About Me
Hi! I’m Kamyar, a PhD Candidate and Research Fellow at the University of Siena’s Artificial Intelligence Lab (SAILab). My academic journey and professional work center around the fascinating world of Artificial Intelligence, with a special focus on Natural Language Processing (NLP) and Natural Language Understanding (NLU).
Throughout my career, I have delved into various aspects of AI, particularly excelling in Machine Learning and Neural Networks. My passion lies in advancing the capabilities of language models, especially for low-resource languages. By improving access to AI technologies for these languages, I aim to contribute to a more inclusive and diverse digital world.
Education
University of Siena
Ph.D. Artificial Intelligence
2022 - 2025
The University of Siena, founded in 1240 in Siena, Italy, is one of the world's oldest universities. It offers diverse undergraduate, postgraduate, and doctoral programs in fields like humanities, sciences, medicine, and engineering. Renowned for academic excellence and situated in a historic city, it attracts global students. The university is also notable for its research activities and international partnerships.
I began my Ph.D. in AI at the SAILab (Siena Artificial Intelligence Lab) at the University of Siena. Initially, my research concentrated on Physics-Informed Neural Networks. Subsequently, I focused on crossword generation for educational purposes. I developed a crossword generator based on NLP techniques and large language models (LLMs), supporting multiple languages including English, Italian, Arabic, and Turkish. Currently, my research is dedicated to knowledge graph generation from text. Throughout my Ph.D., I have collaborated with various companies such as Baker Hughes, Expert.ai, and Leonardo S.p.A.
University of Siena
M.Sc. Applied Mathematics
2017 - 2021
The University of Siena, founded in 1240 in Siena, Italy, is one of the world's oldest universities. It offers diverse undergraduate, postgraduate, and doctoral programs in fields like humanities, sciences, medicine, and engineering. Renowned for academic excellence and situated in a historic city, it attracts global students. The university is also notable for its research activities and international partnerships.
During my Master’s degree in Applied Mathematics at the University of Siena, I had the opportunity to delve deeply into mathematical concepts as well as Artificial Intelligence (AI) and Machine Learning (ML) principles, which are closely aligned. I chose the Data Science track, where half of the courses focused on AI-related subjects such as Machine Learning, Big Data, and Artificial Intelligence. Additionally, the curriculum included advanced mathematics courses like Advanced Algebra, Advanced Calculus, Advanced Analysis, and Advanced Statistics, which significantly bolstered my foundational understanding of AI and ML.
For my thesis, I concentrated on Graph Neural Networks, and titled my research “ECG Signal Classification by Topological Feature Extraction and Utilizing Graph Isomorphism Networks.”
University Of Hormozgan
B.Sc. Applied statistics
2012 - 2017
University Of Hormozgan is a public higher education institution located in Bandar Abbas, Iran. Established in 1991, the university offers a range of undergraduate, graduate, and doctoral programs across various fields of study. These fields often include engineering, humanities, basic sciences, marine sciences, and social sciences, among others. The university aims to provide quality education and foster research and development in the region. It typically collaborates with other academic institutions, industry, and government to contribute to the socio-economic development of Hormozgan Province and beyond.
During my bachelor’s degree in Applied Statistics, I had the valuable opportunity to delve deeply into the concepts of statistics and probability, which constitute the foundational elements of many AI and ML algorithms. My thesis focused on “The Effect of Economic Indicators on the First and Second Markets of the Tehran Stock Exchange.”
Experience
University of Siena
Researcher Phd Candidate
Apr 2022 - Present | Siena, Italy
I specialized in creating crossword puzzles for educational use and developed a sophisticated crossword generator utilizing NLP techniques and large language models (LLMs). This generator supports multiple languages, including English, Italian, Arabic, and Turkish. Currently, my research is centered on generating knowledge graphs from text. During my Ph.D. journey, I have collaborated with notable companies such as Baker Hughes, Expert.ai, and Leonardo S.p.A.
