Information Analysis and Processing for Training

Wed, 07 Aug 2024 11:00:00 +0000

Aims

These courses address the analysis and processing of information for sport and training from two complementary perspectives. They share the same broad teaching area, but they are distinct courses offered at different degree levels and organized around different learning goals, tools, and assessment methods.

The Bachelor’s Degree course introduces the foundations needed to work with data in sport contexts. Students learn what data are, how information is represented in digital systems, and how structured datasets can be prepared, analyzed, summarized, and communicated with spreadsheet tools.

The Master’s Degree course develops a more structured and reproducible view of data work. Students learn how to frame data mining problems, design relational databases, implement them with a DBMS, and query them with SQL to obtain reusable datasets.

The two courses follow a coherent progression from understanding and representing data to organizing, querying, and reusing data in controlled workflows. Their shared goal is to help students transform raw data into meaningful information that can support interpretation and decision making in sport and training contexts.

Courses

Bachelor’s Degree Course: foundational data concepts, digital representation of information, and spreadsheet based data analysis with Microsoft Excel
Master’s Degree Course: data mining foundations, relational database design, database implementation, and SQL querying

Teacher

Michael Soprano - Course Instructor

Bachelor’s Degree Course

The Bachelor’s Degree course introduces the basic vocabulary and tools needed to work with data in sport and training contexts. Students first learn how to describe data, variables, and measurements. They then study how different kinds of information are represented digitally. The final part of the course focuses on Microsoft Excel as a tool for preparing, analyzing, visualizing, and summarizing structured data.

Topics Covered

Module 1 - Introduction to Data Science in Sport

Data, information, knowledge, and the DIKW pyramid in sport and training contexts
Structured and unstructured data from sport-related sources
Sensors, SportsTech, and examples of data-driven decision making in sport
From raw data to information through preparation, analysis, visualization, and communication
Tables, observations, variables, and values
Populations, samples, units of analysis, and levels of measurement
Qualitative and quantitative variables
Sport-related data types, including biometric, position and movement, subjective, and performance data

Module 2 - Representation and Management of Data

Basic computer architecture: memory, storage, peripherals, CPU, and operating system
Data storage in the filesystem
Files, folders, metadata, hierarchical organization, and paths
Bits, bytes, binary code, and digital encodings
Text: character sets, encodings, ASCII, Unicode, and UTF-8
Images: raster graphics, vector graphics, color representation, and main file formats
Sound: wave characteristics, sampling, quantization, channels, MIDI messages, and audio formats
Video: sequences of images, movement, frame rate, resolution, and video formats
The Shannon-Weaver model of communication
Entropy and redundancy of information
Why data compression is useful
Lossless and lossy compression

Module 3 - Data Analysis with Microsoft Excel

Elements of the Excel interface, workbooks, and worksheets
Managing cells, rows, columns, and ranges
Entering, editing, deleting, and formatting data
Applying numeric, date, time, text, and currency formats
Working with contiguous, non-contiguous, and multi-sheet ranges
Managing rows and columns and navigating large worksheets
Relative, absolute, and mixed references
Introduction to formulas and functions
Formula syntax: references, operators, precedence, and common errors
Function syntax: arguments, result types, nesting, and basic summary functions
Practical exercise on Boston Marathon 2025 data, including data cleaning, derived columns, summary statistics, and a mini dashboard
Importing structured data from text files, including delimiters, encodings, and common import problems
Introduction to Power Query for importing and preparing data
Working with Excel tables
Creating, managing, and formatting charts
Examples with line, scatter, and area charts
Basic measures of correlation
Introduction to PivotTables and PivotCharts
Using PivotTables to summarize and explore Fitbit data

Learning Approach

The Bachelor’s Degree course is organized around foundational concepts and applied spreadsheet activities. Students first acquire the vocabulary needed to describe datasets, then study how different types of information are represented digitally, and finally practice data analysis in Excel through guided examples and exercises.

Assessment

The Bachelor’s Degree course is assessed through a written exam with multiple-choice questions.

