Profile Picture

Sonia Nicoletti

Hello! I'm Sonia, a computer science student at the University of Bologna with a passion for innovative technology and problem-solving. In 2022, I graduated with honors in Computer Science for Management, where I developed a strong foundation in both technical and analytical skills. Now, I'm pursuing advanced studies to deepen my expertise in different areas of computer science.

Research Interests

Natural Language Processing

AI Ethics

Software Engineering Practices

Computing Education

News

Projects

Project BI

Customer segmentation analysis for AI-Driven Healthcare Applications

Business Intelligence

Clustering potential users of AI-driven healthcare applications through survey analysis.

Customer segmentation analysis for AI-Driven Healthcare Applications

July 2024

SAS

Python

This project examines customer segmentation for AICare4You, an envisioned AI-powered medical chatbot designed to provide health advice and medication prescriptions based on user data.
By analyzing survey responses from potential users, the study identifies key preferences for chatbot functionalities, such as a focus on general wellness or diagnosis-oriented features. Using Principal Component Analysis (PCA) and hierarchical clustering techniques, five distinct customer segments were identified, ranging from cost-conscious skeptics to fitness enthusiasts and AI enthusiasts who fully embrace the technology. These insights highlight the diverse expectations and challenges in designing AI-driven healthcare solutions.

Project Social Media and Social Network Analysis

Exploring Public Discourse on AI Risks

Social Media Analysis

Analyzing public discourse on AI risks through sentiment and topic modeling.

Exploring Public Discourse on AI Risks

January 2024

R

Python

This social media research explores the public discourse surrounding the risks and opportunities associated with artificial intelligence. I collected and analyzed YouTube comments from AI-related videos published between 2020 and 2023.
The methodology involved sentiment analysis, topic modeling, and dictionary-based text analysis.

GitHub Repo

Syntax Analysis Tutorial

Compilers

LL(1) grammar parsing tutorial with interactive exercises.

Syntax Analysis Tutorial

Syntax Analysis Tutorial

July 2023

Mathematica

This Mathematica tutorial contains a generator of exercises on parsing LL(1) grammars. The objective is to explain to the user the procedure for solving this type of exercise. The exercise consists of generating the Nullable, First and Follow sets and filling in the parsing table.
The primary areas of our focus revolved around the management of nested lists, comprehension of function scope, and the creation of user interfaces. Moreover, we applied our expertise in compiler theory to assess and streamline the core aspects of this subject matter.

GitHub Repo

Tailored Travel Insurance Solutions

Data Mining

Predictive model for recommending travel insurance.

Tailored Travel Insurance Solutions

December 2022

R

WEKA

The objective of this project was to build a predictive model that could suggest the most fitting travel insurance package for a particular customer. The tools used were WEKA and R.

Cluedo Mobile

Mobile Apps

Cluedo-inspired Android app using Clean Architecture.

Cluedo Mobile

June 2022

Kotlin

Firebase

This Android app is inspired by the famous mystery game Cluedo. Friends can play online together by joining the same game. Once the host has started the game each player has to guess Where, by Who and How the murder was committed making accusations and replying to others accusations.
The front-end side has been realised using Kotlin and the Clean Architecture paradigm, in particular MVVM pattern. The back-end and the multiplayer functionality have been realised through Firebase.

GitHub Repo

Animal House

Web Development

A website that integrates APIs for fun facts, videos, and interactive services.

Animal House

May 2022

HTML

CSS

Bootstrap

JavaScript

Vue.js

Express.js

Animal House is website for a shelter where users can look for the pets, book appointments with them, register for events or look for volunteering positions. They can also read fun facts about animals, watch popular videos, play games, like Hangman or Meme Generator, or post stories/pictures on their dashboard and share them with the community. Admins can add or remove pets, events and volunteering positions.
The main focus of this project was, other than building a modern and responsive website, to provide services that were built using different APIs (like the fun facts, the top 10 videos, etc). The front-end was made using HTML, CSS (Bootstrap) and JavaScript (Vue), while the back-end was made using Express.

GitHub Repo

ConfVirtual

Databases

A project that uses relational and non-relational databases to manage virtual conferences.

ConfVirtual

April 2022

HTML

CSS

Bootstrap

JavaScript

PHP

MySQL

MongoDB

ConfVirtual is an online platform used to advertise virtual conferences. Users can register for conferences they like and attend their presentations, hosted by presenters or speakers. Once the presentation has started all registered users can join a chat. The admins can create new conferences, evaluate them and add sponsors to the conferences.
The front-end has been realised using mainly HTML and CSS (JavaScript has been used just for the chat functionality). The back-end was made with PHP, MySQL as the main database and MongoDB to save the logs.

GitHub Repo

Unitech

Software Engineering

University portal developed using Agile methodology with Scrum framework.

Unitech

March 2022

Java

Unitech is a university website where users can find information about the institution and from which professors, students and admins can access their own portal. Professors can create new classes, add exams and give grades while students can register for classes and view their personal information.
The website has been realised entirely in Java, using the GWT framework, and the database was made using MapDB. The Agile methodology, specifically Scrum, was used by the team during the entire project.

GitHub Repo

Breast Cancer Prediction with SVM

Statistics

SVM model predicts breast cancer diagnosis using dataset feature analysis.

Breast Cancer Prediction with SVM

June 2021

R

This research uses a breast cancer dataset from the University of Wisconsin Madison containing information on 569 cases. After cleaning and splitting the dataset, SVM models are trained to predict tumor diagnosis. Cell nucleus characteristics are analyzed via visualizations, and the model training is followed by the performance evaluations and hyperparameter optimization. It is observed that some characteristics, such as radius, perimeter, area and texture, are strongly correlated with tumor diagnosis. Finally, the model predictions are interpreted and the most significant features are selected.

Multiplayer Tetris

Multiplayer Tetris

Java Programming

An online multiplayer version of Tetris.

Multiplayer Tetris

Multiplayer Tetris

June 2021

Java

This game is a spin-off of the original Tetris game, which I'm sure you've heard of. The main difference is that this is a multiplayer version, where each player competes against others to survive longer in the game. Additionally, every time a player completes a row, they can send a "trash row" to another player of their choice to increase the game's difficulty.
This game was developed in Java, with the user interface generated using the Swing framework. For the client-server connection, we utilized the ServerSocket and Socket classes from the java.net package.

Curriculum Vitae

Download my CV

Contact

Contact me by reordering the following strings:

"sonia.nicoletti", "protonmail.com", "@"