What are the major challenges to large corporations in terms of information management for supporting decision making?
The pyramid
- Data Mining (Patern Discovery and Evaluation)
- Data Exploration (OLAP)
- Data Warehouses/Data Marts (Data selection)
What are the main limitations of conventional information systems, i.e. the DBMS technology, in terms of information queries?
- Data is growing at a phenomenal rate (“the yotta world”)
- Data types are more mixed and complicated
- Data rich but information poor!
- New strategies for Decision Supporting Systems (DSS) or BI (Business Intelligence) system
UNCOVER HIDDEN INFORMATION AND PATTERNS
Name three information process for Decision Support Systems (DSS)
- On Line Transactional (information) Process (OLTP)
- On Line Analytical (information) Process (OLAP)
- Knowledge discovery from data (KD)
What is OLTP and what does it do?
On Line Transactional (information) Process (OLTP)
Query is viewed as a read-only transaction
Track/record/retrieve original data records of every day business operations for answering “what, when, where” type of questions: Operational databases (Relational DB and SQL)
What is OLAP and what does it do?
On Line Analytical (information) Process (OLAP)
Summarization, consolidation, and aggregation
Store & manipulate summaries of various groupings of original data records for answering “what happened to the business” type of questions - Analytical databases: Data warehouses and OLAP
What is KD and what does it do?
Knowledge discovery from data
Discover/analyze hidden patterns of abstractive information (knowledge) for answering “why and what to happen next” type of questions: Data Mining (DM)
As a common user of a DB system, such as Dal online or RBC online banking, which type of information process do you deal with & why?
On Line Transactional (information) Process (OLTP)
Students only need a ‘view read-only’ data from the database without any manipulation.
For the president of Dalhousie Univ. or the dean of FCS, what type of information they are interested in getting?
On Line Analytical (information) Process (OLAP)
The president and dean could benefit from aggregations of Dal’s data such as “total male and female students” or “total first year students”
What is the “data rich but information poor situation?
The abundance of data, coupled with the need for powerful data analysis tools
As a store manager of War-Mart or Superstore, what type of information you need to know all the time?
???? Abstractive information (Knowledge discovery)
What is data?
- raw measures
- unprocessed
- some relevancy
- has not structure (per say)
What is information?
- Structured data
- Processed data that brings meaning
- Information can be used to answer questions
What is knowledge?
- Laws or rules
- Generalized to higher levels
- Corresponds to regularity patterns hidden in datasets
- Explain what happened
- Predict what’s next
Use examples to explain the differences between terms Data, Information and Knowledge. How does each term/concept link to business information queries according to three types of information processes?
Data –> OLTP
Information –> OLTP
Knowledge –> KD
Why can’t you ask questions and get answers from a dataset and why?
They are data, but not information!
You need information, that is, structured data, processed and provides meaning.
Why the IT industry needs to develop DM and DW considering that RDBMS/SQL are already available for storing and querying information?
- Efficient for data retrieval
- Not efficient for grouping large data sets
- Difficult to use SQL do define complex queries
- Analyzing data and exploring relationship are not part of the SQL vocabulary.
- It is constrained to retrieve information from single database
Why and in which way DW model is more advanced than RDB model in supporting business management queries?
- Integrated data from one or more disparate sources
- Current and historical summarized data
- -> Can created reports
- -> Complex queries
What is data mining (DM)?
Finding unknown, valid and actionable knowledge (patterns and regularities) from large data.
What are the basic knowledge discovery tasks?
- Classification
- Association
- Clustering
- Generalization
What are the simple rules for choosing solution tools for getting different types of business information?
?????? (Slide 1.3)
What are the two general purposes of DM (or any scientific research)?
– Explanation: understanding/explaining about current behaviours
– Prediction: predicting for future outcomes
How the DM technologies are categorized?
- Predictive (Classification and prediction)
- Descriptive (Clustering, summarization and association rules)
Can you name three major DM tasks & what is each task about?
- Classification is learning a function that maps (classifies) a data item into one of several predefined classes.
- Clustering is a common descriptive task where one seeks to identify a finite set of categories or clusters to describe the data.
- Summarization involves methods for finding a compact description for a subset of data.
What is the Empirical Cycle Model (ECM) of scientific research, describe each stage of the process?
- Observation: we start with a number of observations.
- Analysis: we try to find patterns in these observations.
- Theory: if we have found some regularities, we formulate a theory (hypothesis) explaining the data.
- Prediction: our theory will predict new phenomena that can be verified by new observations.
