Deep Learning applied in Natural Language Processing (2022-2023)

The course takes a tour of the main topics in Natural Language Processing (NLP), presenting both traditional and deep learning algorithmic approaches for text processing and analysis.

Courses

• Course 1. Intro to NLP.
• LAB 1 Wordnet practice.
• Course 2. Syntactic and dependency parsing.
• LAB 2. Parse trees.
• Course 3. Language models.
• LAB 3. Testing language models.
• Course 4. Machine Translation. Question Answering.
• LAB 4. Machine translation evaluation.

Evaluation

Punctaje 2022-2023

Back to top

Scripting languages (2022-2023)

Courses

• Course 1: Intro to scripting languages
• Course 2: Shell commands
• Course 3: GREP command
• Exercises
• More exercises
• Reading XML files in Unix using this input and this output file.
• Test: Write a shell program which transforms this input fle into this output file.
• Python. Variables. If-else control structure.
• Python. Input. While control structure.

Announcements

Back to top

Computational Lexicology, Morphology and Syntax (2022-2023)

This course introduces Lexicology, Morphology and Syntax from a computational point of view.

Courses

• Course 1: Introduction to Natural Language Processing.
• Course 2: Lexicology. Word formation. Dictionaries.
• Lab 2: Tokenization.
• Course 3: Morphology
• Lab 3: Project ideas. Teams. Research on state of the art.
• Course 4: POS Tagging
• Lab 4: Project architecture and responsibles.
• Course 5: Finite state automata. Regular Expressions
• Lab 5: Identifying stemming rules and their priority.
• Course 6: Syntactic parse trees
• Lab 6: PC-PATR parsing using this grammar and this lexicon. The reference manual is here. You can download PC-PATR II.
• Course 9: CKY parsing
• Lab 9: Parsing for English and Romanian.
• Course 10: Earley parsing
• Lab 10: Project.
• Course 11: Dependency parsing
• Lab 11: Project.
• Lab 12: Project.
• Course 13: Evaluation metrics
• Lab 13: Project.
• Lab 14: Project.

Evaluation

Back to top

Machine Translation (2021-2022)

This course introduces notions of machine translation, both from theoretical perspective and from the point of view of applications.

Courses

• Course 1: Introduction to Machine Translation
• Lab 1: Direct Translation exercise with this lexicon and these rules.
• Course 2: Various Approaches to Machine Translation
• Lab 2: Bad Translator.
• Course 3: Statistical MT. Models.
• Lab 3: Implement an n-gram Language Model for Romanian using a smoothing technique at your choice.
• Course 4: Translation Models
• Lab 4: Giza++.
• Course 5: Phrase-based translation
• Lab 5: Implement IBM model 1
• Course 6: Reordering
• Lab 6: Phrase-based translation. Symetrization using as input the files you obtained after running GIZA++.
• Course 7: Syntax-based MT
• Course 9: Decoding
• Lab 9: Transform the lexical n-gram language model for Romanian that you have already implemented into a POS n-gram model.
• Course 11: Evaluation of MT
• Lab 11: Project' progress
• Course 13: Neural Machine Transaltion

Evaluation

Back to top

Procesarea statistica a datelor textuale (FEAA) (2021-2022)

Courses

Announcements

Back to top