Scientific Conference on Computer Linguistics and Intelligent Technologies “Dialogue”
“The rapid growth of artificial intelligence technologies and the powerful development of open source platforms have made natural language analysis one of the most advanced and dynamic areas in science, – comments Vladimir Selegey, Director of Linguistic Studies, ABBYY. – Participants of the “Dialogue” actively use the most modern methods to solve their problems, and in this sense, each conference is not similar to the previous one, because much has changed over the year. As before, Dialogue remains an opportunity to discuss the Russian agenda in an international context, and this year is no exception, even in a new format. At the Dialogue, we will see almost everyone who “makes the weather” in the Russian NLP. ”
“It is well known that the PhysTech School of Applied Mathematics and Informatics (FPMI) MIPT is a powerful center for science and education in the field of technical specialties. However, the arrival of the “Dialogue” conference on our site confirms that we are not limited to leadership in the fields of mathematics and computer science. FPMI is interesting in dialogue with the humanities and activities at the intersection of artificial intelligence and language. ABBYY is our long-standing partner, with as many as two base departments and a laboratory at MIPT. Nevertheless, I regard the current event as a landmark: a strong humanitarian direction appears at the PhysTech ”, – comments Andrey Raygorodsky, Director, FPMI MIPT.
The event will feature recognized experts in the field of computer linguistics:
- Manfred Shtede, professor of applied computer linguistics, University of Potsdam, author of the popular textbook for students in the field of natural language processing Argumentation Mining. Manfred analyzes connected text and uses these links to extract information about the author’s logic. At the conference, he will deliver a report From Coherence Relations to Argumentation Mining.
- Sergey Sharov, professor at Lida University, specialist in language transfer tasks. Sergey will moderate the “Corpusometry and Statistical Analysis of Corps” section and will present the topic of Internet Corpora Topography.
- Igor Boguslavsky, professor at the Technological University of Madrid and head of the Laboratory of Computational Linguistics, IPPI RAS, one of the largest specialists in ontological semantics. At the conference, Igor will present the topic An Approach to Triangle-COPA Social Scenarios Based on Semantic Analysis.
- Olga Inkova, Professor, University of Geneva. Olga will talk about major projects related to discourse analysis and give a presentation on A Quantitative Method for Analysis of Connectives: the “Portrait” of the Russian Conjunction ili ‘or’ in the Supracorpora Database of Connectives.
The conference will traditionally discuss the results of annual competitions DialogueEvaluationwhose main task is to develop common principles for proving the effectiveness of the results of linguistic research. To conduct such competitions, specially prepared data (datasets) are needed to train the tested algorithms.
This year competitions were held in three areas:
- Competitions RuREBus (Russian Relation Extraction for Business) for the automatic extraction of semantic relations in a setting that is close to business tasks.
- GramEval competitions in automatic morphosyntactic text analysis.
- Competitions for the automatic search for hyperonyms in Russian.
Prizewinners and winners of the competition, among which employees of leading IT companies, will talk about their decisions.
Also within the framework of the “Dialogue” there will be several discussions, including a round table “UPGREAT PRO / READING NLP-technology barrier: how to correctly pose (and solve) the problem of understanding the text”. Among the speakers are ABBYY AI evangelist Ivan Yamshchikov, Mikhail Burtsev, head of the neural network and deep learning laboratory at MIPT, and Konstantin Vorontsov, professor at the Yandex Data Analysis School and professor at the Higher School of Economics.
Participation in the “Dialogue” for all but legal entities is free for the first time this year. All registrants will receive links to video recordings of the speakers’ full reports two days before their presentation at the conference, in order to familiarize themselves with them in advance. On-line speakers will present short versions of their reports and answer questions from listeners.
The best materials of the conference will be published in the yearbook “Computer Linguistics and Intelligent Technologies”, which is part of the international citation system Scopus.
You can find out more about the Dialog 2020 conference and register for the event at website. The working languages of the conference are Russian and English.
