Komparasi algoritme support vector machine (svm) dan c4.5 untuk klasifikasi penentuan lahan pertanian tanaman pangan

Ridho, Aji Nur Kusuma (2020) Komparasi algoritme support vector machine (svm) dan c4.5 untuk klasifikasi penentuan lahan pertanian tanaman pangan. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

Preview	Text ABSTRAK.pdf - Accepted Version Download (203kB) | Preview
Preview	Text ABSTRACT.pdf - Accepted Version Download (202kB) | Preview
	Text COVER.pdf - Accepted Version Download (1MB)
	Text BAB I.pdf Download (330kB)
Preview	Text BAB II.pdf - Accepted Version Download (915kB) | Preview
	Text BAB III.pdf - Accepted Version Download (743kB)
	Text BAB IV.pdf - Accepted Version Restricted to Registered users only Download (554kB)
Preview	Text BAB V.pdf - Accepted Version Download (203kB) | Preview
Preview	Text DAFTAR PUSTAKA.pdf - Accepted Version Download (324kB) | Preview

Abstract

Department of Agriculture and Food Security (DPKP) Banyumas to determine soil fertility in determining agricultural land for food crops is still done manually in the sense that officers conduct field surveys directly to observe local farmers regarding soil fertility. The Department of Agriculture and Food Security (DPKP) obtains data on agricultural land use based on the harvest productivity of each sub-district only in the form of yield reports without further data processing. In the field of data mining an algorithm was developed that could classify the determination of agricultural land for food crops to predict agricultural land that would be used for food crops, including SVM and C4.5. SVM is a supervised learming algorithm that classifies classes using hyperplane. SVM has the advantage of the high level of accuracy produced. C45 algorithm is an algorithm that forms a decision tree with decision rules. The C4.5 algorithm has good advantages for using numeric and discrete data types. Data collection in this study was obtained from the Department of Agriculture and Food Security, the Central Statistics Agency, the Regional Research and Development Agency and the Banyumas Regency Water Resources Management Agency which was integrated into a dataset totaling 27 sub-district records and 14 parameters. The results of the accuracy value on the C4.5 algorithm were 77.77% and the SVM algorithm 72.22% using the confusion matrix. C4.5 algorithm gets higher accuracy than SVM because in its classification stage, C4.5 processes attribute data one by one and is good for processing datasets that have a small number of records. In contrast to SVM which classifies in general, the scope is wider and better for processing with a large number of records. Key words : Agriculture, C4.5, Classification, Comparasion Of Algorithms, SVM.

Actions (login required)

View Item