CASPIAN JOURNAL
MANAGEMENT AND HIGH TECHNOLOGIES
Relationship identification between the economic terms and objects in natural language text
 |
Read | Dmitriev A.S., Solovev I.S., Orlova Yu.A., Rozaliev V.L., Konstantinov V.M. Relationship identification between the economic terms and objects in natural language text // Caspian journal : management and high technologies. — 2015. — №4. — pp. 198-210. |
Dmitriev A.S. - senior lecturer, Volgograd State Technical University, 28 Lenin Ave., Volgograd, 400005, Russian Federation, golostos@yandex.ru
Solovev I.S. - software developer, SurfStudio, 1d Srednemoskovskaya St., Voronezh, 394018, Russian Federation, issoloveyv@gmail.com
Orlova Yu.A. - Ph.D. (Engineering), Ph.D. (Pedagogics), Associate Professor, Volgograd State Technical University, 28 Lenin Ave., Volgograd, 400005, Russian Federation, yulia.orlova@gmail.com
Rozaliev V.L. - Ph.D. (Engineering), Associate Professor, Volgograd State Technical University, 28 Lenin Ave., Volgograd, 400005, Russian Federation, vladimir.rozaliev@gmail.com
Konstantinov V.M. - post-graduate student, Volgograd State Technical University, 28 Lenin Ave., Volgograd, 400005, Russian Federation, konstantinovr1@gmail.com
Despite the rapid development of systems for processing text, there are fairly small numbers of software systems for efficient search links within a text. It is particularly acute when processing a large number of economic texts and extraction of important financial and economic terms. As part of this work the software system was developed for the identification of objects and terms of economic subjects. The program also allows you to specify which object is associated one or another economic term. We have developed the template method on the basis of Snowball to identify objects and terms in the text. This paper describes the characteristics of terms and objects and features of the context for the template method. For identify and clarify the relationships between objects and terms a support vector machine (SVM) is used. Then shows the algorithms of described methods and shows the overall architecture of the software system.
Key words: экономические термины, идентификация объектов, текст, извлечение отношений, контекст предложения, программная система, интеллектуальный анализ данных, метод опорных векторов, economic terms, object identification, text, relation extraction, context of the