The Laboratory for Models and Methods of Computational Pragmatics was founded at HSE in 2019 by Prof. Dmitry I. Ignatov.
Aims and objectives of the Laboratory
In addition, we consider it important to collect high-quality labeled cases - sources of language data - and publish them in the public domain, as The creation of new sources of language data is an important contribution to the development of the community and the state of the word processing industry as a whole.
We solve the following tasks:
- Development and comparison of multimodal clustering and classification methods.
- Development and comparison of methods of compression, rarefaction and regularization of neural networks, in particular, recurrent networks for language modeling.
- Marking of cases of question-answer pairs, cases marked up by entities and events and the relationships between them.
- Adaptation of standard architectures for English for the Russian language and development of own solutions for the tasks of finding the answer to a question, semantic parsing, information extraction, detection of quotes and indirect speech, etc.
- Development of models for concise presentation (embeddings) of texts for recommendation systems.
- Methods for extracting semantic patterns from texts based on data mining, analysis of formal concepts and others.
- Understanding the intentions of users of recommendation and other, including mobile services.
The main activities of the Laboratory
- Scientific and research activities: in accordance with the tasks listed above.
- Educational activities: conducting courses on the subject of the Laboratory at the bachelor's and master's programs of the FCS, open seminars for the general public.
- Project activity: implementation of industrial projects in cooperation with companies.
Have you spotted a typo?
Highlight it, click Ctrl+Enter and send us a message. Thank you for your help!
To be used only for spelling or punctuation mistakes.