FIT5145: Workshop 2

FIT5145 Week 3 (Workshop 2)

FIT5145: Workshop 2

Recap on this week's learning objectives

  • Identify different analytic levels in a data science project and comprehend the difference between levels
  • Understand and analyse the growth laws
  • Explain general models for understanding businesses and decision making
  • Analyse business models specific to data science in different organisations
  • Utilise basic statistical analysis
  • Comprehend the power of visualisation
FIT5145: Workshop 2

Act 1: Data models in Orgs (collaborative 🤼)

  1. Get together as a group + pick an industry: clothing retail, car manufacturing, banking, taxi, sporting, law enforcement, etc.

  2. Within the scope of the industry you've chosen:

  • Outlook on pre-internet data in an organisation?
  • Change in the attitude towards/use of data in the last 20 years? (Refer to growth laws below)
    • Power/cost (Moore/Koomey)
    • Technological advancements (Bell)🏃‍♀️
    • Data privacy (Zimmerman)
  • Role of data science in enabling these changes?
FIT5145: Workshop 2

Act 2: Analysing cricket data (self-directed 🚶🏃‍♀️)

  • Go to Lab Work in Moodle
  • Work through the provided RMD file along with CricketData.csv

According to SAS, there are eight levels 1. Standard reports - What happened? and When did it happen? Used commonly in monthly or quarterly financial reports 2. Ad hoc reports - How many? How often? Where? Possible scenarios include custom reports that describe the number of patients for every diagnosis code for each day of the week. 3. Query drill-dow (aka OLAP) - Where exactly is the problem? How do I find the answers? E.g. Sort and explore data about different types of mobile phone users and their calling behaviours 4. Alerts - When should I react? What actions are needed now? e.g., sales execs receive alerts sales targets are falling behind? 5. Statistical analysis - Why is this happening? What opportunities am I missing? e.g., Banks may discover why an increasing number of customers are refinancing their homes. 6. Forecasting - What if these trends continue? How much is needed? When will it ne needed? e.g., Retails can predict how demand for individual products will vary from store to store, region to region. 7. Predictive modelling - What will happen next? How will it affect the business? e.g., Hotels and casinos can predict which VIP customers will be more interested in a specific offer. 8. Optimisation - How do we do things better? What is the best decision for a complex problem? e.g., Given business priorities, resources constraints and available technology, determine the best way to optimise the IT platform to satisfy the needs of every user.

We cover four laws: Moore, Koomey, Bell and Zimmerman - Moore's Law suggests that the number of transistors on a microchip doubles every two years, though the cost of computers is halved i.e. the power of computers doubles every two years. - Koomey's Law is a corollary to Moore's Law, which suggests that the energy efficiency of computers doubles every 1.57 years -> ubiquitous computing - Bell's Law suggests that new computer classes form roughly every decade, opening up new opportunities for computing -> e.g., mainframes, minicomputers, PCs, laptops, smartphones, IoT devices - Zimmerman's Law suggests that the number of people who can access a piece of data increases exponentially over time, leading to a privacy paradox -> e.g., the more data is shared, the more people can access it, but the more people can access it, the more likely it is to be misused.

This task is to get students to think about what data organisations work with, how IT has changed the way data is used, how the attitudes of organisations towards data have changed, and how these changes could be connected through data science.