importance of data(analyis)

  • valuable insights
  • better, more informed decisions
  • help personalizing products services
  • identity potential risks

regular

  • no need to face clients
  • raw Data -> information

steps

  1. Collecting data
  2. Cleaning data
  3. Transforming data
  4. Analyzing data
  5. Visualizing data

purpose of DA

  • make informed decision
  • solve problems
  • test hypotheses

type of DA

  • descriptive
    historical data to describe what has occured in the past
  • diagnostic
    explain why certain events,investigates the courses
  • predictive
    make predcition
  • prescriptive
    specific actions to optimize resute

congnitive analytics

  • use of advanced algorithm
  • machine learning
  • process and analyze complex data
  • Natural Language Processing(NLP)
  • Reasoning and Inference
  • Context awareness
  • pattern recognition

data modeling

goal: illustrate the types of data
3 levels:

  • Conceptual
    entity,classes,characteristics,relationships
  • Logical model
  • Physical model

model

  • data analytics model
  • physical or abstract
  • real world x model world
  • formulation -> analysis -> interpretation

type of model

  • business model
  1. One-time decision model: Used once
  2. Decision support models: intergrated package
  3. Models embedded in computer system
  • decision support models
  1. decision trees(contain decision nodes)
  2. scenario analysis
  3. Linear Programming
  4. Monte Carlo Simulation
  5. Expert Systems(AI-based)

Roles in the CRISP-DM process

  1. business analysis(understanding)
    steps:determine business objectives
  • Backgroud
  • Business Objectives
  • Business Success Criteria
  1. analysis data(understanding)
  • high level
  • low level
  1. data preparation
  2. modeling
  • modeling tech
  • Select Modeling Techniques
  • Generate Test Design
  • Build Model
  • Assess Model
  1. evaluation
  • Evaludate Results
  • Review Process
  • Determine Next Steps
  1. deployment
  • Plan Deployment
  • Plan Monitoring and Maintenance
  • Produce Final Report
  • Review Project

CRISP-DM almost took half of the most commonly use for data science project

6 Step Problem solving

  • who
  • when
  • where
  • what
  • why
  • how