99087 - Data Analysis for Media (1) (LM)

Learning outcomes

The course aims to provide students with a clear understanding of the role of data in the contemporary media landscape. Students will learn to use data-driven methodologies to analyze media products, their production and circulation, and the social discourses related to them. An in-depth theoretical part will be dedicated to the way in which the enormous availability of data, in the digital context, has transformed both the media supply chain and the methodologies of evaluation, prediction and analysis. The student will then be provided with specific skills related to the extraction, transformation and loading of data and modeling tools in the media environment. No previous knowledge of programming languages or specific software is required. At the end of the course the student will have acquired skills related to: - data extraction both from dedicated databases and from social media and web pages; - data analysis, evaluation of data quality, cleaning and construction of datasets; - qualitative and quantitative modeling; - implementation of simple scripts for data manipulation; - introduction to the use of databases; - basic elements of data visualization

Course contents

The course is structured in three distinct parts, each of which covers the essential aspects of data-driven media analysis. Students will explore the theoretical foundations, practical applications, and basic methodologies in the field of data analysis.

Part 1: Theoretical Foundations (3 lectures)

This section begins with an exploration of the importance of data analysis in the contemporary media landscape. Students will examine the fundamental aspects of data analysis, including types of data, methods of access, and ethical considerations relevant to the media industry. Practical skills in data cleaning, preprocessing, and exploratory data analysis will be emphasized, enabling students to draw valuable insights from audiovisual datasets and effectively communicate findings through data visualization.

Part 2: Talking to your data with LLMs (3 lectures)

Students will be introduced to machine learning and deep learning, with a specific focus on Large Language Models (LLMs) and Multimodal Models. Hands-on experience with simple scripts in Python, low-code tools, and cloud-based analysis platforms will enrich their understanding of LLM-based data interactions (ChatGPT and others).

Part 3: Application Projects (4 lectures)

The final segment revolves around hands-on application projects, in which students will apply their acquired knowledge and skills to real-world challenges in the audiovisual industry. Students collaborate on defining project objectives, collecting and pre-processing data, using traditional and AI-supported analysis techniques, and finally presenting their projects.

Ultimately, the course aims to provide students with the skills needed to effectively manage data analysis in the audiovisual industry.

A course plan with content for each lesson will be available in Virtuale.

Readings/Bibliography

The course is geared toward direct interaction with students. Therefore, study and practice materials will be provided in class and released through the Virtuale platform. For each lecture, the materials needed for exam preparation and any optional further study will be indicated.

Students who wish to delve systematically into the course topics can refer to Joel Grus' text, Data Science from Scratch: First Principles With Python, O'Reilly, 2019. The study of this optional text is not necessary to take the exam.

Teaching methods

The course involves constant interaction with students, both in the discussion of theoretical topics and in the actual implementation of application projects and exercises. Students are advised to use their own computer to follow lectures and be able to continue the exercises independently.

Assessment methods

For the exam, the student will be required to carry out, either alone or in a group, a Data Analysis project agreed upon with the professor. The project will be presented at the end of the course (depending on the size of the class) or in the exam. The exam will involve an oral interview devoted both to the analysis of the project and to the verification of knowledge of the course topics.

Teaching tools

Lectures will include Prezi presentations accessible to students, while cloud-based collaborative notebooks (Jupyter Notebook) will be used for coding, tutorials and project development. Additional materials will be provided through the Virtual platform.

Office hours

See the website of Guglielmo Pescatore