USE OF CONVOLUTIONAL NEURAL NETWORKS FOR THE TASK OF CLASSIFYING TEXTS

Abstract

Convolutional neural networks are a powerful tool of machine learning, which is aimed at efficient recognition and classification of images. The success of using convolutional neural networks for images has given rise to many attempts to use this tool in other problems. In this paper, we study the basic methods of using convolutional neural networks for the task of classifying texts. Experiments were performed on large-scale text data, which showed that convolutional neural networks for a word classification problem can achieve a quality similar to or better than traditional methods.

Authors and Affiliations

Artem Karpovych

Keywords

Related Articles

«SOFT LAW» AS A GENERAL-PURPOSE REGULATOR OF INTERNATIONAL RELATIONS

The efficiency of soft law as a regulator of legal relations, which arise in different spheres is examined in this article. Authors analyze international soft legal norms, which are used by international organizations in...

ERODED SOIL OF THE SLOPE OF WESTERN EXPOSURE NEAR THE SETTLEMENT ZABALA, PICHINCH PROVINCE, ECUADOR

The soils of the western exposure slope in the equatorial Andes have been studied. The soils have been found to have low content of organic matter, not exceeding 0.49%. There is also a lack of Nitrogen and Phosphorus. Th...

THE PLACE OF PLUNDER BY MISAPPROPRIATION AND WASTE OF AN OTHERS PROPERTY AMONG THE OTHER FORMS OF PLUNDER

In article the assignment and waste place among other forms of plunders is considered. Definition of a form of plunder allows differentiating correctly responsibility; it is fair to individualize punishment. In article i...

ANALYSIS AND EVALUATION PROGRAMS OF COMPUTER GRAPHICS FOR 3D VISUALIZATION OF INTERIORS AND EXTERIORS

This article will focus on the existing software used to design space, filling objects of the interior and exterior. It will be compared for their different characteristics, evaluated and conclusions about the best and w...

APPLYING MATHCAD SOFTWARE PACKAGE FOR SOLVING DIFFERENTIAL EQUATIONS

This article discusses the use of MathCAD software to solve differential equations of the first order.

Download PDF file
  • EP ID EP610484
  • DOI 10.25313/2520-2057-2018-14-4105
  • Views 88
  • Downloads 0

How To Cite

Artem Karpovych (2018). USE OF CONVOLUTIONAL NEURAL NETWORKS FOR THE TASK OF CLASSIFYING TEXTS. Международный научный журнал "Интернаука", 1(14), 69-73. https://europub.co.uk./articles/-A-610484