ITM 618 FINAL EXAM 100% SOLVED
What is Descriptive Analytics - ANSWER Describes what happened - use in summarizing
historical results (e.g. how many iPhones sold last year)
What is Predictive Analytics - ANSWER Predicts what will happen - use to make an
educated guess at likely results in the future (e.g. predict which advertisement will be
more successful)
What is Prescriptive Analytics - ANSWER Make decisions and tells what should be done -
use when there are important, complex, or time-sensitive decisions to make (e.g. offer
incentives to customers that are likely to leave the business)
What is a Data Warehouse - ANSWER A data store that integrates data from different
sources throughout an organization for decision support purposes
Subject-oriented - organized around major subjects such as customer, product, sales,
etc..
Integrated - constructed by integrating multiple, heterogeneous data sources
Time-variant - Stores historical data, does not store current information like an
operational database
Non-volatile - Physically separate store of data transformed from the operational
environment, much less frequent updates compared to operational data
What is ETL - ANSWER Extract - gathers data from multiple, heterogeneous and external
sources
Transform - Converts data from legacy format to warehouse format
Load - Sorts, summarizes, consolidates, computes views, checks integrity, and builds
indices and partitions
Cleaning: Detects errors in the data and rectifies them when possible
What are the basic OLAP Operations - ANSWER Roll-up - Climbing hierarchy of a
dimension from more to less specific (e.g. given sales by quarter, city, and product, we
can roll up to get sales by province, by quarter, by product)
Drill Down - Inverse of Roll-up, climbing down hierarchy of dimension from less to more
specific (e.g. given sales by product, quarter, province, we can get sales by product,
quarter, city)
Slice - Takes one dimension from a cube and represents a new sub-cube based on the
slice, like a filter function to look at specific areas
, Dice - Selection on two or more dimensions, similar to slice (e.g. Dice for (Location =
Toronto or Ottawa) and (Time = 3Q or 4Q) and (Product = TV or Printer)
Pivot - Re-orient the multi-dimensional view for visualization purposes
Operational Database vs. Data Warehouse - ANSWER Operational Database - OLTP
(online transactional processing), day to day operations: purchasing, inventory,
banking, manufacturing, payroll, registration, accounting
Data Warehouse - OLAP (online analytical processing), Data analysis and aggregation
for decision making
What is ROLAP - ANSWER Relational OLAP - use relational or extended-relational DBMS
to store and manage warehouse data
Scalable: changes to the number of values in a dimension can be handled easily without
complete reorganization
No direct access to cells
What is MOLAP - ANSWER Multidimensional OLAP - array-based multidimensional
storage engine
Fast indexing to pre-computed summarized data
Fast positioning the array with indexes
Less scalability: Even minor changes to structure require a complete reorganization
What is HOLAP - ANSWER Hybrid OLAP - combines ROLAP and MOLAP technology
Benefits from scalability of ROLAP and faster computation of MOLAP
What is a Data Mart - ANSWER Department wide data storage - a subset of
enterprise-wide that is of value to specific groups, with focus on a single subject
What is Data Mining - ANSWER The process of extracting interesting (non trivial,
implicit, previously unknown, potentially useful) knowledge or patterns from data in
large databases
Discover knowledge that characterizes general properties of data
Discover patterns in previous and current data to make predictions for future data
What are the 6 processes in CRISP-DM - ANSWER Business Understanding
Data Understanding
Data Preparation
Modeling
What is Descriptive Analytics - ANSWER Describes what happened - use in summarizing
historical results (e.g. how many iPhones sold last year)
What is Predictive Analytics - ANSWER Predicts what will happen - use to make an
educated guess at likely results in the future (e.g. predict which advertisement will be
more successful)
What is Prescriptive Analytics - ANSWER Make decisions and tells what should be done -
use when there are important, complex, or time-sensitive decisions to make (e.g. offer
incentives to customers that are likely to leave the business)
What is a Data Warehouse - ANSWER A data store that integrates data from different
sources throughout an organization for decision support purposes
Subject-oriented - organized around major subjects such as customer, product, sales,
etc..
Integrated - constructed by integrating multiple, heterogeneous data sources
Time-variant - Stores historical data, does not store current information like an
operational database
Non-volatile - Physically separate store of data transformed from the operational
environment, much less frequent updates compared to operational data
What is ETL - ANSWER Extract - gathers data from multiple, heterogeneous and external
sources
Transform - Converts data from legacy format to warehouse format
Load - Sorts, summarizes, consolidates, computes views, checks integrity, and builds
indices and partitions
Cleaning: Detects errors in the data and rectifies them when possible
What are the basic OLAP Operations - ANSWER Roll-up - Climbing hierarchy of a
dimension from more to less specific (e.g. given sales by quarter, city, and product, we
can roll up to get sales by province, by quarter, by product)
Drill Down - Inverse of Roll-up, climbing down hierarchy of dimension from less to more
specific (e.g. given sales by product, quarter, province, we can get sales by product,
quarter, city)
Slice - Takes one dimension from a cube and represents a new sub-cube based on the
slice, like a filter function to look at specific areas
, Dice - Selection on two or more dimensions, similar to slice (e.g. Dice for (Location =
Toronto or Ottawa) and (Time = 3Q or 4Q) and (Product = TV or Printer)
Pivot - Re-orient the multi-dimensional view for visualization purposes
Operational Database vs. Data Warehouse - ANSWER Operational Database - OLTP
(online transactional processing), day to day operations: purchasing, inventory,
banking, manufacturing, payroll, registration, accounting
Data Warehouse - OLAP (online analytical processing), Data analysis and aggregation
for decision making
What is ROLAP - ANSWER Relational OLAP - use relational or extended-relational DBMS
to store and manage warehouse data
Scalable: changes to the number of values in a dimension can be handled easily without
complete reorganization
No direct access to cells
What is MOLAP - ANSWER Multidimensional OLAP - array-based multidimensional
storage engine
Fast indexing to pre-computed summarized data
Fast positioning the array with indexes
Less scalability: Even minor changes to structure require a complete reorganization
What is HOLAP - ANSWER Hybrid OLAP - combines ROLAP and MOLAP technology
Benefits from scalability of ROLAP and faster computation of MOLAP
What is a Data Mart - ANSWER Department wide data storage - a subset of
enterprise-wide that is of value to specific groups, with focus on a single subject
What is Data Mining - ANSWER The process of extracting interesting (non trivial,
implicit, previously unknown, potentially useful) knowledge or patterns from data in
large databases
Discover knowledge that characterizes general properties of data
Discover patterns in previous and current data to make predictions for future data
What are the 6 processes in CRISP-DM - ANSWER Business Understanding
Data Understanding
Data Preparation
Modeling
