Beschreibung
The aim of this work is to raise patient's awareness by estimating the blood sugar levels in order to avoid miscalculations/misinterpretations of blood sugar measurements due to human error. In order make a prediction on the patient's blood sugar levels, the medication levels, food consumption and post blood sugar measurements of several patients are used as the dataset. An extensive set of features is extracted from this dataset in order to create a blood sugar level corpus, which is used to train a machine learning model for predicting the future blood sugar measurements. These models are then tested on the same patient set, for evaluating the performance of the proposed technique. In this work, a wide range of new contributions is also provided such as comparison of different machine learning algorithms with respect to their prediction quality and analysis on a large range of attributes for predicting blood sugar levels such as insulin dose, food consumption and activity rates.
Autorenporträt
Adil Kondiloglu, lecturer at the Giresun University, Turkey. Master's degree in Computer Engineering from University of Turkish Aeronautical Association. Also he is currently studying Doctorate Program of Computer Engineering at Karadeniz Technical University. The author is interested in artificial intelligence, machine learning and web programming
Herstellerkennzeichnung:
OmniScriptum SRL
Str. Armeneasca 28/1, office 1
2012 Chisinau
MD
E-Mail: info@omniscriptum.com
