Lecture 1: Intro to Digital Data Decision Making

 Data are a set of values of qualitative or quantitative variable about one

or more persons or objects, can be structured (numerical, ordinal,
nominal) or unstructured (text, audio, video)
 Data Science, use of techniques for understanding phenomena via the
analysis of data
 Data-Driven Decision-Making (DDD) refers to the
practice of basing decisions on the analysis of
data, rather than purely on intuition
 2 ways of making decisions, intuition vs Data or experience
 Decision Support Systems (DSS) – a class of computerized
information system that support decision-making activities. Typical
components include a database, knowledge base and user
interface

Lecture 2: Decision-Making and Decision Support Systems

Decision Support Systems
 “A class of computerized
information system that support decision-making activities. DSS are
interactive computer-based systems and subsystems intended to
help decision makers use communications technologies, data,
documents, knowledge and/or models to complete decision process
tasks”
 Typical components include - Database, knowledge base, user interface
Types of DSS
 Data driven DSS – focusing on systems and manipulation of large
databases of structured data
 Model driven DSS – focusing on access to and manipulation of a model,
rather than data
 Knowledge driven DSS – focusing on knowledge and recommending
actions to managers based on an analysis of a certain knowledge base. It
also has a “special problem” solving expertise and are closely related to
data mining
Problem Solving
 Can be characterised by some goal, an initial state and a set of
admissible actions which move the problem from one state to another,
hopefully towards a goal
  Usually, the choice of actions or transitions is quite limited. For example,
one cannot jump directly from initial state to the goal
Importance of expertise
 Experience/Expertise in solving some kinds of problems help to solve
new but similar problems faster
 Expertise is characterised by the ability to recognise patterns in problem
states to apply the correct response
 Expertise requires time
Expert Systems
 Computer programs that mimic human experts in their respective field,
allowing the computer to highlight and solve problems like how an
expert would
 This process has been utilised in many industries, for example medicine,
astrophysics, chemistry and finance

Main areas of application of Expert systems

 Interpretation — drawing high-level conclusions based on data
 Prediction — projecting probable outcomes
 Diagnosis — determining the cause of malfunctions, disease, etc
 Design — finding best configuration based on criteria
 Planning — proposing a series of actions to achieve a goal
 Monitoring — comparing observed behaviour to the expected
behaviour
 Debugging and Repair — prescribing and implementing remedies
 Instruction — assisting students in learning
 Control — governing the behaviour of a system
Advantages of ES
 Increased productivity
 Helps the availability of expertise
 Can be used in areas that are harmful to humans
Limitations of ES
 Difficulty in engineering, especially acquiring the expertise
 Mistrust by the users
 Effective only in specific areas
Rule-based expert systems
 ES based on declarative and procedural knowledge
 Declarative – these are propositions describing facts, objects or events
that are true. Facts are usually represented in a form of logical
propositions
 Procedural – logical rules that we can use to reason using the facts and
make decisions. Rules have the form IF condition THEN action (examples
on L2 slides p13)
Case-based expert systems
 ES based on concepts of similarity and analogy. The systems analyse past
cases and uses past solutions to solve current problems
 Operation Cycle of CBR – Retrieve, Reuse, Revise, Retain

Rule-Based vs Case-Based
• In rule-based systems a solution is achieved through an application of many
rules, inference
of facts, etc.
• In case-based systems the whole problem definition (case) is compared with
similar
problems, and the entire solution is applied at once.
• Rule-based systems usually work better for well-defined problems that do not
change with
time.
• Case-based systems can be used where problems are less understood and are
dynamic.

Lecture 3: Big Data, IoT and Digital Transformation

 Cloud computing: on demand availability of digital resources
 Machine learning: the culmination of online data so that AI can gather
information and perform its tasks.
 Big data: a set of large data that is too much for people to comprehend
and manipulate
 3 Vs of Big Data: Volume – amount and form of data, velocity – time of
collection and analysis of data, variety – type of data collected
 Importance of big data: cost savings, time reductions, understanding
market conditions, boost customer retention, marketing insights,
innovation and product development
  Internet of Things: devices with sensors and remote control, e.g. smart
houses with phone-controlled locks, lights etc
 Other definition: Idea of interconnected everyday smart objects
 IoT generates big data as each use of these smart objects can be used as
separate pieces of data.
 Digitization: Taking analogue information and encoding into digits so that
it may be used by computers OR creating a digital representation of
physical objects or attributes
 Digitalization: use of digital technologies to transform businesses and
social structures and provide new value creation opportunities
 Digitalization cannot occur without digitization
 Digital Transformation: all processes have been digitalized and cultural
change to adopt these new ways has been accepted by all members
internally involved.

15/10/24:
Big Data:
 Excel vs big data – benefits of using spreadsheets to manage big data
 Drawbacks – excel spreadsheets are highly vulnerable to errors.
Making Sense of Big Data
 Data -> Information -> Knowledge
 Data gets to information through conceptualisation and categorisation,
information gets to knowledge through contextualisation and
personalisation and can be reversed by explication and
conceptualisation.
 Use of Tableu, used in hospitals
 A startup is called digital when its amin assets are linked to technological
investments
 Digital startups are contributing to and feeding off technological changes
at the same time
 Digitally enabled startups are key vehicles channelling digital technology
Sharing Economy
 Sharing of assets, such as uber or Airbnb, without transferring ownership
 Many of the market leaders in some areas (uber, Airbnb, Facebook,
Alibaba) all have no content / inventory of their own, but rather connect
people through their platform and apps.
 Digital divide – the disparity of those with many phones and devices,
compared to those who don’t even have internet access.
 Success factors of sharing economy – developing a culture of trust, ease
of access to services etc.
  Lock in effect – when a company makes it hard to leave them (e.g.
Gillette with cheaper razor but expensive blades)

22/10/24
Work Automation
Digital Labour platforms
 Location based platforms, where the services are provided by local
individuals in a specific location, and online platforms, where workers
provide their services remotely
 Digital labour platforms:
1. Open new markets for businesses
2. Create new employment and income opportunities
3. Offer flexibility for some workers -> characterised by low entry barriers
 Significantly transformed how work is organised and performed, with
challenges in ensuring that workers on platforms have access to decent
work
 An example of a digital labour platform is uber, where locals provide the
service with help from an online platform and remote workers.
Digital Nomads
People who are location independent with digitally based jobs/roles that allows
them to be remote.
Jobs at risk - Tasks least likely to be automated
1. Perception tasks
2. Manipulation tasks – irregular forms
3. Creative tasks – problem solving, art and design
4. Social tasks – teaching, negotiating, advertising

This differs depending on a country’s development were limited education and

resources can affect the risk of automation.

Gig Economy
The emergence of new forms of contingent employment that:
 Maximises flexibility
 Minimises security for workers
 Transfer risk to worker: variability of demand, downtime, sickness
This is done using on demand work (0-hour contracts), self-employed workers
and digital labour platforms

24/10/24 Workshop Debates

Digital Transformation: all processes have been digitalized and cultural
change to adopt these new ways has been accepted by all members
internally involved

1. Allows people to manage their own time, digital nomads, flexibility of

time.
2. More access to education if everything is put online, allows the world to
become more connected, meaning commuting times are decreased.
3. From this, there are environmental benefits, where commuting times are
reduced. When fully working remotely, research shows that there is a
54% decrease in emissions.
4. Just because teaching is now online doesn’t mean that you can’t access
specialists to help with specific problems.