Teaching
 


Teaching by year



Year 2012-2013

Year 2011-2012

Year 2010-2011

Year 2009-2010

Year 2008-2009

Year 2007-2008



Home

Department of Computer Science and Information Engineering

iKernels

 

 






Advanced Natural Language Processing

and Information Retrieval (ANLP-IR)



Part1:


1. Basic Information Retrieval, Machine Learning Natural Language Processing (PDF)




2. Kernel Methods for NLP (PDF)



3. Named Entity Recognition and POS-Tagging (PDF)



4. Syntactic Parsing (PDF)



5. Coreference and Anaphora Resolution (PDF)





LAB1: Indexing, Word Features, Document and Term Frequency, Text Categorization)



Download:

TCF - Text Categorization Framework



LAB2: Support Vector Machines and Kernel Methods


Download:

(use the TCF above)




LAB2.b: Combining Tree Kernels for Question/Answer classification

Download:

LAB2.b.zip




LAB2.c: Smoothed Partial Tree Kernel for Question Classification

Download:

LAB2.c.zip



LAB3: Ranking with Tree Kernels


Download:

LAB3.zip


Projects 2015-2016

Format of Reports for the Projects


Additional Material





Information Retrieval Lectures



- Motivation and presentation of the course + inverted index
In depth on tokenization, normalization and optimization

Preprocessing, data structures, n-grams and wildcards
- Vector Space Model and weighting schemes
- Efficient
methods for document retrieval
Performance Measures and Query Expansion











The above presentations are  based on the IR courses:
http://nlp.stanford.edu/IR-book/newslides.html

whereas the book (also adopted in my course) is available at:
http://nlp.stanford.edu/IR-book/


  Machine Learning Lectures

- Text Categorization and Feature Selection
- Statistical Learning Theory: linear classifiers
-
Support Vector Machines
- Structured Output Spaces
- Kernel Methods


As referring text please use my new chapter:

Kernel-Based Machines for Abstract and Easy Modeling of Automatic Learning

along with the old book (with some typos)

Roberto Basil and Alessandro Moschitti, Automatic Text Categorization: from Information Retrieval to Support Vector Learning. Aracne editrice, Rome, Italy.

  Natural Language Processing Lectures

- POS-Tagging and Named Entity Recognition
- Syntactic Parsing
- Semantic Role Labeling
- UIMA Introduction
- Coreference Resolution
- Latent Semantic Analysis
- Kernel Methods for Natural Language Processing


 



























































Academic Year: 2012-2013






Informatica (MAT/FIS)

Presentatione del corso 

Introduzione all'Informatica

- Introduzione alla programmazione

- Compilatori, interpeti, e introduzione al C

Costrutti del Linguaggio C

- Array-Stringhe-Matrici-Preprocessore

- L'algebra dei calcolatori

- Tipi di dati avanzati

- Argomenti avanzati

- Struttura dei calcolatori


Materiale aggiuntivo

Slides del corso (Prof. Bianchini)

Altre slides recenti della Prof Bianchini


Overflow

Stack e Record di Attivazione

Complessità Computazionale


Link alle lezioni di laboratorio

  

  



Natural Language Processing and Information Retrieval

 
Information Retrieval Lectures

- Motivation and presentation of the course + inverted index
In depth on tokenization, normalization and optimization
 (optional ppt)
Preprocessing, data structures, n-grams and wildcards
- Vector Space Model and weighting schemes
- Efficient
methods for document retrieval
Performance Measures and Query Expansion

The above presentations are heavily if not totally based on the IR courses of my friends Chris and Hinrich, who with
Prabhakar Raghavan have built an excellent didactic tool. I would like to express my sincere thanks and appreciation for their nice work: their ppts are available at:

http://nlp.stanford.edu/IR-book/newslides.html

whereas the book (also adopted in my course) is available at:

http://nlp.stanford.edu/IR-book/


  Machine Learning Lectures

- Text Categorization and Feature Selection
- Statistical Learning Theory: linear classifiers
-
Support Vector Machines
- Structured Output Spaces
- Kernel Methods


As referring text please use my new chapter:

Kernel-Based Machines for Abstract and Easy Modeling of Automatic Learning

along with the old book (with some typos)

Roberto Basili and Alessandro Moschitti, Automatic Text Categorization: from Information Retrieval to Support Vector Learning. Aracne editrice, Rome, Italy.

