Title Natural Language Processing
Lesson Code 321-6100
Semester 7
ECTS 5
Hours (Theory) 3
Hours (Lab) 0
Faculty Stamatatos Efstathios

Syllabus

Introduction: basic concepts, applications. Morphological analysis, text tokenization and sentence splitting. Language modeling using n-grams. Basic supervised learning methods. Deep learning architectures for classification and sequence labeling. Language modeling using neural networks. Vector semantics, word and document embeddings. Text classification and applications. Part-of-Speech tagging and named-entity recognition. Constituency grammars and parsing, stochastic parsing. Dependency parsing. Logical representations of sentence meaning. Semantic analysis.

Learning Outcomes

After successfully completing the course the students should be able to:
 
• describe the basic principles and analysis levels of natural language processing;
 
• understand and use word and document representation techniques;
 
• understand algorithms and use tools for sequence labelling;
 
• understand algorithms and use tools to perform syntactic analysis;
 
• understand algorithms and use tools to classify documents;
 
• get familiar with deep learning methods and their application to natural language processing applications. 

 

Prerequisite Courses

Not required.

Basic Textbooks

1. Russell and Norvig, Artificial Intelligence: A Modern Approach (4th ed.), Pearson, 2005.
 
2. Jurafsky and Martin, Speech and Language Processing (3rd ed.)
 
3. Manning and Schütze, Foundations of Statistical Natural Language Processing, MIT Press, 1999.
 
 

Additional References

1. Bird, Klein & Loper: Natural Language Processing with Python
 
2. Eisenstein: Natural Language Processing (notes)
 

Teaching and Learning Methods

Activity Semester workload
Lectures 52 hours
Laboratory hours 26 hours
Personal study 44 hours
   
Final exams 3 hours

Student Performance Evaluation

Course grading comes from participation in individual laboratory exercises (50%) and written exam (50%). In both cases, lab exercises and written exams, at least 5.0 out of 10 is required. Students are informed about grading policy from the beginning.