Reading Material

Peter O’Donoghue, Lucy Holmes, Data Analysis in Sport. Routledge Studies in Sports Performance Analysis, First edition, 2014
Michael Alexander, Dick Kusleika, Excel 365 Bible. Wiley, First edition, 2022

Master’s Degree Course

The Master’s Degree course moves toward structured, reproducible, and technically controlled data workflows. The course connects data mining, relational modeling, database implementation, and SQL querying.

Topics Covered

Preparatory Recall - Data Science Foundations

Data, information, and knowledge in reproducible data workflows
Sport-related data types, including biometric, movement, subjective, and performance data
Tables as representations of observations, variables, and values
Units of analysis, variables, scales of measurement, and metadata
Data quality, coherent types, identifiers, and first normal form
Privacy and protection of personal and health-related data

Module 1 - Introduction to Data Mining

Data science as a reproducible process from data to decision
Project workflow: understanding, preparation, exploration, modeling, interpretation, and deployment
Populations, samples, representativeness, and sampling error
Variables, features, targets, and units of analysis
Problem formulation in sport-related scenarios
Typical data mining tasks: classification, regression, clustering, association, anomaly detection, and time series analysis
Learning paradigms: supervised, unsupervised, and other basic settings
Baselines, training and test data, validation strategies, and evaluation on unseen data
Metrics for model evaluation and comparison
Methodological issues: missing values, outliers, temporal consistency, leakage, imbalance, and data quality
Interpretability, reproducibility, and documentation of modeling choices

Module 2 - Relational Databases

Databases as persistent, coherent, and shared collections of data
DBMSs and the role of persistence, scale, globality, reliability, efficiency, and privacy
Conceptual, logical, and physical levels of database design
Entity-Relationship modeling: entities, attributes, relationships, cardinalities, identifiers, and constraints
Binary, recursive, and n-ary relationships
Hierarchies and alternative modeling choices in ER schemas
Relational model: relations, tables, tuples, attributes, domains, schemas, instances, and NULL values
Relational constraints: domains, primary keys, uniqueness, and referential integrity
Translation from ER schemas to relational schemas
Many-to-many, one-to-many, one-to-one, and recursive relationships in the relational model
Redundancy, update anomalies, insertion anomalies, and deletion anomalies
Normalization up to 3NF as an operational method for reducing uncontrolled redundancy
Practical case study on road cycling races, from requirements to ER schema and relational schema
SQLite as an embedded relational DBMS based on a single database file
DBeaver as a graphical client for exploring schemas, tables, data, and relationships
SQL scripts and controlled population of a database
DDL for schema definition: CREATE TABLE, DROP TABLE, ALTER TABLE, types, keys, and constraints
DML for data manipulation: INSERT, UPDATE, and DELETE
SQL queries for data retrieval: SELECT, FROM, WHERE, ORDER BY, LIMIT, AS, and calculated columns
Aggregations and grouping with COUNT, MIN, MAX, AVG, SUM, GROUP BY, and HAVING
Joining tables to integrate information, including JOIN and LEFT JOIN
Exporting query results as reusable datasets for subsequent analysis and reporting

Learning Approach

The Master’s Degree course emphasizes structured and reproducible workflows. Students connect data science concepts to relational database design, SQL extraction, and reusable data preparation. The course combines conceptual modeling, database implementation, query writing, and the interpretation of database outputs as datasets for further analysis.

Assessment

The Master’s Degree course is assessed through a written exam with open-ended questions.

Reading Material

Ian H. Witten, Eibe Frank, Mark A. Hall, Christopher J. Pal, James R. Foulds, Data Mining: Practical Machine Learning Tools and Techniques. Morgan Kaufmann, Fourth edition, 2016
Giorgio M. Di Nunzio, Emanuele Di Buccio, Basi di dati. Progettazione concettuale, logica e SQL. Esculapio, First edition, 2017

Data Mining | Michael Soprano

Information Analysis and Processing for Training

Aims

Courses

Quick Access

Teacher

Bachelor’s Degree Course

Topics Covered

Module 1 - Introduction to Data Science in Sport

Module 2 - Representation and Management of Data

Module 3 - Data Analysis with Microsoft Excel

Learning Approach

Assessment

Reading Material

Master’s Degree Course

Topics Covered

Preparatory Recall - Data Science Foundations

Module 1 - Introduction to Data Mining

Module 2 - Relational Databases

Learning Approach

Assessment

Reading Material