  Natural Language Processing Lectures

- POS-Tagging and Named Entity Recognition
- Syntactic Parsing
- Semantic Role Labeling
- UIMA Introduction
- Coreference Resolution
- Latent Semantic Analysis
- Kernel Methods for Natural Language Processing


  Laboratory Lectures

- Setting Search Engines in Java
        - Zip file for the exercise
        - Answerbag dataset
- Answer reranking in Answerbag
       
- Zip file for the exercise





Computational Methods for Data Analysis

  

- Introduction to Machine Learning:  Decision Tree and Bayesian Classifiers

- Vector Space Learning

- Introduction to Statistical Learning Theory

- VC-dimension

- Perceptron

- Support Vector Machines

- Kernel Methods for Structured Data


As referring text please use my new chapter:

Kernel-Based Machines for Abstract and Easy Modeling of Automatic Learning

along with the old book (with some typos)

Roberto Basili and Alessandro Moschitti, Automatic Text Categorization: from Information Retrieval to Support Vector Learning. Aracne editrice, Rome, Italy.








Academic Year: 2011-2012



Computational Methods for Data Analysis

  

- Introduction to Machine Learning:  Decision Tree and Bayesian Classifiers

- Vector Space Learning

- Introduction to Statistical Learning Theory

- VC-dimension

- Perceptron

- Support Vector Machines

- Kernel Methods for Structured Data


As referring text please use my new chapter:

Kernel-Based Machines for Abstract and Easy Modeling of Automatic Learning

along with the old book (with some typos)

Roberto Basili and Alessandro Moschitti, Automatic Text Categorization: from Information Retrieval to Support Vector Learning. Aracne editrice, Rome, Italy.





PhD Course: Natural Language Processing in Watson

  

- Preparation to the Watson Tutorial

- IBM Watson Tutorial (not available yet)


Additionally, choose three lectures from the followings:

    - Introd. to Machine Learning:  Decision Tree and Bayesian Classifiers

    - Vector Space Learning

    - Perceptron + Support Vector Machines

    - Introduction to Statistical Learning Theory + VC-dimension

    - POS-Tagging and Named Entity Recognition

    - Syntactic Parsing





Informatica (MAT/FIS)

Presentatione del corso 

Introduzione all'Informatica

- Introduzione alla programmazione

- Compilatori, interpeti, e introduzione al C

Costrutti del Linguaggio C

- Array-Stringhe-Matrici-Preprocessore

- L'algebra dei calcolatori

- Tipi di dati avanzati

- Argomenti avanzati

- Struttura dei calcolatori


Materiale aggiuntivo

Slides del corso (Prof. Bianchini)

Altre slides recenti della Prof Bianchini


Overflow

Stack e Record di Attivazione

Complessità Computazionale


Link alle lezioni di laboratorio

  



Natural Language Processing and Information Retrieval

 
Information Retrieval Lectures

- Motivation and presentation of the course + inverted index
In depth on tokenization, normalization and optimization
 (optional ppt)
Preprocessing, data structures, n-grams and wildcards
- Vector Space Model and weighting schemes
- Efficient
methods for document retrieval
Performance Measures and Query Expansion

The above presentations are heavily if not totally based on the IR courses of my friends Chris and Hinrich, who with
Prabhakar Raghavan have built an excellent didactic tool. I would like to express my sincere thanks and appreciation for their nice work: their ppts are available at:

http://nlp.stanford.edu/IR-book/newslides.html

whereas the book (also adopted in my course) is available at:

http://nlp.stanford.edu/IR-book/


  Machine Learning Lectures

- Text Categorization and Feature Selection
- Statistical Learning Theory: linear classifiers
-
Support Vector Machines
- Structured Output Spaces
- Kernel Methods


As referring text please use my new chapter:

Kernel-Based Machines for Abstract and Easy Modeling of Automatic Learning

along with the old book (with some typos)

Roberto Basili and Alessandro Moschitti, Automatic Text Categorization: from Information Retrieval to Support Vector Learning. Aracne editrice, Rome, Italy.

