Chapter 12: Business Intelligence Flashcards
Process + Phases in Making a Decisions (check notes, this may be inaccurate)
i.e a rainy day outside, what do you take to cover yourself
- You examine the reality/problem, this examination leads to the
Intelligence Phase, where you see “What is the problem”
In this case, it’s trying to cover yourself from the rain
- Next you see/list the options for the decisions in
the Design Phase “What are My Options”
i.e.: You could use a Hoodie Jacket, an Umbrella, or your Backpack to cover yourself from the rain
You can validate the model (???) when you apply it to the reality of the Situations
- The Choice phase is where you” Pick an Option + Decide how to Implement it”
The Verification + Testing of the proposed solution can be applied to the problem in reality
i.e. you pick the Umbrella
- The implementation of the solution:
See if the option proposed worked
If it did = Success
If it didn’t = Failure, go back to the decision It’s process
Not every option works all the time, sometimes you can wear a Hoodie on lighter rains than carrying around an umbrella
3 Basic Roles of a Manager
Interpersonal
Informational
Decisional
Decision making is difficult due to following reasons
- # of alt’s is constantly increasing
- Most decisions must be made under time pressure
- Increased Uncertainty in the dec environment
- Often necessary to Rapidly Access Remote Info, Consult with Experts, or conduct a Group dec - making session
A Framework for Computerized Dec Analysis
Includes:
- Problem Structure (Structured (seen before_, Semi - Struct (somewhat seen before?), Unstruc decs (unique?))
- Nature of Decisions:
3 broad categories of Managerial Dec’s/of managerial control in dec - making:
Operational Control (1st Line Mang/Emps?)
Management Control (Higher up Mang’s)
Strategic Planning (Top Mang/CEO?)
- The Decision Matrix (links prob Strcut + Decision Nature together + IS used to support solution/dec?)
Computer Support for Structured Dec’s
Dec - Making + Problem - Solving occur
At each level in an org
Operational Dec - Making
Employees Develop, Control, _ Maintain core Bu activities required to run the day - 2 - day operations
Structured Decisions
Situations where established processes offer potential solutions (i.e. Problem in Factory machine, Struc Dec = Push button to call engineer)
Managerial Dec - Making
Emp’s/Mang’s? evaluate Company operations to Identify, Adapt to, + Leverage Change
Semistructured Dec’s
Occur in situations in which a few established processes help to evaluate potential solutions, but not enough to lead to definite recommended decisions
(i.e. Mang wants to fire emp based on claims of workplace shenanigans, knows criteria for firing, but how to decide on seeing if allegations true???)
Strategic Dec Making
Mang’s Develop overall Strats, Goals + Objectives
Unstruct Dec’s
Occurs in situations in which No procedures or rules exist to guide dec - making toward the correct choice
Scope of Bu Intelligence
The Development of 1 or a Few Related BI Applications
The Development of Infrastructure to Support Enterprise-wide BI
Support for organizational Transformation
Problem: Data Rich, Info Poor
Bu’s face a Data Explosion with Digital Images, Email in-boxes + broadband connections
The amount of data generated is doubling/2x ing every yr
Some believe it will son double monthly
BI Can Answer Tough Q’s
It starts from answering broader to answering specific questions
(i.e. Q: Why are Sales below Target —> A: Because we sold less in Western Region
…….
Q: Why did Cu Complaints inc? —-> A: Because late Deliveries went up 60%
Bu Analytics Process
- Starts with bU Problem/Org Pain points
- Moves that into Data Management Ware house (with things like External, Internat, Big Data being ETL(clean) to be put? in Data Warehouse)
- It then goes through Descriptive Analytics (i.e. Descriptive Statistics, Data Mining, Dec Support Systems) which shows:
“What has happened?”
- The through Predictive Analytics (i.e. Linear Regression, Logistic Regression, Multiple Regression) to see:
“What could happen?”
This uses Presentation tools like: Dashboards, Figures, Graphs + Reports
- Then it goes through Prescriptive Analytics (i.e. Optimization, Simulation + Dec Trees) to see:
“What should we do?”
