Round 2 Flashcards
BI is:
A set of methodologies, tools and technology driven processes used to analyse data. Computerised support for data driven decision making. It helps transform data into information (and knowledge) to decisions and actions
Data Visualisation is:
Use of visual representation to explore, make sense of and communicate data.
What Data Visualisation does:
1) Pushes people into action
2) Big data is hard to understand and explore
3) traditional graphs can only handle 2-3 variables but Histograms, geographic maps, highlight tables and tree maps handle more.
Information Visualisation is:
Always useful. (No errors)
Summarised: Aggregation and contextualisation.
Challenges of BI implementation:
Develop or acquire?
CVB analysis
Security and Privacy protection
Integration of systems and applications
Agile BI Benefits:
Close collaboration 2-6 week cycles Based on feedback Consumers feel good having input Less risk of low adoption
Choosing Bi Tools
S
CRISP - Data Mining:
1 - Business Understanding 2 - Data understanding 3 - Data preparation 4 - Model building!! 5 - Testing and Evaluation 6 - Deployment
Some Decision Tree Terminology
Root Node - Attribute
Leaf Node - Final Class Label
Branch = Outcome
First Branch = Highest Information again
Content Mining
Page content (HTML format, text, images)
Classify and cluster web pages
Sentiment Analysis
Structure Mining
Hyperlink structure
Authoritative web pages
Usage Mining
Sequence of clicks by session
Make better recommendations
Information Governance:
Policy based control of information to meet all legal, regulatory, risk and business demands
Data or Info Governance, which is which?
If policy is relevant regardless of context - Data. Ie storing CCV
If only relevant during some business processes - Info. DNCall register
Ethical Reasoning involves 4 fundamental questions:
Who is involved?
What action taken or being considered?
What are consequences?
Is it fair and just for all stakeholders?
