HSE Graduate Receives Bertrand Meyer Award
Polina Kazakova, graduate of the 'Fundamental and Computational Linguistics' Bachelor’s Programme, has won the Bertrand Meyer award for the best research project at the Software Engineering Conference in Russia 2018.
The SECR has been held annually for the last thirteen years and brought together researchers, developers and representatives of market leaders in IT. Participants present reports and articles at the conference, with the the best ones being awarded the prize established by Bertrand Mayer, a famous French programmer, and creator of the Eiffel language.
In 2018, the prize was awarded to the ‘Applying Topic Segmentation to Document-Level IR’ project. This project is based on the work carried out by HSE graduate Polina Kazakova in cooperation with her colleague Gennady Shtekh, MSU student Nicholay Skachkov, and led by Nikita Nikitinsky and Konstantin Vorontsov, Academic Supervisor of the project, Professor at the Faculty of Computer Science.
‘The idea is that short and topically homogeneous (coherent) text fragments are more suitable for presentation in vector, numerical form,’ says Polina Kazakova. ‘These concepts are applied by various systems associated with natural language processing (NLP). We tested our hypothesis by solving the problem of information retrieval, where we often have to work with large and topically heterogeneous documents. It turned out that if we manage to divide large documents into smaller semantic pieces, we can really improve the retrieval quality. We’ll keep on experimenting with this topic, and we may also try to use this segmentation in other tasks.’
Polina has been engaged in natural language processing since studying on the 'Fundamental and Computational Linguistics' Bachelor’s Programme (she graduated in 2017). Now she works in data science for the IRELA project.
‘Soon we will launch our first big project - a data analysis platform consisting of several modules that perform different functions of data analysis and natural language processing. For example, we have a multilingual search module — it allows you to search for documents written in different languages without manually requesting translating. This is quite a difficult task, much more difficult than monolingual searching. And in this we apply the same topic segmentation of documents to get the best results. So we prepare articles not merely for the sake of articles, but in order to use the results in practice,’ notes Polina Kazakova.
An article by HSE Faculty of Computer Sciences senior Artyom Gadetsky was accepted for presentation at the Association for Computational Linguistics international conference, the only A* level conference on computational linguistics. According to the CORE system, which ranks large conferences in computer science, the category A* is the highest level for a conference.
The 8th RuSSIR-2014 (Russian Summer School in Information Retrieval), organized by the Higher School of Economics in Nizhny Novgorod and ROMIP, has ended.