|
HU Credits:
3
Degree/Cycle:
1st degree (Bachelor)
Responsible Department:
Computer Sciences
Semester:
1st Semester
Teaching Languages:
Hebrew
Campus:
E. Safra
Course/Module Coordinator:
Omri Abend
Coordinator Office Hours:
Teaching Staff:
Prof Omri Abend
Course/Module description:
Natural Langue Processing (NLP) addresses the automatic analysis of text and speech, and interfaces with various fields such as formal languages, machine learning, linguistics and cognitive psychology. It also has a variety of applications, e.g., in machine translation, information retrieval, and human-computer interaction. The course will present the main challenges the field is facing and (mostly statistical) techniques for addressing them.
Course/Module aims:
Acquaintance with the main problems of the field, its existing capabilities and hands-on experience in implementing them.
Learning outcomes - On successful completion of this module, students should be able to:
Implement basic techniques in Natural Language Processing, and get a basic understanding of the current literature in the field.
Attendance requirements(%):
75
Teaching arrangement and method of instruction:
Oral presentations, accompanied by theoretical and programming exercises.
Course/Module Content:
1. Language models, smoothing and neural language models
2. Bag of words models
3. Log-linear models and feed-forward neural networks
4. Linear chain methods,
tagging, named entity recognition
5. Recurrent neural networks
6. Vector space models of semantics
7. Sentiment analysis
8. Syntactic parsing:
grammar-based and grammar-less methods
9. Information extraction and semantic role labeling
10. Transfer learning in NLP
11. Machine translation
Required Reading:
--
Additional Reading Material:
https://web.stanford.edu/~jurafsky/slp3/
Course/Module evaluation:
End of year written/oral examination 70 %
Presentation 0 %
Participation in Tutorials 0 %
Project work 0 %
Assignments 30 %
Reports 0 %
Research project 0 %
Quizzes 0 %
Other 0 %
Additional information:
There will be five exercises (all mandatory). Some of them theoretical and some will involve programming.
|
