- To develop NLP applications using existing ready-to-use tools
- To analyze the interdisciplinarity involved in developing NLP applications
The objective of this course is to introduce the field of Natural Language Processing (NLP) through the most used applications both in academia and the industry. The course contents will include basic techniques of NLP: document classification, sequence labeling for opinion mining, vector-based word representations (embeddings), and normalization and pre-processing of texts. The course will have a practical focus based on laboratories and practical tasks learning to use NLP tools based on machine and deep learning.
| Slides | Labs |
|---|---|
| 1. Introduction to Multilingual NLP tasks | 1. Spacy basics. Training text classifiers with Spacy |
| 2. Text Classification | 2. Feature Based Stance Detection. Stance Detection with Logistic Regression. |
| 3. Fake News, Fact-checking, Stance | 3. Flair basics. Flair text classification training |
| 4. Word Representations for Named Entity Recognition | 4. Flair NER tagging training |
| 5. Contextual Lemmatization and Morphology | 5. Neural contextual lemmatization as sequence labelling |
| 6. Opinion Mining - Aspect Based Sentiment Analysis | 6. Aspect-based Target Extraction - Sentiment Analysis |
| 7. Question Answering and other intermediate tasks | 7. Sequence Labelling and Multilabel classification with Transformers |
| 8. Argumentation and Inference | 8. Argument Mining and Argument Relation Extraction |
| 9. Text generation tasks | 9. Generating counter-arguments |
-
Labs. Students will be assessed on the activities proposed during the module. Max. 5 points (50%)
-
Project. Students can undertake an optional submodule-specific project of their choice. Max. 3.5 points. (35%). The evaluation will be based on the following criteria: initiative, workload (min. 15 hours), application of concepts learned in the course and documentation (2 people group):
- Evaluation criteria: initiative, workload (min. 15 hours), application of concepts learned in the module and documentation.
- Important dates:
- February 15: publish project proposals on egela
- March 02: Students decide on a project or submodule to work on and notify me
- April 7: deadline for submission (submissions after this date will be penalized by 0.5 points)
- April 17: hard deadline (no submission will be accepted after this date)
-
Attendance. 1 point if you attend 80% of the classes (including invited talks)
-
Participation. The remaining 0.5 points (5%) will be based on the students participation in class discussions and activities.