Implementasi Metode Klasifikasi Dengan Algoritma Support Vector Machine Kernel Gaussian RBF Untuk Prediksi Partisipasi Pemilu Terhadap Demografi Kota Surabaya

Safitri, Dewi Eka; Fitrani, Arif Senja

doi:10.21927/ijubi.v5i1.2259

Cited by 1 publication

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…SVM merupakan algoritma dengan metode supervised learning yang menganalisa data dan mengenali pola [12]. Adapun keunggulan dari algoritma SVM, yaitu dapat digunakan untuk masalah klasifikasi dan regresi dengan kernel linier ataupun non-linier kernel yang dapat menjadi satu kemampuan algoritma pembelajaran untuk klasifikasi serta regresi dengan akurasi yang tinggi [13]. SVM dikembangkan oleh Vapnik, Guyon & Boser [14].…”

Section: Pendahuluanunclassified

Penerapan Metode Support Vector Machine (SVM) Dalam Klasifikasi Produktivitas Padi

Tohari,

Harini,

Yaqin

et al. 2023

JoSYC

View full text Add to dashboard Cite

Indonesia is one of the largest rice producing countries in the world. Almost 95% of the Indonesian population consumes rice as a mandatory staple food, so that every year the demand for rice increases along with the increase in population. East Java is known as the largest rice producing province in Indonesia. To optimize rice production, Central Java province can group rice producing cities or districts. This aims to see and find out cities or districts that have the potential to produce rice as well as find out areas that have less than optimal rice production. To see whether the need for rice in East Java province is met, it is necessary to predict rice productivity in the East Java region so that it can be used as a basis for efforts to increase rice yields for the next period. In this research, the Support Vector Machine (SVM) method was used to classify data for predicting crop yields in East Java province. The advantage of the SVM algorithm is that it can be used for classification and regression problems with linear kernels or non-linear kernels. The data used is agricultural statistical data obtained from the website jatim.bps.go.id. The data is then analyzed using a data mining process. The results of this research are in the form of a prediction pattern with a decision tree which can be used as a basis for predictions in estimating harvest results in the next period. From dividing 80% training data and 20% testing data, results were obtained with 80% accuracy when predicting category '0' (not on target), and 100% accuracy when predicting category '1' (on target). And overall, the classification model has an accuracy of 88%. The contribution to be achieved in this research is to provide ideas for data processing in the agricultural sector.

show abstract

Section: Pendahuluanunclassified