|
Docente
|
BASILI ROBERTO
(programma)
Richiami ai metodi base di ML. Metodi Supervised vs. Unsupervised. Paradigmi di ML, Metodi di addestramento. Machine Learning Metrics and Evaluation.
Introduzione alla modellazione dei documenti per il Web: dall’Information Retrieval al Natural Language Processing.
Modelli di Linguaggio. Processi Markoviani. Modelli Generativi: HMM. Use Case: Probabilstic POS tagging
PAC Learnability. Perceptron
SVM. Hard Margin. Soft margin SVM.
Kernel polinomiali. Sequence Kernels. Kernel for NLP. Tree Kernels. Semantic Tree kernels
Deep Learning. Intro e Background. NNs: tasks and Training. Convolutional Neural Networks. Recurrent Neural Networks. Attention. Trasformers. Deep Learning Software Development. NN in Python.
Language Modeling con modelli neurali. Temi avanzati: attention; encoding-decoding; adversarial NNs; transformers.
Web and Lexical Semantics: the role of lexical learning in Web scenarios.
Opinion Mining e Sentiment Analysis: il taks, le risorse e le metodologie principali
Advanced Statistical NLP for QA (from NERC & SRL to QA)
Advanced Machine Learning for the Web: Learning to Rank, Recommending systems
 • Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze, Introduction to Information Retrieval, Cambridge University Press. 2008. Consultabile on-line
• Roberto Basili, Alessandro Moschitti, Text Categorization: from Information Retrieval to Support Vector Learning, ARACNE Editore, 2005.
• Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT press, 2016.
• Note del docente e articoli scientifici distribuite durante il corso.
|
