Chapter 9 In Class Notes Flashcards
information
systems that process operational and other data to
identify patterns, relationships, and trends for use by
business professionals and other knowledge workers.
Business intelligence (BI) systems
patterns, relationships,
trends, or predictions
Business intelligence
The software component of a BI system is referred to
as a
BI application
Big Data
• A problem for data processing, organization and
storage
– Nonrelational data stores (a.k.a. NoSQL databases)
enter stage right. With celebrity endorsements by the
likes of Amazon (Dynamo), Google (Bigtable), and
Facebook (Cassandra), NoSQL architectures are gaining
in popularity and acceptance and are likely to have a big
impact on the BI space in the near future
BI Applications process data to produce business
intelligence via:
– Reporting
– Data Mining
– Knowledge Management
Where Does BI Data Come From?
- Operational Data
- Purchased Data (Data Vendors)
- Human Knowledge
Q1: How Do Organizations Use Business
Intelligence (BI) Systems?``
They make BI Applications that manage operational and purchased Data, and Human knowledge
What is BI used for? 4 things
Project management
problem solving
deciding
informing
Q2: What Are the Three Primary
Activities in the Business Intelligence
Process?
• Collecting Data: Obtaining, cleaning, organizing,
relating, and cataloging source data.
• Performing Analysis: Creating business intelligence
via reporting, data mining, and knowledge
management
• Publishing Results: Delivering business intelligence
to the knowledge workers who need it.
– Push publishing: BI is delivered without any request on
the part of the user (scheduled or event driven)
– Pull publishing: BI is delivered to users upon request
Push vs. Pull: Implications for BI?
• What are some suggestions for maintaining a healthy
dynamic between your reporting team and the end
users? 4 of them
– Voice of customer – Steering committee – Data sufficiency – Islands of automation and sharing of reporting information
Why Do Organizations Use Data
Warehouses and Data Marts to Acquire
Data?
• Creating reports and performing analyses from operational data is
generally not recommended
– Security and control
– Data structure is structured for fast and reliable transaction
process and not for BI analysis (one of the potential
downsides of Nonrelational database systems)
– BI analysis can be very resource intensive
• For these reasons, data is often extracted to a data warehouse
for BI analysis
Components of a Data Warehouse 3 of them
Cental is Data warehouse DBMS. 2 components coming off center are data warehouse metadata and data warehouse database
Data Warehouses
• Functions: (4 of them)
– Obtain data
– Cleanse data
– Organize and relate data
– Catalog data
Data Warehouses vs. Data Marts
• Data warehouse – Obtain data – Cleanse data – Organize and relate data – Catalog data • Data Mart: A collection of data smaller than a data warehouse that addresses the needs of a particular department or function
Q4 How Do Organizations Use Typical
Reporting Applications
Reporting applications apply reporting operations to
data to produce business intelligence.
Q4 How Do Organizations Use Typical Reporting Applications • Reporting applications apply reporting operations to data to produce business intelligence. • Basic operations: (5 of them)
Sorting filtering grouping calculating formatting
RFM Analysis and Report
Customers are rank-ordered according to their
purchasing patterns (1-5, 20th percentiles) with
1. how recently a customer has ordered.
2. how frequently a customer orders
3. how much money a customer has spent
Provides an
ability to sum, count, average, and perform
calculations on groups of data.
Online analytical processing (OLAP)
data item of interest that is to be summed or
averaged (e.g. total sales, returns, etc.)
Measure
a characteristic of a measure (date of
purchase, customer zip code, type of customer, etc.)
dimension
The application for statistical techniques to find patterns
and relationships among data for classification and
prediction
data mining
Q5 How Do Organizations Use Typical
Data-mining Applications?
Data mining is the central location. coming off of data mining is artificial intelligence machine learning. data management technology. cheap computer procesing and storae , statistics/mathematics, huge databases
Unsupervised vs. Supervised Data Mining
Unsupervised 1. no model before running analysis 2. hypotheses created after analysis 3. cluster analysis to find groups Supervised 1. model created before analysis 2. hypotheses created before analysis 3. regression analysis make predictions.
- Used for predicting values and making classifications
* Complicated set of nonlinear equations
Neural Networks (Supervised)
• Data mining technique for determining sales patterns
• Shows products that customers tend to buy together
– Cross-selling opportunities
– Support: The probability that two items will be purchased
together
– Lift: How much the base probability increases or
decreases when other products are purchased
Market Basket Analysis (Unsupervised)
• Data mining technique for determining sales patterns
• A hierarchical arrangement of criteria that predict a
classification or a value.
Decision Tree (unsupervised)
Ethics Guide: The Ethics of Classification
• Classifying applicants for college
• University collects demographics and performance
data of all its students
• Uses decision tree data mining program
• Uses statistically valid measures to obtain statistically
valid results
• No human judgment involved
“the process of
creating value from intellectual capital and sharing
that knowledge with employees, managers,
suppliers, customers and others who need it.”
knowledge management (KM)
Q6. Role/Benefits of KM Systems
(Santosus and Surmacz) 5 of them
- Encourage free flow of ideas.
- Improve customer service by streamlining response
time. - Boost revenues by getting products and services to
market faster. - Enhance employee retention rates by recognizing
and rewarding knowledge sharing. - Streamline operations and reduce costs
Three Major Categories of Knowledge
Assets
- Data
- Documents
- Employees
most important content function in KM
applications
Indexing
subscribing to
content sources
Real Simple Syndications
place where employees share their
knowledge that may include RSS feeds
Blogs
• Encode human knowledge as Rule-based systems
(IF/THEN)
• Rules created by interviewing experts
Expert Systems
• Major problems with ES:
– Expensive to develop
– Unpredictable maintenance
– Over hyped
Q7 What Are the Alternatives for
Publishing Business Intelligence?
Email or collaboration tool
Web server
SharePoint
BI Server
Components of a Generic Business
Intelligence System
You have metadata on top. Then on the second row you start with the BI application, which is BI data source, BI Application, and BI application results. Then it goes to BI Server, then you have push an pulls from BI server to Any Device. You have BI users using “Any Device”
machines can build their own
information systems. What is this called?
Social singularity
Q8: 2022
• Companies will know more about your purchasing
habits and psyche.
• Social singularity — machines can build their own
information systems.
• Will machines possess and create information for
themselves?