University of Siena
Teacher Assistant
Oct 2022 - Present | Siena, Italy
As a Machine Learning course Teaching Assistant at the University of Siena, I had the opportunity to work closely with the course instructor to create a dynamic and engaging learning environment. My responsibilities included guiding students through course material and assisting them with assignments, projects, and lab exercises. I focused on fostering a collaborative learning atmosphere by encouraging student interaction and participation. In addition to providing one-on-one assistance and conducting small group sessions, I helped develop and refine educational materials to enhance the learning experience. Monitoring student progress and offering constructive feedback were key parts of my role, all while contributing to a positive and inclusive classroom atmosphere. This position allowed me to leverage my expertise in machine learning and strengthen my communication and interpersonal skills.
University of Siena
Machine Learning Specialist
Apr 2022 - May 2023 | Siena, Italy
As a machine learning specialist, I applied machine learning and artificial intelligence to real-world scenarios, developing and deploying advanced algorithms to solve complex problems. I focused on creating physically-informed neural networks that adhered to underlying physical laws, bridging the gap between theoretical research and practical application. I collaborated with cross-functional teams to integrate these solutions into existing processes, provided insights for decision-making, and continuously improved model performance through rigorous testing and validation.
Baker Hughes
Data Scientist
Jan 2022 - Mar 2023 | Florance, Italy
During my tenure, I collaborated with Baker Hughes to develop machine learning models aimed at predicting the pollution output generated by gas turbines. We created various shallow and deep learning models to achieve this objective.
State manegement and planing organization
Data Analyst
Jun 2016 - Mar 2017 | Bandarabbas, Iran
As a Demographic and Statistical Analyst, I performed comprehensive demographic and statistical analyses using Excel, SPSS, and R. My role involved gathering and managing datasets from various sources, ensuring their accuracy and relevance, as well as cleaning and preprocessing data for in-depth analysis. Utilizing advanced statistical methods, I conducted descriptive, inferential, and regression analyses to identify trends, patterns, and insights. Additionally, I analyzed population demographics to provide insights into current trends and forecast future changes, ultimately supporting strategic decision-making and research initiatives.
Gilda Research Company
Statistican
Feb 2015 - May 2017 | Bandarabbas, Iran
As a dedicated statistician, I leveraged SPSS, Python, and R to conduct comprehensive statistical analyses for surveys, providing accurate and actionable insights. My expertise included designing, implementing, and interpreting complex survey data with a variety of techniques such as descriptive and inferential statistics, and predictive modeling. I ensured data integrity through meticulous cleaning and validation while creating clear visualizations and reports to communicate findings effectively. Committed to methodological rigor, I continuously enhanced my skills and stayed updated with the latest advancements in statistical methodologies and software.
Publications
Kamyar Zeinalipour, Yusuf Gökberk Keptiğ, Marco Maggini, Leonardo Rigutini, Marco Gori | AIED 2024
Clue-Instruct Text-Based Clue Generation for Educational Crossword Puzzles
https://aclanthology.org/2024.lrec-main.297
Andrea Zugarini, Kamyar Zeinalipour, Surya Sai Kadali, Marco Maggini, Marco Gori, Leonardo Rigutini | LREC-COLING 2024
Building Bridges of Knowledge Innovating Education with Automated Crossword Generation
https://ieeexplore.ieee.org/abstract/document/10459980
Kamyar Zeinalipour, Tommaso Iaquinta, Giovanni Angelini, Leonardo Rigutini, Marco Maggini, Marco Gori | ICMLA 2024
ArabIcros AI-Powered Arabic Crossword Puzzle Generation for Educational Applications
https://aclanthology.org/2023.arabicnlp-1.23
Kamyar Zeinalipour, Mohamed Zaky Saad, Marco Maggini, Marco Gori | Arabic NLP 2023
Giovanni Angelini, Marco Ernandes, Tommaso Iaquinta, Caroline Stehlé, Fanny Simões, Kamyar Zeinalipour, Andrea Zugarini, Marco Gori | EAI INTETAIN 2023
Italian Crossword Generator Enhancing Education through Interactive Word Puzzles
https://arxiv.org/abs/2311.15723
Kamyar Zeinalipour, Asya Zanollo, Giovanni Angelini, Leonardo Rigutini, Marco Maggini, Marco Gori | CLiC-it 2023
Graph Neural Networks for Topological Feature Extraction in ECG Classification
https://link.springer.com/chapter/10.1007/978-981-99-3592-5_2
Kamyar Zeinalipour, Marco Gori | WIRN 2022