Claim Missing Document
Check
Articles

Found 4 Documents
Search
Journal : Journal of Applied Sciences, Management and Engineering Technology (JASMET)

Support Vector Machine optimization with fractional gradient descent for data classification Hapsari, Dian Puspita; Utoyo, Imam; Purnami, Santi Wulan
Journal of Applied Sciences, Management and Engineering Technology Vol 2, No 1 (2021)
Publisher : Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.jasmet.2021.v2i1.1467

Abstract

Data classification has several problems one of which is a large amount of data that will reduce computing time. SVM is a reliable linear classifier for linear or non-linear data, for large-scale data, there are computational time constraints. The Fractional gradient descent method is an unconstrained optimization algorithm to train classifiers with support vector machines that have convex problems. Compared to the classic integer-order model, a model built with fractional calculus has a significant advantage to accelerate computing time. In this research, it is to conduct investigate the current state of this new optimization method fractional derivatives that can be implemented in the classifier algorithm. The results of the SVM Classifier with fractional gradient descent optimization, it reaches a convergence point of approximately 50 iterations smaller than SVM-SGD. The process of updating or fixing the model is smaller in fractional because the multiplier value is less than 1 or in the form of fractions. The SVM-Fractional SGD algorithm is proven to be an effective method for rainfall forecast decisions.
Fractional Gradient Based Optimization for Nonlinear Separable Data Dian Puspita Hapsari; Muhammad Fahrur Rozi
Journal of Applied Sciences, Management and Engineering Technology Vol 3, No 1 (2022)
Publisher : Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.jasmet.2022.v3i1.2881

Abstract

The Support Vector Machine or SVM classifier is one of the machine learning algorithms whose job is to predict data. Traditional classifier has limitations in the process of training large-scale data, tends to be slow. This study aims to increase the efficiency of the SVM classifier using a fractional gradient descent optimization algorithm, so that the speed of the data training process can be increased when using large-scale data. There are ten numerical data sets used in the simulation that are used to test the performance of the SVM classifier that has been optimized using the Caputo type fractional gradient descent algorithm. In this paper, we use the Caputo derivative formula to calculate the fractional-order gradient descent from the error function with respect to weights and obtain a deterministic convergence to increase the speed of the Caputo type fractional-order derivative convergence. The test results show that the optimized SVM classifier achieves a faster convergence time with iterations and a small error value. For further research, the optimized SVM linear classifier with fractional gradient descent is implemented on the problem of unbalanced class data.
Support Vector Machine optimization with fractional gradient descent for data classification Dian Puspita Hapsari; Imam Utoyo; Santi Wulan Purnami
Journal of Applied Sciences, Management and Engineering Technology Vol 2, No 1 (2021)
Publisher : Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.jasmet.2021.v2i1.1467

Abstract

Data classification has several problems one of which is a large amount of data that will reduce computing time. SVM is a reliable linear classifier for linear or non-linear data, for large-scale data, there are computational time constraints. The Fractional gradient descent method is an unconstrained optimization algorithm to train classifiers with support vector machines that have convex problems. Compared to the classic integer-order model, a model built with fractional calculus has a significant advantage to accelerate computing time. In this research, it is to conduct investigate the current state of this new optimization method fractional derivatives that can be implemented in the classifier algorithm. The results of the SVM Classifier with fractional gradient descent optimization, it reaches a convergence point of approximately 50 iterations smaller than SVM-SGD. The process of updating or fixing the model is smaller in fractional because the multiplier value is less than 1 or in the form of fractions. The SVM-Fractional SGD algorithm is proven to be an effective method for rainfall forecast decisions.
HOSPITAL LENGTH OF STAY PREDICTION WITH ENSEMBLE LEARNING METHODE Dian Puspita Hapsari; Waras Lumandi; Arief Rachman
Journal of Applied Sciences, Management and Engineering Technology Vol 4, No 1 (2023)
Publisher : Institut Teknologi Adhi Tama Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31284/j.jasmet.2023.v4i1.4437

Abstract

The hospital length of stay (LoS) is the number of days an inpatient will stay in the hospital. LoS is used as a measure of hospital performance so they can improve the quality of service to patients better. However, making an accurate estimate of LoS can be difficult due to the many factors that influence it. The research conducted aims to predict LoS for treated patients (ICU and non-ICU) with cases of brain vessel injuries by using the ensemble learning method. The Random Forest algorithm is one of the ensembles learning methods used to predict LoS in this study. The dataset used in this study is primary data at PHC Surabaya Hospital. From the results of the simulations performed, the random forest algorithm is able to predict LoS in a dataset of treated patients (ICU and non-ICU) with cases of brain vessel injuries. And the simulation results show a type II error value of 0.10 while the value of type I error is 0.16.