PERBANDINGAN ALGORITMA SUPPORT VECTOR MACHINE DAN RANDOM FOREST UNTUK KLASIFIKASI PRODUK PALING DIMINATI PADA TOKO SEMBAKO MENTARI BERBASIS DATA TRANSAKSI HARIAN

ALDITO FITER VANESA REZA, NPM 2209100007 (2026) PERBANDINGAN ALGORITMA SUPPORT VECTOR MACHINE DAN RANDOM FOREST UNTUK KLASIFIKASI PRODUK PALING DIMINATI PADA TOKO SEMBAKO MENTARI BERBASIS DATA TRANSAKSI HARIAN. Skripsi thesis, Universitas Labuhanbatu.

	Text COVER.pdf Download (1MB)
	Text BAB I.pdf Download (416kB)
	Text BAB II.pdf Download (685kB)
	Text BAB III.pdf Download (621kB)
	Text BAB IV.pdf Download (562kB)
	Text BAB V.pdf Download (218kB)
	Text DAFTAR PUSTAKA.pdf Download (244kB)
	Text LAMPIRAN.pdf Download (560kB)

Abstract

Penelitian ini menganalisis dan membandingkan kinerja algoritma Support Vector Machine (SVM) dan Random Forest (RF) dalam mengklasifikasikan produk paling diminati di Toko Sembako Mentari, sebuah UMKM di sektor ritel kebutuhan pokok. Penelitian difokuskan pada data transaksi harian yang terbatas, dengan variabel utama jumlah pembelian, frekuensi transaksi, dan harga satuan. Metode yang digunakan mengikuti pendekatan CRISP-DM, meliputi pemahaman bisnis dan data, persiapan data, pemodelan, evaluasi, dan penerapan, dengan data transaksi enam bulan terakhir. Evaluasi kinerja algoritma menggunakan Accuracy, Precision, Recall, F1-score, dan AUC. Hasil penelitian menunjukkan bahwa kedua algoritma mampu melakukan klasifikasi dengan baik, namun SVM menunjukkan performa sedikit lebih unggul dibandingkan RF secara keseluruhan, dengan nilai accuracy 95,71%, precision kelas diminati 90,67%, recall 97,14%, F1-score 93,79%, dan AUC 98,96%, sementara RF memiliki nilai accuracy 95,47%, precision 90,07%, recall 97,14%, F1-score 93,47%, dan AUC 99,08%. Penelitian ini menegaskan bahwa penerapan machine learning dapat membantu UMKM meningkatkan efisiensi pengelolaan stok dan mendukung pengambilan keputusan bisnis berbasis data. Kata Kunci: Support Vector Machine (SVM), Random Forest (RF), Klasifikasi Produk, UMKM, Toko Sembako ================================================================================ This study analyzes and compares the performance of the Support Vector Machine (SVM) and Random Forest (RF) algorithms in classifying the most in demand products at Toko Sembako Mentari, a micro, small, and medium enterprise (MSME) operating in the basic retail sector. The research focuses on limited daily transaction data, with key variables including purchase quantity, transaction frequency, and unit price. The method follows the CRISP-DM approach, which includes business and data understanding, data preparation, modeling, evaluation, and deployment, using transaction data from the past six months. The performance of the algorithms is evaluated using Accuracy, Precision, Recall, F1-score, and AUC metrics. The results show that both algorithms perform well in classification tasks; however, SVM demonstrates slightly superior overall performance compared to RF, with an accuracy of 95.71%, precision (in-demand class) of 90.67%, recall of 97.14%, F1-score of 93.79%, and AUC of 98.96%. Meanwhile, RF achieves an accuracy of 95.47%, precision of 90.07%, recall of 97.14%, F1-score of 93.47%, and AUC of 99.08%. This study confirms that the application of machine learning can help MSMEs improve inventory management efficiency and support data driven business decision-making. Keywords: Support Vector Machine (SVM), Random Forest (RF), Product Classification, MSMEs, Grocery Store

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	Support Vector Machine (SVM), Random Forest (RF), Klasifikasi Produk, UMKM, Toko Sembako=============Support Vector Machine (SVM), Random Forest (RF), Product Classification, MSMEs, Grocery Store
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	27 Apr 2026 03:56
Last Modified:	27 Apr 2026 03:56
URI:	http://repository.ulb.ac.id/id/eprint/2169

Actions (login required)

View Item