- This all helps lead to Actionable Bu Dec’s/Recommended Actions
- After that you ask new Q’s
Underlying Technologies for the Previous Process
Increasingly Powerful Chips (GPUs)
Cheaper, Higher - Volume Storage
Increasingly Faster Broadband Internet
Neural networks
Machine learning
Deep Learning
Bu Analytics Tools
Excel
Multidimensional Analysis or Online Analytical Processing (OLAP)
Data Mining
Dec Support Systems (DSS)
Data Mining + Bu Analytic Tools:
Predictive Data - Mining Applications (Apps) are used in:
Retailing + Sales
Banking
Manufacturing + Production
Insurance
Police Work
Healthcare
Marketing
Prescriptive Analytics
Statistical procedures include: Optimization, Simulation + Dec Trees
Dec Support System (DSS)
Sensitivity Analysis (0
What - if Analysis ()
Goal - Seeking Analysis ()
Artificial Intelligence
The Theory + Development of IS’s able to perform tasks that normally require human intelligence
Intelligent Behaviour
Learning or Understanding from experience, making sense of ambiguous or contradictory messages, + responding quickly + successfully to new situations
Algorithm
A Problem - solving method expressed as a finite sequence of steps
Technological Advancements that led to Advancements in AI
Advancements in chip Technology
Big Data
The Internet + Cloud Computing
+ Improved Algorithims
Natural vs Artificial Intelligence
Large table, look at it on Page 26
AI Creativity could be a bit diff than what on table (but that’s Teacher’s opinion)
Ai Technoogies 1: Expert Systems
Transfers expertise from a domain expert (or other source) to the system
AIT 2: Machine Learning
Is the Ability to perform new, unseen tasks built on known properties from training or historical data that are labeled
AIT 3: Deep Learning
Subset of Machine Learning in which the system discovers new patterns without being exposed to labeled or historical or training data
AIT 4: Neural Network
A set of virtual neurons or central processing units (CPU’s) that work in parallel in an attempt to simulate the way the human brain works, although in a greatest simplified form
AI Applications
- Computer Vision:
Refers to the ability of IS’s to identify objects, scenes, and activities in Images
Natural Language Processing
Robotic
Speech Recognition
Intelligent Agents
Infor Agents search for information + dispatch it to users
Best known Info Agents = Buyer Agents
Buyer agent (or Shopping Bot): Helps Cu’s find the Pr & Sr’s they need on a Web Site
Dashboards
Brain is wired to look at visual inputs
You look at gauge not spreadsheet when driving car
Dashboards can make data easier to visualize/understand
i.e. Gauges to see Department Performance, Revenue per sale, etc
Dashboard Capabilities 1: Drill Down
The ability to go to details, in several levels
It can be done in a series of menus or by clicking in a drillable portion of the screen
DB2: Critical Sucess Factors (CSFs)
The factors most critical for the success of the Bu
These can be: Organizational, Industry, Departmental, or individ workers
DB3: Key Performance Indicators (KPI’s)
The specific measures of CSF’s
DB4: Status Access:
The latest data available on a KPI or some other metric, often on real time (current?)
DB5: Trend Analysis
S, Medium, + L - term trend of KPI’s or metrics, which are Projected using forecasting methods
DB6: Exception Reporting
Reports that highlight deviations larger than certain thresholds
Reports may include only deviations
Data Visualization Technologies
Geographical IS’s:
Geocoding
Reality Mining:
Graphical IS (GIS) + Geographic Positioning Systems (GPS) together produce an interesting new type of tech which allows analysts to extract info from usage patterns of mobile phones + other wireless devices
i.e
NYC
Singapore
Other Major cities
(found in these cities or examples of cities where Reality Mining is used???)
Bu in Action: Corporate Performance Management (CPM)
Uses Key Performance Indicators (KPI)
CPM =
KPI’s
The quantifiable metrics a comp uses to evaluate progress toward Critical Success Factors
i.e.:
Turnover Rate of emp’s
of Product Returns
of New Cu’s
Avg Cu Spending
