patterns based on: if a & b & c then RESULT Classification spits out the best match of class. Pattern mining focuses more on the rules: a & b & c Data Warehouse Definition - correct answers a data repository for an organization set up to support strategic decision making. They are clean and have noise removed. Building from (potentially) other databases. How big is a data warehouse - correct answers Huge in size storing billions of records. Data Warehouse is the following: - correct answers Subject Oriented, Integrated, Time-variant, Non-volatile Data Warehouse - Subject Oriented - correct answers gives information about a particular subject and not (necessarily) a company's ongoing operations Data Warehouse - Integrated - correct answers Data is gathered into a data warehouse from a variety of sources and merged into a coherent whole Data Warehouse - Non-volatile - correct answers More data is added but it is never removed. Allows management to gain a consistent picture of the business Data Warehouse - Time-variant - correct answers All data is identified with a particular time period Enterprise Data Warehouse - correct answers collects all information about subjects (customers, products, sales, assets, personnel) that span the entire organization Data Mart - correct answers Departmental subsets that focus on selected subjects DSS Stands for - correct answers Decision Support System Decision Support System (Definition) - correct answers Information technology to help the knowledge worker make faster and better decisions OLAP stands for - correct answers Online Analytical Processing Online Analytical Processing (Definition) - correct answers an element of decision support systems (DSS)