Lelio Campanile

Lelio Campanile

Python lovers ❤️, researcher in the making and apprentince sorcerer 😁

Thesis Topics

3 minute read

Available Thesis Topics

Introduction

The thesis is a key milestone in both the Bachelor’s and Master’s programs. It provides students with the opportunity to explore a topic in depth, apply their knowledge to a concrete project, and contribute—albeit modestly—to the advancement of scientific research. Beyond the technical results, what truly matters is the rigor of the methodology, the ability to analyze and synthesize information, and the development of autonomous and critical thinking.

Research Areas

1. Natural Language Processing (NLP) and Large Language Models (LLM): Methodological and Applied Perspectives

This research track focuses on natural language processing, with special attention to the use of large language models. Topics may include:

  • Design and evaluation of NLP pipelines
  • Stylometry and authorship attribution
  • Performance evaluation and fine-tuning of LLMs
  • Applications in education, privacy, misinformation, and more

2. Machine Learning and Deep Learning: Explainable AI and KAN Networks

This area explores both classic and advanced machine learning techniques, with a particular focus on:

  • Interpretable neural networks (XAI)
  • Hybrid architectures such as Kolmogorov–Arnold Networks (KANs)
  • Applications in classification, regression, and image analysis
  • Comparative studies of interpretable vs black-box models

3. Simulation and Modeling of Systems Using Queuing Networks and Discrete Event Simulators

This line of work is suited for students interested in system modeling and performance evaluation. Topics may include:

  • Modeling using queuing theory
  • Discrete-event simulation with SimPy or similar tools
  • Performance metrics: throughput, latency, resource saturation
  • Case studies in smart cities, edge computing, and blockchain architectures

4. Software Engineering Aspects in Data Science

This area is still under development and is aimed at students interested in improving the quality, reproducibility, and efficiency of data-driven projects. Topics may include:

  • Design patterns and software architectures for data science workflows
  • Automation of data pipelines
  • Code testing, documentation, and version control
  • Collaborative development practices and DevOps tools in data analytics

Guidelines and Ethical Recommendations

To ensure that thesis work is serious, ethical, and high-quality, students are expected to follow these essential rules:

  • Academic integrity: All submitted material must be original or properly referenced. Copying text, code, or content without citation is strictly forbidden.
  • Proper use of sources: Every external resource—research paper, website, dataset, third-party code—must be clearly cited both in-text and in the bibliography.
  • Responsible use of generative AI (e.g., ChatGPT, Copilot, Claude): These tools may be used for assistance, but must never replace the student’s own work. Any generated content must be reviewed, understood, verified, and critically reworked before inclusion. Failing to do so may lead to inconsistencies or even plagiarism.
  • Autonomy and responsibility: A thesis is not a passive assignment. It requires planning, regular discussion with the advisor, clear documentation, and the ability to overcome technical and conceptual challenges.
  • Ethics in data and technology: Projects involving personal, biometric, or sensitive data must be handled with particular care, with attention to privacy, transparency, and responsible use of algorithms.

Respecting these principles is an integral part of the final evaluation. The goal is not merely to “complete” the thesis, but to learn how to work with method, rigor, and scientific integrity.