Erkan Özhan

Prof. Dr. Erkan Özhan

Computer Engineering, Tekirdag Namik Kemal University, Tekirdağ, Turkey (Türkiye)
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 May 2017 in Wireless Personal Communications
Verified
Journal Article
Closed access
PDF file
EUErdem Ucar
Erdem Ucar
Erkan ÖzhanErkan Özhan
Erkan Özhan

28 October 2021 in Vision: The Journal of Business Perspective
Verified
Journal Article
Open access
PDF file
SOSeniz Ozhan
Seniz Ozhan
Erkan ÖzhanErkan Özhan
Erkan Özhan
GPGamze Yakar Pritchard
Gamze Yakar Pritchard
When the successful businesses of today are examined, it is seen that the main factor in their success is the value they give to the customers rather than the production power. One of the most important factors in ensuring customer satisfaction and loyalty is customer orientation (CO). In this study, it is aimed to investigate the perceived management and customer support for corporate social responsibility, the identification of the employees with the business and the customers and its effect on CO. Another aim of the study is to obtain a model that classifies employee–customer identification (ECI)-CO levels for employees by using artificial intelligence methods not used in previous studies. The research data were obtained from salesperson working in shopping malls in Istanbul. Hypothesis testing with structural equation modelling (SEM) has shown that perceived management and customer support for corporate social responsibility have an impact on employee identification with the business and customers. It has been observed that ECI affects CO, while organizational identification has no significant effect on CO. The structural equation modelling and artificial intelligence findings have empirically demonstrated that high accuracy practical classification models can be obtained and used to detect and solve different marketing problems.

31 August 2020 in European Journal of Science and Technology
Verified
Journal Article
Open access
PDF file
Erkan ÖzhanErkan Özhan
Erkan Özhan
Sera gazlarının atmosferdeki miktarı gün geçtikçe artmaktadır. Bu artışın başta küresel ısınma olmak üzere neden olduğu çok sayıda olumsuz etki ortaya çıkmaktadır. Geleceğe dönük sera gazı emisyonunun tahminlenmesi özellikle karar alıcılar ve CO2 salınımında payı olan sektörler açısından bakıldığında bu salınımın azaltılması ve alternatif kaynakların aranması için cesaret verici olabilir. Zaman serileri zaman düzleminde düzenli olarak belirli aralıklarla elde edilmiş verilerin literatürdeki adıdır ve bu serilerin analizinin nasılını inceleyen süreçlere ise zaman serisi analizi denir. Araştırmada Türkiye’ye ait sera gazı emisyonu (CO2 eşdeğeri) değerlerini içeren Dünya Bankası veri tabanındaki 55 yıllık verileri içeren veri seti kullanımıştır. Bu veri seti içerisinden yapay sinir ağları ve üstel düzleştirme yöntemleri ile faydalı örüntüler elde edilmesi amaçlanmıştır. Analizler için zaman serisi formatına dönüştürülen veri seti daha sonra eğitim ve test verisi olarak iki bölüme ayrılmıştır. Zaman serisi tipindeki eğitim verileri üstel düzleştirme yöntemini temel alan Holt’un lineer trend modeli ve yapay zekanın alt dallarından biri olan yapay sinir ağları (YSA) ile analizi edilmiştir. Bu analizler sonucunda ortaya çıkan modellere göre eğitim ve test verileri üzerinden tahmin modelleri elde edilmiştir. YSA’nın ve Holt’un lineer trend yönteminin test verileri için ortaya koyduğu tahminler ile modelleri değerlendirmek için RMSE, MAPE gibi değerlendirme metrikleri elde edilmiştir. Bu değerlere göre iki model karşılaştırılmış ve en az hata oranına sahip modelin YSA olduğu tespit edilmiştir. Çalışmada elde edilen bulgulara göre YSA 0.1607’lik RMSE değeri ile, Holt’un liner trend yöntemine göre çok daha az hata oranına sahiptir. YSA’nın daha doğru tahminler yapacağı bulgusu elde edildikten sonra bu yöntemin önerdiği model kullanılarak 2021 yılına kadar tahminler gerçekleştirilmiştir. Model Türkiye için 2021 yılı sera gazı eşdeğeri CO2 emisyonunu 366,3972 milyon ton olarak tahminlemiştir. Araştırmada görülen bir diğer sonuç ise CO2 emisyonunun dalgalı bir seyir izlediği ancak genel olarak yükselme eğiliminde olduğudur.

18 November 2020 in IEEE Access
Verified
Journal Article
Open access
EUErdinç Uzun
Erdinç Uzun
Erkan ÖzhanErkan Özhan
Erkan Özhan
HAHayri Volkan Agun
Hayri Volkan Agun
TYTarik Yerlikaya
Tarik Yerlikaya
HBhalil nusret bulus
halil nusret bulus
Web pages contain irrelevant images along with relevant images. The classification of these images is an error-prone process due to the number of design variations of web pages. Using multiple web pages provides additional features that improve the performance of relevant image extraction. Traditional studies use the features extracted from a single web page. However, in this study, we enhance the performance of relevant image extraction by employing the features extracted from different web pages consisting of standard news, galleries, video pages, and link pages. The dataset obtained from these web pages contains 100 different web pages for each 200 online news websites from 58 different countries. For discovering relevant images, the most straightforward approach extracts the largest image on the web page. This approach achieves a 0.451 F-Measure score as a baseline. Then, we apply several machine learning methods using features in this dataset to find the most suitable machine learning method. The best f-Measure score is 0.822 using the AdaBoost classifier. Some of these features have been utilized in previous web data extraction studies. To the best of our knowledge, 15 new features are proposed for the first time in this study for discovering the relevant images. We compare the performance of the AdaBoost classifier on different feature sets. The proposed features improve the f-Measure by 35 percent. Besides, using only the cache feature, which is the most prominent feature, corresponds to 7 percent of this improvement.

1 September 2018
Verified
Conference Paper
Closed access
Erkan ÖzhanErkan Özhan
Erkan Özhan
EUErdinç Uzun
Erdinç Uzun
The Web is an invaluable source of data stored on web pages. These data are contained in HTML layout elements of a web page. It is a crucial issue to extract data automatically from a web page. In this study, a dataset, which is annotated with seven different layouts including main content, headline, summary, other necessary layouts, menu, link, and other unnecessary layouts, is used. Then, 49 different features are computed from these layouts. Finally, we compare the different classification methods for evaluating the performance of these methods in layout prediction. The experiments show that the Random Forest classifier achieves a high accuracy of 98.46%. Thanks to this classifier, the prediction of link layout has a higher performance (approximately 0.988 f-Measure) according to the performance of the prediction of other layouts. On the other hand, the prediction of the summary layout has the worst performance with about 0.882 f-Measure.

1 September 2018
Verified
Conference Paper
Closed access
EUErdinç Uzun
Erdinç Uzun
Erkan ÖzhanErkan Özhan
Erkan Özhan
The user reviews in web pages can provide useful information about the content of the web page for text processing applications. Automatically extracting data from a web page is a crucial process for these applications. One of the used methods in this process is to construct a learning model with an appropriate classification method using features that are derived from data. However, some features can be either redundant or irrelevant for this model. In this study, an imbalanced dataset including 47 shallow text features obtained from web pages is utilized for extracting of the user reviews. Then, various well-known feature selection techniques are applied to reduce the number of these features. The effects of this reduction on the classification methods are also examined. The experimental results indicate that approximately half of the features are sufficient for the classification task. Additionally, the AdaBoost classifier gives the best results concerning precision of about 0.930 for the review layout prediction.