Claim Missing Document
Check
Articles

Found 2 Documents
Search

Bank Kalbar's Educational Programs for MSME Revitalization and Community Economic Development Didi Haryono; Gumilar Rusliwa Somantri; Hanif Saha Gafur; Muhammad Syaroni Rofii; Ajeng Hidayati
Nomico Vol. 1 No. 5 (2024): Nomico-June
Publisher : PT. Anagata Sembagi Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62872/sjpxb984

Abstract

Micro, Small, and Medium Enterprises (MSMEs) are crucial to Indonesia's economy, contributing approximately 60.5% to the Gross Domestic Product (GDP) and employing about 96.9% of the national workforce. Despite the large number of MSMEs and their potential GDP contribution, there is still room for improvement in exports. The government has taken various measures to support MSME development, including policies that encourage digital business growth and education-based training for MSMEs. In West Kalimantan Province (Kalbar), MSMEs play a vital role in regional economic development. Despite their potential, they face several challenges such as access to capital, financial management, and market access. The regional government and the Bank of West Kalimantan have collaborated to revitalize MSMEs post-COVID-19 through various programs, credit schemes, and educational initiatives or training sessions. This study employs both qualitative and quantitative methods with a case study approach. The research results indicate that MSMEs significantly contribute to West Kalimantan's economic development, despite being impacted by the pandemic. The internal and external revitalization strategies implemented by Bank Kalbar have assisted MSMEs in recovery and adapting to digital-based business models. The study concludes that MSMEs in West Kalimantan have the potential to continue growing and thriving as independent economic entities in a digital-based business environment. The revitalization of MSMEs is a result of collaboration between MSME actors, the government, and banking institutions to achieve regional economic stability and recovery.
Klasifikasi Multi-Kelas Serangan Jaringan Menggunakan Algoritma Random Forest pada Dataset NSL-KDD Ajeng Hidayati
JIMAD : Jurnal Ilmiah Multidisiplin Vol. 4 No. 1 (2026): JIMAD : Jurnal Ilmiah Multidisiplin (October)
Publisher : Asosiasi Guru dan Dosen Seluruh Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59585/jimad.v4i1.1680

Abstract

Serangan siber pada jaringan komputer terus meningkat baik dari sisi volume maupun kompleksitas, sehingga dibutuhkan sistem deteksi intrusi yang mampu mengklasifikasikan jenis serangan secara akurat. Penelitian ini bertujuan membandingkan kinerja algoritma Random Forest, Decision Tree, dan Naive Bayes dalam mengklasifikasikan lalu lintas jaringan ke dalam lima kategori, yaitu Normal, Denial of Service (DoS), Probe, Remote to Local (R2L), dan User to Root (U2R). Metode yang digunakan adalah eksperimen klasifikasi multi-kelas pada dataset benchmark NSL-KDD, meliputi tahap praproses data, pelatihan model, dan evaluasi menggunakan metrik akurasi, presisi, recall, dan F1-score pada data uji independen (KDDTest+). Hasil penelitian menunjukkan Decision Tree memperoleh akurasi tertinggi (76,15%), sedikit di atas Random Forest (75,00%), sementara Naive Bayes jauh tertinggal (28,71%). Kedua model berbasis pohon keputusan mampu mendeteksi kategori Normal dan DoS dengan baik, namun mengalami kesulitan signifikan mendeteksi serangan R2L dan U2R akibat ketidakseimbangan kelas pada data latih serta adanya pola serangan baru pada data uji. Analisis feature importance menunjukkan src_bytes, same_srv_rate, dan flag sebagai fitur paling berpengaruh. Temuan ini menegaskan perlunya strategi penyeimbangan data untuk meningkatkan deteksi serangan minoritas pada sistem keamanan jaringan berbasis machine learning