Francesco Sannicola
Machine Learning | Software Engineering
Resume
Resume
Email
Email
LinkedIn
LinkedIn
GitHub
GitHub
☀
☾
Home
Projects
Articles
About
Projects
Document-Claim Evidence Mapper
A document-claim analysis tool that uses LLMs to extract and rank relevant text supporting or refuting claims. Combines prompt-based NLP with customizable evaluation for fact-checking and analysis in legal and financial contexts.
ML-based classification of customer complaint drivers
A data-driven approach to understand and predict customer dissatisfaction through data processing, exploratory analysis, predictive modeling, rigorous evaluation, and actionable visualization.
LLM-Powered Email Generation and Multi-Class Classification
Building and evaluating a machine learning pipeline for email classification utilizing synthetic data generated by a GPT-based email generator with Few-Shot Learning.
Blind Binary Classification: ML Model Development & Evaluation
Development, assessment, and evaluation of machine learning models for a binary classification task. However, the dataset, although divided into two parts, completely lacks any field descriptions.
Conquering Coding Challenges: My LeetCode Journey (WIP)
LeetCode problems solved in Python. Each solution is well-tested and easy to execute.
Links Academy - Anomaly detection
Anomaly detection from statistical baseline to density-based and distance-based methods.
Links Academy - Time-series analysis
Time series component analysis, forecasting utilizing AutoRegressive models, and time series anomaly detection employing LSTM.
Greek and Latin Part of Speech Tagging
PoS tagging for Greek and Latin texts using dynamic programming and Viterbi algorithm.
Natural Language Technologies: Leveraging Ontologies and Lexical Resources
Four tutorials showing the application of linguistic resources and lexical ontologies such as WordNet, BabelNet and NASARI. It includes methods for measuring word similarity, mapping between Framenet and WordNet, text summarization, and evaluating semantic similarity in Italian.
Natural Language Technologies: Exploring Definition Consistency, Hanks' theory and Automatic Summarization
The exercises explore the consistency of student personal definitions, the analysis of definitions in WordNet, the characterization of verbs according to Hanks' theory, the experimentation with content-to-form using genus and differentia, and the creation of automatic summaries using vector representations of Babelnet synsets.
Document-Claim Evidence Mapper
A document-claim analysis tool that uses LLMs to extract and rank
relevant text supporting or refuting claims. Combines prompt-based
NLP with customizable evaluation for fact-checking and analysis in
legal and financial contexts. 
ML-based classification of customer complaint drivers
A data-driven approach to understand and predict customer
dissatisfaction through data processing, exploratory analysis,
predictive modeling, rigorous evaluation, and actionable
visualization.
LLM-Powered Email Generation and Multi-Class
Classification
Building and evaluating a machine learning pipeline for email
classification utilizing synthetic data generated by a GPT-based
email generator with Few-Shot Learning.
Blind Binary Classification: ML Model Development &
Evaluation
Development, assessment, and evaluation of machine learning models
for a binary classification task. However, the dataset, although
divided into two parts, completely lacks any field descriptions.
Conquering Coding Challenges: My LeetCode Journey (WIP)
LeetCode problems solved in Python. Each solution is well-tested and
easy to execute.
Links Academy - Anomaly detection
Anomaly detection from statistical baseline to density-based and
distance-based methods.
Links Academy - Time-series analysis
Time series component analysis, forecasting utilizing AutoRegressive
models, and time series anomaly detection employing LSTM.
Greek and Latin Part of Speech Tagging
PoS tagging for Greek and Latin texts using dynamic programming and
Viterbi algorithm.
Natural Language Technologies: Leveraging Ontologies and Lexical
Resources
Four tutorials showing the application of linguistic resources and
lexical ontologies such as WordNet, BabelNet and NASARI. It includes
methods for measuring word similarity, mapping between Framenet and
WordNet, text summarization, and evaluating semantic similarity in
Italian.
Natural Language Technologies: Exploring Definition Consistency,
Hanks' theory and Automatic Summarization
The exercises explore the consistency of student personal
definitions, the analysis of definitions in WordNet, the
characterization of verbs according to Hanks' theory, the
experimentation with content-to-form using genus and differentia,
and the creation of automatic summaries using vector representations
of Babelnet synsets.