TY  - JOUR
TI  - The Role of Machine Learning in Productivity: A Case Study of WineQuality Prediction
AB  - Machine learning has been used in many areas in recent years and has achieved quite successful results. Machine learning methods havebeen used from healthcare to driverless vehicles, it might also play a big role to increase productivity in the production sector. In thisstudy, we have compared the performance of some machine learning strategies on an abnormally distributed dataset. Any machinelearning methods can be easily applied to normally distributed data sets. However, it is necessary to alter the theoretical structure of thealgorithm or data transformation process while a dataset is abnormally distributed. In this regard, three different methodologies arecompared in this study. Initially, Support Vector Machines, which are often used in the literature, is used. Besides, Weighted SupportVector Machines, which is the revised version of the Support Vector Machines to produce successful results in abnormally distributeddata sets. Finally, the Synthetic Minority Oversampling Technique (SMOTE) is applied, and the distribution of the dataset wasartificially changed to normal distribution. Three techniques are compared in terms of sensitivity, specificity, precision, prevalence, F 1 score, and G-Mean evaluation criteria. Based on the results of the methods, Weighted Support Vector Machines produced the mostsuccessful results according to the chosen evaluation criteria.
AU  - ÜNLÜ, RAMAZAN
DO  - 10.31590/ejosat.773736
PY  - 2020
JO  - Avrupa Bilim ve Teknoloji Dergisi
VL  - 0
IS  - 20
SN  - 2148-2683
SP  - 280
EP  - 286
DB  - TRDizin
UR  - http://search/yayin/detay/466075
ER  -