scholarly journals RESEARCH OF APPROACHES TO THE RECOGNITION OF SEMANTIC IMAGES OF SCIENTIFIC PUBLICATIONS BASED ON NEURAL NETWORKS

2021 ◽  
Vol 2 ◽  
pp. 38-43
Author(s):  
Iuliia Bruttan ◽  
Igor Antonov ◽  
Dmitry Andreev ◽  
Victor Nikolaev ◽  
Tatyana Klets
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Information Retrieval ◽  
Scientific Publication ◽  
Scientific Publications ◽  
Network Training ◽  
Retrieval Systems ◽  
Intelligent Information Retrieval ◽  
Information Retrieval Systems ◽  
Intelligent Information

The paper is devoted to the problems of orientation and navigation in the world of verbal presentation of scientific knowledge. The solution of these problems is currently hampered by the lack of intelligent information retrieval systems that allow comparing descriptions of various scientific works at the level of coincidence of semantic situations, rather than keywords. The article discusses methods for the formation and recognition of semantic images of scientific publications belonging to specific subject areas. The method for constructing a semantic image of a scientific text developed by Iuliia Bruttan allows to form an image of the text of a scientific publication, which can be used as input data for a neural network. Training of this neural network will automate the processes of pattern recognition and classification of scientific publications according to specified criteria. The approaches to the recognition of semantic images of scientific publications based on neural networks considered in the paper can be used to organize the semantic search for scientific publications, as well as in the design of intelligent information retrieval systems.

A linked open data framework to enhance the discoverability and impact of culture heritage

2018 ◽  
Vol 45 (6) ◽  
pp. 756-766 ◽  
Author(s):  
Gustavo Candela ◽  
Pilar Escobar ◽  
Rafael C Carrasco ◽  
Manuel Marco-Such
Keyword(s):  
Information Retrieval ◽  
Open Data ◽  
Knowledge Bases ◽  
Linked Open Data ◽  
Miguel De Cervantes ◽  
Data Framework ◽  
Retrieval Systems ◽  
All Solutions ◽  
Information Retrieval Systems ◽  
Culture Heritage

Cultural heritage institutions have recently begun to consider the benefits of sharing their collections using linked open data to disseminate and enrich their metadata. As datasets become very large, challenges appear, such as ingestion, management, querying and enrichment. Furthermore, each institution has particular features related to important aspects such as vocabularies and interoperability, which make it difficult to generalise this process and provide one-for-all solutions. In order to improve the user experience as regards information retrieval systems, researchers have identified that further refinements are required for the recognition and extraction of implicit relationships expressed in natural language. We introduce a framework for the enrichment and disambiguation of locations in text using open knowledge bases such as Wikidata and GeoNames. The framework has been successfully used to publish a dataset based on information from the Biblioteca Virtual Miguel de Cervantes, thus illustrating how semantic enrichment can help information retrieval. The methods applied in order to automate the enrichment process, which build upon open source software components, are described herein.

Structural Text Mining

2005 ◽  
pp. 2658-2661
Author(s):  
Vladimir A. Kulyukin ◽  
John A. Nicholson
Keyword(s):  
Information Retrieval ◽  
Natural Language Processing ◽  
Text Mining ◽  
Language Processing ◽  
World Wide ◽  
Knowledge Bases ◽  
Domain Specific ◽  
Retrieval Systems ◽  
The World ◽  
Information Retrieval Systems

The advent of the World Wide Web has resulted in the creation of millions of documents containing unstructured, structured and semi-structured data. Consequently, research on structural text mining has come to the forefront of both information retrieval and natural language processing (Cardie, 1997; Freitag, 1998; Hammer, Garcia-Molina, Cho, Aranha, & Crespo, 1997; Hearst, 1992; Hsu & Chang, 1999; Jacquemin & Bush, 2000; Kushmerick, Weld, & Doorenbos, 1997). Knowledge of how information is organized and structured in texts can be of significant assistance to information systems that use documents as their knowledge bases (Appelt, 1999). In particular, such knowledge is of use to information retrieval systems (Salton & McGill, 1983) that retrieve documents in response to user queries and to systems that use texts to construct domain-specific ontologies or thesauri (Ruge, 1997).

Relevance Predictability in Information Retrieval Systems

1967 ◽  
Vol 06 (02) ◽  
pp. 45-51 ◽  
Author(s):  
A. Kent ◽  
J. Belzer ◽  
M. Kuhfeerst ◽  
E. D. Dym ◽  
D. L. Shirey ◽  
...  
Keyword(s):  
Information Retrieval ◽  
Experimental Conditions ◽  
Treatment Groups ◽  
Retrieval Systems ◽  
Significant Difference ◽  
Information Retrieval Systems ◽  
High Predictability ◽  
Intermediate Response ◽  
Quantitative Indicators ◽  
Level Of Processing

An experiment is described which attempts to derive quantitative indicators regarding the potential relevance predictability of the intermediate stimuli used to represent documents in information retrieval systems. In effect, since the decision to peruse an entire document is often predicated upon the examination of one »level of processing« of the document (e.g., the citation and/or abstract), it became interesting to analyze the properties of what constitutes »relevance«. However, prior to such an analysis, an even more elementary step had to be made, namely, to determine what portions of a document should be examined.An evaluation of the ability of intermediate response products (IRPs), functioning as cues to the information content of full documents, to predict the relevance determination that would be subsequently made on these documents by motivated users of information retrieval systems, was made under controlled experimental conditions. The hypothesis that there might be other intermediate response products (selected extracts from the document, i.e., first paragraph, last paragraph, and the combination of first and last paragraph), that would be as representative of the full document as the traditional IRPs (citation and abstract) was tested systematically. The results showed that:1. there is no significant difference among the several IRP treatment groups on the number of cue evaluations of relevancy which match the subsequent user relevancy decision on the document;2. first and last paragraph combinations have consistently predicted relevancy to a higher degree than the other IRPs;3. abstracts were undistinguished as predictors; and4. the apparent high predictability rating for citations was not substantive.Some of these results are quite different than would be expected from previous work with unmotivated subjects.

Sign in / Sign up

Export Citation Format

Share Document