Abstract:Emergency departments are one of the most important units in the hospital where there are special units and many problems. At the beginning of these problems, emergency services are crowded and urgent patient care planning is difficult. The applications such as triage system are used for these problems. However it is known that such applications do not fully solve these problems. In this study, a fuzzy logic based clinical decision support system (CDSS) was developed for the classification of emergency patient… Show more
The need for shelter has been among the most basic needs of people since ancient times. Houses used to meet people's accommodation needs differ in terms of geography, materials used and etc. These differences can be categorized under five categories: structural, security, heating, hardware and proximity. Each category has its own subgroups. Existing home purchasing systems list the postings taking into account the presence or absence of residential properties. However, it is not appropriate for buyers to mean the properties only with logic 0 or logic 1 values. To solve such problems, a model is created where buyers give a value between-1 and +1 to each category and the attributes under it. In the list created using this model, each house has a score and is shown to the buyer in order. Thus, a more realistic list is created that will assist the buyer in making decisions, compared to existing systems. In this study, a total of 34844 houses in 39 districts of Istanbul province were examined and a model formed from 29 criteria was developed. The developed model was tested for 40 samples and the adequacy of the decision support system was demonstrated. In addition, an alternative decision support system offered exclusively to the person and the ranking table is shown.
The need for shelter has been among the most basic needs of people since ancient times. Houses used to meet people's accommodation needs differ in terms of geography, materials used and etc. These differences can be categorized under five categories: structural, security, heating, hardware and proximity. Each category has its own subgroups. Existing home purchasing systems list the postings taking into account the presence or absence of residential properties. However, it is not appropriate for buyers to mean the properties only with logic 0 or logic 1 values. To solve such problems, a model is created where buyers give a value between-1 and +1 to each category and the attributes under it. In the list created using this model, each house has a score and is shown to the buyer in order. Thus, a more realistic list is created that will assist the buyer in making decisions, compared to existing systems. In this study, a total of 34844 houses in 39 districts of Istanbul province were examined and a model formed from 29 criteria was developed. The developed model was tested for 40 samples and the adequacy of the decision support system was demonstrated. In addition, an alternative decision support system offered exclusively to the person and the ranking table is shown.
Günümüzde şirketlerin artan rekabet şartlarından dolayı bilişim teknolojilerine olan ihtiyaç her geçen gün artmaktadır. Bilişim teknolojilerinin (BT) sorunsuz, hızlı ve güvenilir çalışması yapılan yatırım kadar önemlidir. Şirketlerde bulunan bilişim teknoloji(BT) departmanları yapılacak yatırımların planlanmasından, mevcut sistemin sorunsuz çalışmasından ve oluşabilecek sorunlara hızlı çözümler üretilmesinden sorumludurlar. Şirket çalışanlarının BT departmanına sorunları iletmesi, süreci takip etmesi ve raporlandırması amacıyla bilgi işlem takip programları kullanmaktadır. Böylece sorunların ve çözüm sürecinin yönetimi, bilgilendirmesi ve raporlandırılması profesyonel ve kolay yapılabilmektedir. Sorunların hızlı çözülmesi ve çözüm süresinin bilinmesi çalışanlarının zamanı iyi kullanmasını sağlayacak ve belirsiz bekleme süresinin yaratacağı olumsuz etkileri ortadan kaldıracaktır. Bu çalışmada bir şirketin kullanıcıdan gelen talepleri çözme süresi makine öğrenmesi yöntemiyle tahmin edilmiştir. Bunun için kullanıcılardan gelen 2320 talep; departman, destek türü, sorumlu ve kategori olarak kayıt altına alınmaktadır. Sonuçtaki destek süreleri için bir sınıf yapısı ile test edilmiştir. Harcanan süreler 0-10 dakikadan başlayıp 90 dakika ve daha fazlası şeklinde 10’ar dakika arayla on adet sınıfa ayrılmıştır. Bu veri setleri üzerinde makine öğrenme yöntemleri kullanılmıştır. Gerçekleştirilen testler sonucunda en iyi sonuç Destek Vektör Makineleri (Support Vector Machine-SVM) metodu kullanılarak tasarlanan model ile, eğitim başarısı %99.82, test başarısı ise %93.11 olarak ölçülmüştür. Bu sistem sayesinde kullanıcının bekleme süresi ve sorunun ortalama çözüm süresi tahmin edilmektedir.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.