MSc Data Science: Revision Guide

Advanced Analytics, Architecture, and Governance

1. Analyse the significance of data cleaning and transformation in analytics readiness

Analytics readiness is the foundation of high-quality decision-making. Without these steps, models suffer from "Garbage In, Garbage Out" (GIGO).

2. Discuss the role of data governance in supporting organizational compliance

Data Governance (DG) provides the legal and operational "rulebook" for handling information assets.

3. Evaluate privacy preserving techniques for sensitive data analytics
Technique Concept Critical Evaluation
Differential Privacy Adding mathematical "noise" to data. High security, but can make small datasets less accurate.
Anonymization Removing Names/IDs. Fast and easy, but vulnerable to "re-identification" by clever hackers.
Homomorphic Encryption Processing encrypted data. Highest security; extremely slow and computationally expensive.
4. Analyse the role of analytics engineering in modern enterprises, distinguish it from data science and software engineering and evaluate its impact on organizational decision making and operational efficiency

Analytics Engineering is the bridge between raw data collection and final analysis.

5. Compare monolithic, microservices and event driven architectures for analytics applications. Evaluate their suitability for large scale real time financial analytics platform
6. Analyse the performance optimization techniques used in analytics application

To ensure speed in big-data environments, developers use:

7. An e-commerce company collects customer browsing behaviour, purchase history and geolocation data to train a recommendation model. Analyse whether the system complies with data privacy regulations and identify compliance gaps

Potential compliance failures include:

8. Evaluate how the general data protection regulation and the local data protection regulation affect data analytics design

Modern design must be "Privacy by Design":