Most part of the data available on the Web is produced via people interaction in natural language. Automatic extraction of knowledge from this data demands the use of efficient Natural Language Processing (NLP) techniques such as text aggregation, text summarization, and text generation. NLP algorithms are essential for creating human-data applications, for example, Google Translator which uses a large amount of text crawled from Web pages for training its machine translation models. Recently, NLP has gained significant momentum with the advancement of artificial neural networks which have shown to surpass most earlier approaches on important problems in language. However, they are known for requiring large amounts of data to perform reasonably. Hence, processing human knowledge on raw text for acquiring semantics still remains a challenge. In this seminar, students will study novel research results from NLP and adjacent fields with the aim of deepening their knowledge of the field. A beneficial side-effect will be a deepening of the students’ expertise in scientific writing.
The course consists of single sessions which will introduce the students to topics related to NLP, scientific writing, and scientific presentations Each student will be assigned a paper to read and present within the first two sessions.
The students’ performance will be evaluated as follows:
Scientific Presentation (33%): Every student will present the paper they were assigned/selected for presentation within a mini-conference format with allocated time slots The attendance to the mini-conference is mandatory.
Final Report (67%): The students will submit a report (20 pages, written using the provided LaTeX template) on current research in the area of their paper containing a formal presentation of the problem, an explanation of related works, existing methodologies along with a discussion about their pros and cons.
L.079.07501 Proseminar: Natural Language Processing (in English)