Coreference resolution for Russian: the impact of semantic features
This paper presents the results of our experiments on building a general coreference resolution system for Russian. The main aim of those experiments was to set a baseline for this task for Russian using the standard set of features developed and tested for coreference resolution systems created for other languages. We propose several baseline systems, both rule-based and ML-based. We show that adding some semantic information is crucial for the task and even the small amount of data can improve the overall result. We show that different types of semantic resources affect the performance differently and sometimes more does not imply better.
This book constitutes a collection of selected contributions from the 12th International Conference on Perspectives in Business Informatics Research, BIR 2013, held in Warsaw, Poland, in September 2013. Overall, 54 submissions were rigorously reviewed by 41 members of the Program Committee representing 21 countries. As a result, 19 full and 5 short papers from 12 countries have been selected for publication in this volume. This book also includes the two keynotes by Witold Abramowicz and Bernhard Thalheim. The papers cover many aspects of business information research and have been organized in topical sections on: business process management; enterprise and knowledge architectures; organizations and information systems development; information systems and services; and applications.
The paper describes the development of a portal about development and use of tools based on the (meta) modeling (using DSM, DSL, etc.). The architecture of a portal, information retrieval subsystem and document management are described.
The purpose of the portal is the creation of "selfdeveloping" resource, which provides intelligent search and automatic processing of the results (documents and sources), easy navigation on the found resources. Implementation is based on the ontologies approach.
The main feature of suggested methods is an integrated approach to development. The approach bases on a multi-level ontology repository. The portal allows searching and analyzing information, creating and researching model, publishing research results. Software gives an opportunity of a flexible customizing. The main topic of this paper is an intelligent information search means based on semantic indexation, automatic document classification, tracking of semantic links between documents and automatic summarization.
Quotation, being one of the most common structural and semantic constituents of media discourse, is widely employed by a great number of journalists time and again. Since it may perform various functions, quotation constitutes a powerful tool at reporters’ disposal used to suit their own purposes. Consequently the analysis of this issue can be found in numerous works, but authors mostly study quotation from a stylistic perspective, paying less attention to its lexical aspect, which requires a closer examination. Thus the present research targets at the analysis of quotes’ structural types and their semantic features in English media discourse, subjecting to conscious scrutiny the borrowing of a quote’s form and content in the process of introducing the quoted text into a new one. For carrying out this research, quotations from the following newspapers: the New York Times, the Financial Times, the Economist, the Spectator, the Sun, The Wall Street Journal were examined according to existing classifications. The research is based on a descriptive analysis of various quote types and their specific features. Despite the fact that some statistical data are provided in this paper, the approach adopted is more observational. The findings suggest that although the quote is expected to meet the structural and semantic organization of an authentic text, in media discourse authors often break this rule for pursuing particular objectives. The study reveals that intentional changes in reported speech and the influence of a new context environment may alter the meaning of the quoted utterance. The research outcomes verify the idea that quotation, being a journalist’s powerful tool, plays an important role within a text and the understanding of quotes’ structural and semantic peculiarities is likely to assist readers in perceiving and analyzing information.
This paper concerns discourse-new mention detection in Russian. This might be helpful for different NLP applications such as coreference resolution, protagonist identification, summarization and different tasks of information extraction to detect the mention of an entity newly introduced into discourse. In our work, we are dealing with the Russian where there is no grammatical devices, like articles in English, for the overt marking a newly introduced referent. Our aim is to check the impact of various features on this task. The focus is on specific devices for introducing a new discourse prominent referent in Russian specified in theoretical studies. We conduct a pilot study of features impact and provide a series of experiments on detecting the first mention of a referent in a non-singleton coreference chain, drawing on linguistic insights about how a prominent entity introduced into discourse is affected by structural, morphological and lexical features.
This book constitutes the refereed proceedings of the 4th Conference on Knowledge Engineering and the Semantic Web, KESW 2013, held in St. Petersburg, Russia, in October 2013. The 18 revised full papers presented together with 7 short system descriptions were carefully reviewed and selected from 52 submissions. The papers address research issues related to knowledge representation, semantic web, and linked data.
Today many problems that are dedicated to a particular problem domain can be solved using DSL. Thus to use DSL it must be created or it can be selected from existing ones. Creating a completely new DSL in most cases requires high financial and time costs. Selecting an appropriate existing DSL is an intensive task because such actions like walking through every DSL and deciding if current DSL can handle the problem are done manually. This problem appears because there are no DSL repository and no tools for matching suitable DSL with specific task. This paper observes an approach for implementing an automated detection of requirements for DSL (ontology-based structure) and automated DSL matching for specific task.
The paper concerns discourse-new referent detection. The task of coreference resolution is essential in many text-mining applications. The focus in this task is to detect noun phrases (NPs) that refer to the same entity. In languages without articles, there are no overt grammatical clues in an NP for whether it introduces a new referent into discourse or it refers to one of before-mentioned entities. However, there are some theoretical researches which claim that referent first-mentioning NPs have some specific features. In our research, we examine features that serve as discourse-new detectors for NPs corresponding to discourse salient referents and provide an experiment on different features contribution to this detection. The first-mention detection could help the quality of coreference resolution systems.
The aim of this article is to highlight the relationships between contemporary tendencies in the humanities (the new ontologies) and contemporary architectural practices. The author articulates the distinction between the optics of the «old ontologies» and the new ones. The ontologies considered to be new ones are flat, free from classical opposition between the whole and the parts and based on modality of possibility, but not obligation. Objects and practices traditionally referred to as architecture appear to be based on the principles of the «old ontologies». For them human being is an extraordinary object compared to others, the part-to-whole relationships appear to reflect either the superiority of the whole (society) or the superiority of the part (individual), finally, they are aimed at creating an “it has to be this way” picture. The new ontologies seem to be impossible to apply to architecture in its traditional meaning. Nevertheless, a two-fold link between the new ontologies and architecture can be posed. On the one hand, the former offer a new language to describe the variety of traditional architecture and accept that all of directions, styles and buildings are ontologically coordinate. On the other hand, the new ontologies enable some new architectural practices (computer architecture, architecture of virtual space and speculative architecture) which do not substitute for traditional architecture, but accompany it.
Keywords: new ontologies, flat ontologies, architecture, computer architecture, architecture of virtual space, speculative architecture