Prediction of Addictive Substance Usage Risk by using Machine Learning Methods

Murad Naghiyev; İclal Çetin Taş

Makale detayları

Prediction of Addictive Substance Usage Risk by using Machine Learning Methods

Yazarlar:

Murad Naghiyev | İclal Çetin Taş

Yayıncı:

Bolu Abant Izzet Baysal University

Ana Başlıklar:

Yazılım Mühendisliği (Diğer)

Dergi:

Abant Sağlık Bilimleri ve Teknolojileri Dergisi Vol.3 Issue 2, pp.1-13

Belge Tipi:

Makale

Yayın Tarihi:

8/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001297430

pISSN:

2791-8904

Anahtar kelimeler:

Özet:

 With the rapid development of technology, data-driven studies are increasingly being carried out in various fields. Machine learning algorithms are commonly used in these studies, and their use for early diagnosis and detection in the medical field is becoming more widespread. Machine learning techniques help in the early diagnosis of diseases, treatment planning, and making more effective and accurate decisions in disease management in the medical field. This leads to increased knowledge about patients health conditions, thereby improving the quality and efficiency of healthcare services. The aim of this study is to predict which other addictive substances individuals who use addictive substances may use in the future using machine learning methods. KNN knearest neighbor, Gaussian SVM support vector machine, and decision tree methods were used in the application on the dataset of addictive substance use, and the results obtained were examined. The highest success rate was obtained with the Gaussian SVM method, which was 90.60.

 With the rapid development of technology, data-driven studies are increasingly being carried out in various fields. Machine learning algorithms are commonly used in these studies, and their use for early diagnosis and detection in the medical field is becoming more widespread. Machine learning techniques help in the early diagnosis of diseases, treatment planning, and making more effective and accurate decisions in disease management in the medical field. This leads to increased knowledge about patients health conditions, thereby improving the quality and efficiency of healthcare services. The aim of this study is to predict which other addictive substances individuals who use addictive substances may use in the future using machine learning methods. KNN knearest neighbor, Gaussian SVM support vector machine, and decision tree methods were used in the application on the dataset of addictive substance use, and the results obtained were examined. The highest success rate was obtained with the Gaussian SVM method, which was 90.60.

 With the rapid development of technology, data-driven studies are increasingly being carried out in various fields. Machine learning algorithms are commonly used in these studies, and their use for early diagnosis and detection in the medical field is becoming more widespread. Machine learning techniques help in the early diagnosis of diseases, treatment planning, and making more effective and accurate decisions in disease management in the medical field. This leads to increased knowledge about patients health conditions, thereby improving the quality and efficiency of healthcare services. The aim of this study is to predict which other addictive substances individuals who use addictive substances may use in the future using machine learning methods. KNN knearest neighbor, Gaussian SVM support vector machine, and decision tree methods were used in the application on the dataset of addictive substance use, and the results obtained were examined. The highest success rate was obtained with the Gaussian SVM method, which was 90.60.

 Teknolojinin hızlı gelişimiyle birlikte, farklı alanlarda veri odaklı çalışmalar giderek artmaktadır. Bu çalışmalarda, makine öğrenme algoritmaları sıklıkla kullanılmakta ve özellikle tıp alanında erken tanı ve teşhis amaçlı kullanımı giderek yaygınlaşmaktadır. Makine öğrenme teknikleri, tıp alanında hastalıkların erken teşhis edilmesi, tedavi planlaması ve hastalık yönetiminde daha etkili ve doğru kararlar alınmasına yardımcı olmaktadır. Bu sayede hastaların sağlık durumları hakkında daha fazla bilgiye sahip olunarak, sağlık hizmetlerinin kalitesi ve verimliliği artırılmaktadır. Bu çalışmanın amacı, bağımlılık yapıcı madde kullanan kişilerin ileride başka hangi bağımlılık yapan maddeleri kullanma risklerinin olabileceğini makine öğrenmesi yöntemleri ile tahminlemektir. Bağımlılık yapıcı madde kullanımlarına ilişkin veri setinde yapılan uygulamada KNN k-nearest neighbour, Gaussian SVM support vector machine, karar ağacı DT-decision tree yöntemleri kullanılmış olup elde edilen sonuçlar incelendiğinde en yüksek başarı 90,60 ile Gaussian SVM yönteminden elde edilmiştir.

Makale detayları

Prediction of Addictive Substance Usage Risk by using Machine Learning Methods

Yazarlar:

Murad Naghiyev | İclal Çetin Taş

Yayıncı:

Bolu Abant Izzet Baysal University

Ana Başlıklar:

Yazılım Mühendisliği (Diğer)

Dergi:

Abant Sağlık Bilimleri ve Teknolojileri Dergisi Vol.3 Issue 2, pp.1-13

Belge Tipi:

Makale

Yayın Tarihi:

8/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001297430

pISSN:

2791-8904

substance addiction

machine learning

prediction

KNN

SVM

decision tree.madde bağımlılığı

makine öğrenmesi

tahmin

KNN

SVM

karar ağacı.

Sayı

Makale detayları

Yazarlar: Murad Naghiyev | İclal Çetin Taş

Yayıncı: Bolu Abant Izzet Baysal University

Ana Başlıklar: Yazılım Mühendisliği (Diğer)

Dergi: Abant Sağlık Bilimleri ve Teknolojileri Dergisi Vol.3 Issue 2, pp.1-13

Belge Tipi: Makale

Yayın Tarihi: 8/2023

Belge Dili: Türkçe

Document ID (Org No.): 2001297430

pISSN: 2791-8904

Sayı

Yazarlar:

Murad Naghiyev | İclal Çetin Taş

Yayıncı:

Bolu Abant Izzet Baysal University

Ana Başlıklar:

Yazılım Mühendisliği (Diğer)

Dergi:

Abant Sağlık Bilimleri ve Teknolojileri Dergisi Vol.3 Issue 2, pp.1-13

Belge Tipi:

Makale

Yayın Tarihi:

8/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001297430

pISSN:

2791-8904