Automatic Building of a Semantic Model of Disease Symptoms Based on Text Corpus
Journal Title: Computer Science and Mathematical Modelling - Year 2014, Vol 0, Issue 14
Abstract
The research described in the article refers to the study of data from the domain of medicine. The diagnostic test results are recorded in different ways. They may take the form of tables, graphs or images. Regardless of the original data format, it is possible to draw up their verbal description, which focuses on the description of the observed symptoms. Such descriptions make up the text corpora concerning individual diagnostic technologies. Knowledge on disease entities is stored in a similar manner. It has the form of text corpora, which contain descriptions of symptoms specific to individual diseases. By using natural language processing tools semantic models can be automatically extracted from the texts to describe particular diagnostic technologies and diseases. One of the obstacles is the fact that medical knowledge can be written in a natural language in many ways. The application of the semantic format allows the elimination of record ambiguities. Ultimately, we get a unified model of medical knowledge, both from the results of diagnostic technologies describing the state of the patient and knowledge of disease entities. This gives the possibility of merging data from different sources (heterogeneous data) to a homogeneous form. The article presents a method of generating a semantic model of medical knowledge, using lexical analysis of text corpora.
Authors and Affiliations
Grażyna Szostek, Marek Jaszuk, Andrzej Walczak
Keywords
Related Articles
- EP ID EP63206
- DOI 10.5604/15084183.1152151
- Views 129
- Downloads 0
SAVE - State of the Art and Visionary Energetics
The paper focuses on the attempt to improve the process of modeling the electrical grid infrastructure, designed within project SAVE (State of the Art and Visionary Energetics). The system, which is a response to a still...
Interoperability and Security Standards and Rules in the Polish Law on Informatization
Observing time-period from 2003 till 2008, polish governmental efforts in informatization was focused on e-government legal foundations and IT standards too. Main purpose of this article is to achieve short discussion of...
Wykorzystanie metod badania podobieństwa grafów i sieci do analizy sieci złożonych
W artykule zaproponowano koncepcję wykorzystania metod badania podobieństwa grafów i sieci do analizy sieci złożonych. Omówiono podstawowe modele sieci złożonych oraz metody badania podobieństwa grafów i sieci. Następnie...
Metody badania wybranych usług dodanych w sieciach rozległych
W publikacji przedstawiono wybrane usługi dodane wykorzystywane w sieciach rozległych, a także scharakteryzowano podstawowe typy architektur sieci, w których posiadają one zastosowanie. Omówiono również metody formalnego...
Electronic services security management for the public administration
The public administration in Poland is constantly extending and improving electronic services provided to citizens, business and itself (e.g. one ministry providing services to another public administration institution)....