Natalia V. Loukachevitch

Learn More
This paper describes the SemEval 2016 shared task on Aspect Based Sentiment Analysis (ABSA), a continuation of the respective tasks of 2014 and 2015. In its third year, the task provided 19 training and 20 testing datasets for 8 languages and 7 domains, as well as a common evaluation procedure. From these datasets, 25 were for sentence-level and 14 for(More)
In the paper we present description of Thesaurus on Sociopolitical life, which was constructed as a tool for automatic text processing of large text collections. Specific features of the thesaurus in comparison to conventional information-retrieval thesauri for manual indexing are described. Evaluation of thesaurus-based information retrieval for short(More)
Semantic relatedness of terms represents similarity of meaning by a numerical score. On the one hand, humans easily make judgements about semantic relatedness. On the other hand, this kind of information is useful in language processing systems. While semantic relatedness has been extensively studied for English using numerous language resources, such as(More)
In this paper lexical cohesion modeling is considered. We argue that to model lexical cohesion in connected texts it is not enough to find related words in a text. It is important to take into account relations between entities participating in the described situations. Consideration of this factor gives the possibility to develop more flexible lexical(More)
In this paper we consider a method for extraction of sets of semantically similar language expressions representing different participants of the text story – thematic nodes. The method is based on the structural organization of news clusters and exploits comparison of various contexts of words. The word contexts are used as a basis for multiword expression(More)
In this paper we consider a new approach for domain-specific sentiment lexicon extraction in Russian. We propose a set of statistical features and algorithm combination that can discriminate sentiment words in a specific domain. The extraction model is trained in the movie domain and then utilized to other domains. We evaluate the quality of obtained(More)