Hybrid Search for Russian: Combining Graph Information and Question Answering

Student: Katricheva Nadezhda

Supervisor: Anastasiya A. Bonch-Osmolovskaya

Educational Programme: Computational Linguistics (Master)

Year of Graduation: 2020

Question answering (QA) task presupposes drawing answers from texts, a knowledge base (KB) or a combination of both. Questions can be complex (e.g. multi-hop) and simple (one-hop), and the majority of research is conducted for the English language. In this work, a hybrid QA system for one-hop questions in Russian is proposed, which employs Wikipedia articles and the Wikidata knowledge graph (KG), a large but (in truth and for experiments) incomplete KB, as data sources. The recent neural network (NN) advances in QA are combined with traditional natural language processing (NLP) methods to show that text and KB synergy outperform mono-source approaches. System evaluation required building a new dataset, where answers can be found in each of the data sources; together with the source code and evaluation results, it is available on GitHub.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses