Module details for Introduction to Data Science

Description

*

Aims

The aim of the module is to introduce students to data science and the typical life cycle of data. The relevant theoretical foundations of data science are highlighted, and the student is equipped with the fundamental understanding and knowledge of tools underpinning data science in light of rapid technological advancement.

Learning Outcomes

By the end of this module the student should be able to:

1.  Understand all stages of the data science life cycle: Capture, Maintain, Process, Analyze & Communicate.

2.  Use data science and associated APIs (Application Programming Interfaces) to solve a clearly-framed and well-posed data science problem.

3.  Appreciate real-world problems and challenges involving the use of data in industry, science, and the public sector.

Indicative Content

1 Introduction

Overview of the data science life cycle

2 Intro to Scripting

Use of Python through the data science lifecycle

3 Data Extraction

Database querying, including essential SQL and noSQL

4 Data Cleansing

Regular expressions, find and replace, vocabulary construction and harmonization, outlier detection and missing data imputation, anonymizing data, with emphasis on ethics and legal compliance with regulations such as GDPR

5 Feature Engineering

Scaling, translation, bucketization, embeddings, hash values, feature crosses

6 Overview of Models

High-level overview of the breadth of ML/AI models available

7 Basic Model Assessment and Selection

Accuracy, Confusion Matrices, Receiver Operating Characteristic (ROC) and Precision-Recall Curves

8 Serving Models

Exporting models, containerization, container orchestration, high-performance remote procedure call (RPC) frameworks

9 Visual Analytics and Data Visualisation

History and goals of visual analytics. Types of data and encodings. Data processing and clustering. Information visualisation techniques. Analytics process and pipeline

Teaching and Learning Work Loads