PS 1st Week

• what are the current problems? (Input outside)

  • f.e. fit the bottles with square bottle necks

  • screw itself is a bit wobbly -> cork distruction

  • too big too heavy

• What are interbal expectations? (external input?)

  • conform to our corporate branding standards

• generarting

• collecting

• defining

• processing

• governing

• extractring insight

• getting rid

Definition:

Artificially generated data that mimics real-world data, Created via algorithms or simulations to preserve statistical properties of real data while protecting privacy.

How is it built? -> Generation Methods

• Rule-based simulation

• Statistical modeling: based on prob distribution

• Generative models (AI)

• Hybrid approaches

Why use it?

• Privacy protection

• data augmentation ->

• testing & prototyping -> test sytems safely

• Innovation sandbox -> no legal/ reg risk

• Even when the reasons for collecting data are transparent, the methods used to gather it may be unethical

• ”…good intentions are not enough to

  make data collection ethical…”

-> example French bank, sexual harassment detevtion via mails

Example target:

• assigns each shopper a unique Guest ID

• adding data -> historicaö buying data

• find patterns -> unsecnted lotion in second trimester

• could estimate which phase and which due date

Data science is the field of study that combines domain

expertise, programming skills, and knowledge of math and statistics to extract meaningful insights from data.

• multi-disciplinary field

• extract knowledga and insights from data in various forms -> translate into business value

• understand and analyze actual phenomena

• helping indivduals and organizations to make better decisions

• apply ML Alogorithms text, video, audio images

Descriptive analytics explain past events ("what happened?"), predictive analytics forecast future events ("what might happen?"), and prescriptive analytics recommend the best actions to take ("what should be done?")

BDDM RV PD IM

1. Business Understanding

2. Data Acquisition

3. Data Curation

4. Modelling

5. Results & Evaluation

6. Visualization & Interface

7. Pilot & Fine-tune

8. Deployment & Roll-out

9. Insights & Decision-making

10. Maintain & Improve

GDPR:

• describe the nature, scope, context and purposes of the processing

• assess necessity, proportionality and compl. measures

• identify assess risks to individuals

• and indetify any additional measures ti mitigate those risks

• Project scoping worksheet

• Project Charter

Differences:

• The Scoping Worksheet is like a diagnostic form doctors fill out to understand a patient’s symptoms.

• The Project Charter is like the official treatment plan signed by doctor, patient, and insurance to proceed with the cure.

• A document (word/ ppt) that contains detailed information on what is the project about and how it will be

developed

• It should be developed with information provided by the client and desk research of the team

• Last version should include client’s feedback

Possible sections:

• Team Introduction (Roles and Responsibilities)

• Problem definition and background

• Project Purpose

• Stakeholders’ Introduction (Roles and Responsibilities)

• Actions to be informed

• Assumptions, Risks and Constraints

• Project Deliverables and Milestones

• Deliverable Review and Acceptance

• Timeline

• Communication and Meetings

• Partner Data & Tech Requirements

• Ethics, Privacy and Data Protection

• KPIs and Expected Impact

• Solution Mock-up

• Budget

• Authorization

• internal sources: ERP, Int. Docs, Website logs

• External Data sources: soc. Media, official statistics, open data platforms, private data

• understanding the data -> data story

• cleaning (Missing values)

• exploring -> Descriptive statistics (Correlation)

• preparing