User Guide
Why can I only view 3 results?
You can also view all results when you are connected from the network of member institutions only. For non-member institutions, we are opening a 1-month free trial version if institution officials apply.
So many results that aren't mine?
References in many bibliographies are sometimes referred to as "Surname, I", so the citations of academics whose Surname and initials are the same may occasionally interfere. This problem is often the case with citation indexes all over the world.
How can I see only citations to my article?
After searching the name of your article, you can see the references to the article you selected as soon as you click on the details section.
 Views 20
 Downloands 2
 Audio Listening 1
Üretken Rakip Ağ ile Türkçe Metin Üretimi
2021
Journal:  
Avrupa Bilim ve Teknoloji Dergisi
Author:  
Abstract:

Makine öğrenmesi yöntemlerinde tahmin aşamasının başarısı için kullanılan eğitim veri seti kümesi oldukça önemlidir. Doğal dil işlemede en çok karşılaşılan problemlerden birisi yeterli veri bulunamaması veya bulunan verilerin etiketsiz olmasıdır. Özellikle sınıflandırma problemlerinde belirli bir sınıftaki verinin azlığı sınıflandırmanın başarısını düşürmektedir. Bu çalışmada veri kümesinde bulunan eksik sınıfa ait metinlerin arttırılması amacı ile üretken rakip ağlar yöntemi kullanılmıştır. Haber metinleri üzerinde veri çoğalma işlemi gerçekleştirilmiştir. Elde edilen sonuçlar n-gram, destek vektör makinesi, TF-IDF ve lojistik regresyon gibi makine öğrenmesi teknikleriyle birlikte kullanılarak performansları değerlendirilmiştir. Sonuçlara göre üretken rakip ağların Türkçe metin üretimi için kullanılması sınıflandırma başarısını yaklaşık % 47 oranında arttırmıştır.

Keywords:

Production of Turkish texts with competitive network
2021
Author:  
Abstract:

The training data set used for the success of the forecast stage in machine learning methods is quite important. One of the most common problems in natural language processing is the lack of sufficient data or the lack of labelling of the available data. Especially in classification problems, the lack of data in a particular class reduces the success of classification. In this study, the method of producing competitive networks was used in order to increase the missing class texts found in the data set. Reproduction of data on the newspapers has been carried out. The results obtained have been evaluated by performance using machine learning techniques such as n-gram, support vector machine, TF-IDF and logistics regression. According to the results, the use of producing competitive networks for Turkish text production increased the classification success by about 47%.

Keywords:

Turkish Text Generation With Generative Adversarial Network
2021
Author:  
Abstract:

The training data set used for the success of the training phase in machine learning methods is very important. One of the most common problems in natural language processing is the lack of sufficient data or the unlabeled data. Especially in classification problems, the scarcity of data in a certain class reduces the success of the classification. In this study, generative adversarial network method was used in order to increase the texts belonging to the missing class in the data set. Data augmentation is performed on news texts. The results obtained were evaluated together with machine learning techniques such as n-grams, support vector machine, TF-IDF and logistic regression. According to the results, the use of generative adversarial network for Turkish text generation increased the classification success by approximately 47%.

Keywords:

Citation Owners
Information: There is no ciation to this publication.
Similar Articles










Avrupa Bilim ve Teknoloji Dergisi

Field :   Fen Bilimleri ve Matematik; Mühendislik

Journal Type :   Uluslararası

Metrics
Article : 3.175
Cite : 5.513
Avrupa Bilim ve Teknoloji Dergisi