  Natural Language Processing Lectures

- POS-Tagging and Named Entity Recognition
- Syntactic Parsing
- Semantic Role Labeling
- UIMA Introduction
- Coreference Resolution
- Latent Semantic Analysis
- Kernel Methods for Natural Language Processing


  Laboratory Lectures

- Setting Search Engines in Java
        - Zip file for the exercise
        - Answerbag dataset
- Answer reranking in Answerbag
       
- Zip file for the exercise






Academic Year: 2010-2011

PhD course: Machine Learning (to be updated)


- Kernel Methods (advanced lecture)

- Kernel Engineering




Informatica Generale

Presentatione del corso 

Introduzione al Corso

Slides del corso (Prof. Bianchini)

Altre slides recenti della Prof Bianchini

Prima e seconda lezione  (prima, seconda)

Overflow


Stack and activation record

Computational Complexity






Machine Learning for Laurea Specialistica

and

Master on Human Language Technology and Interfaces

(Course on Machine Learning for Natural Language Processing and Information Retrieval)


Introduction to Machine Learning

PAC Learning

VC-Dimension

- Perceptrons
- Support Vector Machines

Slides below to be updated

-
Automated Text Categorization (practical machine learning)
-
Lab for Automated Text Categorization
- Kernel Methods
- Tree Kernels (lab)

Projects

Format of Reports for the Projects







Academic Year: 2009-2010


PhD course: Machine Learning 


Kernel Methods (advanced lecture) 

Kernel Engineering 






Informatica Generale

Presentatione del corso 

Introduzione al Corso

Slides del corso (Prof. Bianchini)

Altre slides recenti della Prof Bianchini

Prima e seconda lezione  (prima, seconda)

Overflow


Stack and activation record

Computational Complexity




Machine Learning for Laurea Specialistica

and

Master on Human Language Technology and Interfaces

Course on Machine Learning for Natural Language Processing and Information Retrieval


- Introduction to Machine Learning 

- Automated Text Categorization (practical machine learning)

- PAC Learning

- VC-Dimension

- Perceptrons
- Support Vector Machines
- Lab for Automated Text Categorization
- Kernel Methods
- Tree Kernels (lab)

Projects

Format of Reports for the Projects








 

Academic Year: 2008-2009


PhD course: Machine Learning 


- Kernel Methods (advanced lecture) 

- Kernel Engineering






Informatica Generale


Presentatione del corso 

Introduzione al Corso

Slides del corso

Altre slides recenti della Prof Bianchini

Prima e seconda lezione  (prima, seconda)

Overflow


Stack and activation record

Computational Complexity




Master on Human Language Technology and Interfaces

Course on Machine Learning for Natural Language Processing and Information Retrieval


- Introduction to Machine Learning 

- PAC Learning

- Basic Concepts

- VC-Dimension

- Automated Text Categorization
- Lab for Automated Text Categorization
- Perceptrons
- Support Vector Machines
- Kernel Methods
- Tree Kernels

Projects

Format of Reports for the Projects




Laurea Specialistica


- Introduction to Machine Learning 

- PAC Learning

- VC-Dimension

- Automated Text Categorization
- Lab for Automated Text Categorization
- Perceptrons
- Support Vector Machines
- Kernel Methods
- Tree Kernels










Academic Year: 2007-2008


Master on Human Language Technology and Interfaces

Course on Machine Learning for Natural Language Processing and Information Retrieval




Informatica Generale I

Course Presentation  (PDF)

Link 1  (all slides)

Other More Recent Slides of Bianchini

Link 2  (prima, seconda)

Overflow



Informatica Generale II

From 10 to 15

Stack and activation record

Computational Complexity

Classes of Computational Complexity

Object Oriented Introduction

C++

Exrcise classes (INF.GEN I and II)