Learning SessionTraining

DATA ANALYSIS IN THE ERA OF LARGE LANGUAGE MODELS WITHIN FOSSR

Large language models (LLMs) are best known for their ability to generate written text and other content in human-like ways. But the usefulness of these AI algorithms extends far beyond. Large language models are used in numerous data science applications. Their ability to process and interpret vast amounts of text data have made them an indispensable part of many data science workflows.

The second FOSSR learning session, which will be held in a hybrid format, in presence (preferable) – in Rome, at the CNR-ISTC institute – and online, promises an in-depth exploration of how LLMs are transforming data analysis.

The Learning Session on Data Analysis in the era of Large Language Models within FOSSR is based on a two-day programme, scheduled for December 10-11, 2024.

For application: https://l.cnr.it/fossr-learning-session-registration-form

For reasons related to the course format and group management, participation in the learning session will be limited to a maximum of 25 participants. If the number of registrations exceeds this limit, a selection will be made based on the educational/professional background related to the course topics, as well as the motivations expressed in the registration form.

On Day 1, between 2:00 PM and 6:00 PM, participants will engage in a series of activities, which take an approach oriented to participatory learning. The session will start with a welcome and an introduction explaining the goals and format of the two-day learning event. The lecturer will introduce important keywords relevant to data analysis and large language models (LLMs). These keywords will lay the groundwork for understanding core LLM-based data analysis concepts with benefits and potential issues. The participants will break into groups to brainstorm ideas, define concepts, and discuss challenges related to the assigned keywords on LLM-based data analysis. This exercise will foster collaboration and a deeper understanding of the terminology. Hence, each group will present their findings, sharing insights on the meaning, application, and implications of the assigned keywords in the context of data analysis with LLMs. Then, the lecturer will provide theoretical insights into key principles of data analysis, such as prompt engineering, model interpretability, and ethical considerations, drawing connections to the ideas shared in group discussions. The day will conclude with a Q&A session.

On Day 2, starting at 10:00 AM and lasting until 1:40 PM, the lecturer will introduce an exercise on data analysis that asks participants to interact with an LLM to find a solution. Accordingly, the participants will work in groups to find their own solution to the assigned exercise, using an LLM-based tools to analyse a dataset relevant to FOSSR. Finally, we will conclude with a discussion about how LLMs are transforming data analysis with a focus on FOSSR settings, and the potential they bring to various research areas . Emphasis will be placed on the importance of responsible and informed use of LLMs by social scientists.

The objective of the learning session is to provide participants with a comprehensive understanding of how LLMs can be applied to data analysis. Through a mix of theoretical insights and group exercises, participants will gain both conceptual knowledge and hands-on experience, fostering a participatory and engaging learning environment.

Download the programme

FOSSR locandina learning session Nuzzolese_fronte

Data di inizio
10 December 2024 00:00

Data di fine
11 December 2024 00:00

Audience
Researchers, Data Scientists, Software Developers, Research Data Managers