MSc Data Analytics Overview 2024-2025

Data Management Overview

• The course is MSc Data analytics for business, 2024-2025, Bordeaux, offered by KEDGE Business School.
• Instructor is Dr. Milad Poursoltan.
• The course content includes introduction to data and data management, data modeling & relational databases, Structured Query Language (SQL), MariaDB and MySQL software, emerging techniques and methods.

Assessment

• 8 sessions.
• Written exam - 50% (individual).
• Practical exam - 40% (group of two).
• Presentation - 10% (group of three).

Class Activities

• Students will collect data about classmates in 30 minutes, and are not allowed to share the data with each other. (first name, field of study, fav job in data science and knowledge of database/data management).
• Students will form groups of four and discuss solutions for challenges in data collection/quality (data collection protocols, defining data quality standards, etc.). Time limit 20 minutes.
• Students will undertake data analysis activities over a 10-minute period.
• Course participants will collect info about classmates (first name, field of study, fav job in data science and knowledge of database/data management), in 30 minutes.

Data and Data Management

• Data represents facts about the world, but these facts are not always straightforward.
• Data can be considered 'raw material' for information, while information can be seen as data in a defined context.
• Information can be seen as a synonym for data or a synonym for facts, and as something new or as something that changes beliefs, knowledge or expectations.
• Wisdom is knowledge put into action.

Data Architecture

• Introduced as the "fundamental organization of a system".
• Enterprise Architecture: visual blueprint of the organization, showing key data/process/applications/technology interrelationships.
• Data Architecture: identification of enterprise data needs, and design/maintenance of master blueprints to meet these needs.

Data Life Cycle

• The process involves collection, processing, storage and securing, using and sharing, and archiving, reusing and destroying data.

Data Management

• Data Management: the process of collecting, organizing, and accessing data to boost productivity/efficiency/decision making.

Data Management Challenges

• Data differs from other assets.
• Data valuation remains a challenge.
• Maintaining data quality is a critical concern.
• Ethical considerations in data handling are key.

Data Management Frameworks

• DMBOK (Data Management Body of Knowledge): Strategic alignment model.
• DCAM (Data Management Capability Assessment Model).
• The Data Management Association (DAMA).

Data Management Framework

• Components of a framework include data architecture, data modeling & design, data quality, data storage and operations, data security, data integration and interoperability, document & content management, metadata, data warehousing & business intelligence, and reference & master data.

Data Governance

• Data Governance is the authority/control framework over data assets.
• The goal is having data managed properly based on best practices and policies (Ladley, 2012).
• Data governance (DG) activities include controlling, and monitoring activities to ensure that data is managed properly.

Data Management Functions and Initiatives

• Policies
• Roles and Responsibilities
• Controls
• Guidelines
• Decision Rights
• Metrics
• Processes
• Rules
• Accountabilities
• Standards
• Issue Management

Data Architecture: Data Lineage

• Demonstrated through examples of tabular relationships linking major entities such as Product, Product Part, Manufacturing Plant. These relationships show what data is used to create a specific data element.

Data Modeling and Design

• A data model is a representation of data requirements.
• Some common data modeling schemes are Relational, Dimensional, Object-Oriented, Fact-Based, Time-Based, and NoSQL schemes.

Data Storage and Operations

• Database: Collection of stored data of varied structure/content. Some large databases relate to instances and schemas.
• Database Architecture Types: Centralized, and Distributed (not federated).
• Basic requirements for distributed architecture (CAP theory): Consistency, Availability, Partition Tolerance.
• Scaling techniques: Horizontal, Vertical.

Data Security

• Data security policies & procedures will provide proper "authentication, authorization, access and auditing of assets". Key concerns include stakeholder concerns, government regulations, security aspects, and ensuring necessary business access.

Data Integration and Interoperability

• Data Integration and Interoperability describes processes related to movement & consolidation of data within & between data stores. Common technologies include ETL and ELT processes.

Document and Content Management

• Document & Content Management controls capture, storage, access, and use of data outside of established (relational) databases.

Reference and Master Data

• Reference data characterizes/classifies other data relevant to an organization.
• Master Data is data about entities that create context for business transactions, such as employees, customers, products, locations, and financial information.

Data Warehousing & Business Intelligence

• Data Warehouse stores raw data using ETL, and/or ELT processes for analytical processing.
• Data Warehousing and Business Intelligence encompass all activities that enable decision making from data, using BI techniques, such as reporting & analysis. The goal is to enable knowledge workers to gain value.

Metadata Management

• Metadata is descriptive 'data about data'. Business intelligence, metadata has multiple kinds and should be managed as 'data'. Metadata is categorized into descriptive, structural and administrative types.

Data Quality

• Data quality refers to the degree to which data meets requirements, and is managed through the data lifecycle.
• Six key dimensions of data quality: Completeness, Uniqueness, Timeliness, Validity, Accuracy, and Consistency.

Group Exercises

• Group exercise N.2 focuses on data governance concepts and on the ability to analyze related problems in various scenarios. Students should consider interaction between the knowledge areas.
• Objectives: practical understanding of data governance concepts and the ability to analyze related problems in various scenarios.

Knowledge Areas

• Knowledge areas involved include Data Governance, Data Architecture, Data Modeling & Design, Data Storage and Operations, Data Security, Data Integration and Interoperability, Document and Content Management, Reference and Master Data, Data Warehousing and Business Intelligence, and Metadata Management and Data Quality.