Natural Language Processing for Massive Textual Data Management
(PLN-PMT)


Basic Information
=================
Course type: optional	
Semester: third
ECTS : 6 (12 "punts de docència")	
Periodicity: anual
UPC structural unit responsible of the course: Software Department (LSI)
Teachers: Lluís Màrquez (main), Jordi Turmo
Language: Catalan / Spanish / English
Requirements: you have to course the following subjects before doing
  PLN-PMT: Processament del Llenguatge Natural, Aplicacions de la
  Intel·ligència Artificial, Aprenentatge


Description and Goals
=====================

The main goal of this course is to provide the students with an in
depth knowledge of the techniques, methods and tools, both symbolic
and empirical, of Natural Language Processing (NLP). The course
focuses on the systems dealing with the analysis and processing of
massive quantities of textual data. The applications in this domain
usually work in a batch mode and have their basic framework in
Internet and very large textual data bases.

After taking this course we expect students to be familiar with the
basic bibliography of this area of NLP and have the capacity and
skills for performing a future in-depth research in any of the themes
covered by the course. Also, the range of applications studied allows
the students to bridge the gap between the language technologies
studied and the real-world applications in which they take part.
  
This course is highly coupled with the course covering Natural
Language applications for person-machine communication (Natural
Language Processing for Human-Machine Communication). By taking both
courses, the student will be able to get a sufficient knowledge of the
two basic paradigms of NLP in the framework of the two most frequent
scenarios.

A final goal of the course is the presentation of the most active
research areas within the topics of the course.


Contents
========

The content of the course is organized into three main blocks:

(1) The most representative applications based on a massive processing
    of textual data. These applications are currently being used
    mainly in the context of Internet processing and the automatic
    organization of very large document data bases, but they are still
    the focus of very active research. The concrete set of
    applications that will be covered by the course are: Document
    Categorization, Information Extraction, and Automatic
    Summarization.

(2) Basic generic tasks, which can be very useful for the applications
    listed above (apart from others). We will cover only those that
    have not been introduced by previous mandatory courses of the
    Master. More concretely, the generic tasks studied will be:
    partial parsing, word sense disambiguation, and semantic role
    labeling.

(3) The introduction of advanced Machine Learning Techniques for
    Natural Language Processing. These algorithms and techniques are
    very useful for implementing most of the generic tasks described
    in the previous point. By "advanced" we mean that the ML topics
    covered extend the basic techniques that are already known by
    students from the previous mandatory courses of the Master.

The table of contents is structured in four themes. The numbers
accompanying each of the titles give an orientation of the percentage
of the course that will be devoted to the corresponding theme. As can
be seen, the main focus will be on the applications.


Table of Contents
=================

1. Introduction (5%)

    1.1 The necessity of automatically processing massive quantities
        of textual data. Main applications in this domain.

2. Advanced Topics in Machine Learning (30%)

    2.0 Review of the main concepts of Machine Learning
    2.1 Statistical Methods: Maximum Entropy modeling
    2.2 Discriminative Learning Methods: Boosting, Support Vector Machines
    2.3 Learning & Inference for relational and structured domains
    2.4 Semi-supervised Learning: Bootstrapping, co-training and variants

3.  Generic Tasks (25%)

   3.1 Partial parsing: chunking, clause boundary detection
   3.2 Word Sense Disambiguation
   3.3 Semantic Role Labeling

4. Applications (40%)

   4.2. Information Extraction: typology, adaptability,
        multilinguality, evaluation
   4.1. Document Categorization: thematic classification, using
        hierarchies of concepts from the Web, subjective
        classification (intention, sentiment, etc.)
   4.3. Automatic Summarization: single document, multi-document, multilingual


Evaluation Method
=================
There will be three aspects subject to evaluation:

1) Development and implementation of a linguistic tool related to the
   tasks described in points 3 and 4 of the program. 40% of the final
   score. Work in group.

2) Oral defense of the preceeding work. 20% of the final score. 

3) Writting a survey (state-of-the-art) about one of the suggested
   themes or, alternatively, oral presentation and class discussion of
   one of the complementary readings suggested along the course. 40%
   of the final score.