Hüseyin TURGUT | Mehmet Akif Ersoy University (original) (raw)

Uploads

Makale - Papers by Hüseyin TURGUT

Converting to information by analyzing the data is very important for businesses in today's knowl... more Converting to information by analyzing the data is very important for businesses in today's knowledge
economy. In this study, data of the Vefa Gıda Pazarlama Company are used. This data consists of buying
and selling between September 1, 2012 and February 17, 2015 dates. 8 successful data mining algorithms
were tested during the study period. The aim of this study which of the data mining method is most suitable
for date analyzing when the structure of the data is missing or insufficient in business database. More than
78 000 data was obtained. The transaction is made by 63958 lines of data. 80% of the data are used as
training data and 20% was used as the test data. In this study, the most appropriate model logitboost
classification algorithm was determined for data base with missing data.

bildiri by Hüseyin TURGUT

"Objective: Today, quality of life and increased health consciousness also reduced the use of hum... more "Objective: Today, quality of life and increased health consciousness also reduced the use of human resources leads to the prolongation of human life. Lengthening human life expectancy increases in the incidence of dementia and many other diseases. To assist physicians in the diagnosis of the disease and doctors in this field applied to data mining process in order to save time.

Methods: The research for the Suleyman Demirel University, Faculty of Medicine, used data from patients admitted to the neurology clinic. All stages of diagnosis and diagnostic data acquisition and data necessary for the evaluation of the processes studied. Applied by the physician on the patient in order to simulate the process of data mining method was used. As part of data mining optimized data with each other, has been filtered. Of the algorithms used to clean the data obtained from the WEKA software.

Conclusion: The study tested the algorithm and as a result the appropriate 67 and over 95% accuracy algorithm described in a total of 9. An algorithm is applied which have a value of 100% accuracy.

Discussion: Research and applications, data mining can help in the process of diagnosing health physician, a physician may provide educational support to the candidate reveals. The value of accuracy and clarity of medical data is less than the amount of data that was caused by the need to research and practice understood."

Papers by Hüseyin TURGUT

Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 2018

Konut, barınma açısından vazgeçilmez bir araç olmasına rağmen son yıllarda dünyada yaşanan ekonom... more Konut, barınma açısından vazgeçilmez bir araç olmasına rağmen son yıllarda dünyada yaşanan ekonomik, kültürel ve teknolojik değişimlerin bir sonucu olarak konut, barınma temel ihtiyacını karşılama amacı dışında, gelecek için güvence kaynağı, yaşam tarzı ve zenginliğin en önemli göstergelerinden biri olmakla kalmamış toplum içindeki saygınlığın da ifadesi olmuştur. Bu durum konutun heterojen bir mal olmasına ve birçok özelliği içinde barındırmasına neden olmaktadır. Konutun sahip olduğu farklı özellikleri konutun piyasa değeri üzerinde ne derece etkin olduğu mikro iktisat teorisine dayalı hedonik fiyat modeli ile tahmin edilmektedir. Çalışmanın amacı, hanehalkının kiralık konut talebini etkileyen faktörleri Veri Madenciliği süreç ile analiz etmektir. 2015 yılı Hanehalkı Bütçe Anket verilerinin kullanıldığı çalışmada, 2341 hane verisi çalışmaya kaynaklık etmiş ve 49 farklı veri başlığı temel alınmıştır. Çalışma sonucunda, Veri Madenciliği yöntemlerinden Karar Ağacı Algoritmasının en iyi sonuç verdiği tespit edilmiştir.

Journal of the Turkish Chemical Society, Section A: Chemistry, 2018

Nature and life include many mysterious events, behaviors and format within themselves. There is ... more Nature and life include many mysterious events, behaviors and format within themselves. There is harmony between the environmental conditions, behavior and forms of all living organism. Computer science, especially data and information science, is based on the structure or behavior of living things in the creation of many artificial intelligence algorithms by examining this attitude of life. The rapid progress of the developing artificial intelligence and information technology has increased the data and hidden data in our lives many times and has tried to solve (1). Artificial intelligence has examined many areas or environments and has developed approaches based on it. Expert systems, artificial neural networks, genetic algorithms, inductive learning, explanation-based learning, similarity-based learning, common sense information processing, model based reasoning, model based reasoning, rational protection mechanism, distributed artificial intelligence, natural language processing, chaos theory, logic programming are the artificial intelligence algorithms used for these approaches (2). Among artificial intelligence algorithms; The ant colony algorithm imitates the behavior and direction of ants, and artificial neural networks imitates the behavior and functions of neurons in the nervous system and genetic algorithms imitates the theoretical form of genetic science (3, 4). Many algorithms such as these algorithms are based on the vital form and behavior of living things. The purpose of this review is the relations between the mentioned algorithms and the living science are examined.

Conclusion: The study tested the algorithm and as a result the appropriate 67 and over 95% accuracy algorithm described in a total of 9. An algorithm is applied which have a value of 100% accuracy.

Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 2018

Journal of the Turkish Chemical Society, Section A: Chemistry, 2018