Claim Missing Document
Check
Articles

Perbaikan Peramalan Produksi Padi di Kabupaten Kendal dengan Menggunakan Metode Support Vector Machine (SVM) Andita, Ayu; Sulistijanti, Wellie
Prosiding University Research Colloquium Proceeding of The 7th University Research Colloquium 2018: Mahasiswa (student paper presentation)
Publisher : Konsorsium Lembaga Penelitian dan Pengabdian kepada Masyarakat Perguruan Tinggi Muhammadiyah 'Aisyiyah (PTMA) Koordinator Wilayah Jawa Tengah - DIY

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Sektor pertanian memegang peranan penting bagi perekonomiandaerah sebagai peningkatan produksi bahan pangan khususnya bahanmakanan pokok bagi kehidupan manusia, tentunya masyarakat diKabupaten Kendal. Pada Tahun 2015, produksi padi di KabupatenKendal menempati peringkat 18 se-Jawa Tengah, Dinas PertanianKendal sangat memperhatikan perkembangan produksi padi sehinggamemerlukan perencanaan dalam peramalan yang harus dilandasidengan kekuatan model dan parameter yang signifikan. Penelitian inimenggunakan data dari bulan Januari 2013- Januari 2016, 70% dari48 data sebagai data training dan 30% data testing. Peramalan yangbaik memiliki nilai Mean Square Error (MSE) yang kecil. Peramalanproduksi padi dengan metode Seasonal Autoregressive IntegratedMoving Average (SARIMA) diperoleh MSE training sebesar 0.043918dan MSE testing 77.118.361,62, terlihat bahwa nilai MSE antarakeduanya sangat jauh berbeda. Dengan metode Support VektorMachine (SVM) diperoleh MSE training sebesar 0,14 dan MSE testingsebesar 0,57. Terlihat bahwa nilai MSE yang dihasilkan sangat dekattidak jauh berbeda dengan keduanya. Inilah kelebihan metode yangtelah diusulkan oleh peneliti dengan menggunakan Metode SVMmerupakan metode yang baik untuk meramalkan produksi padi diKabupaten Kendal, karena nilai MSE training dan MSE testing yangkecil, tidak memerhatikan parameter yang signifikan dan dapatmenanggani permasalahan linier ataupun non linier tanpamemerhatikan pola data.
Peramalan Kurs Dolar Amerika Serikat dan Riyal Arab Saudi Terhadap Rupiah dengan Neural Network Conjugate Gradient Polak Ribiere Susasimy, Lita Citra Dewi; Sulistijanti, Wellie
Prosiding University Research Colloquium Proceeding of The 14th University Research Colloquium 2021: Bidang Ekonomi dan Bisnis
Publisher : Konsorsium Lembaga Penelitian dan Pengabdian kepada Masyarakat Perguruan Tinggi Muhammadiyah 'Aisyiyah (PTMA) Koordinator Wilayah Jawa Tengah - DIY

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Pandemi Covid-19 yang terjadi sejak 2 Maret 2020 berdampak pada pertumbuhan ekonomi di Indonesia. Indikator terpenting dalam perekonomian adalah nilai tukar mata uang. Dolar Amerika (USD) merupakan acuan nilai tukar Rupiah (IDR) karena memegang peranan penting dalam transaksi perdaganganan internasional sedangkan Riyal Arab Saudi (SAR) sebagai nilai tukar SAR/IDR karena masyarakat Indonesia melaksanakan ibadah haji dan umroh ke tanah suci Mekkah. Peramalan dilakukan untuk mengetahui kenaikan dan penurunan kurs Dolar dan kurs Riyal terhadap Rupiah yang akan datang. Penelitian ini menggunakan data kurs Dolar dan Riyal bulan Januari-Mei tahun 2021 dari laman Investing.com. Metode peramalan Neural Network dengan algoritma Conjugate Gradient-Backpropagation Polak Ribiere (TrainCGP) algoritma yang arah pencariannya berdasar pada arah konjugasi dan lebih cepat konvergen. Nilai MSE yang didapat pada penelitian ini yaitu untuk data training kurs Dolar terhadap Rupiah sebesar 0,0009 dan data testing sebesar 0,1940 dengan arsitektur jaringan terbaik yaitu 7-9-1. Kemudian diperoleh hasil peramalan tanggal 02 Juni-10 Juni 2021 kurs Dolar terhadap Rupiah melemah pada 04 Juni seharga Rp.14.364,- dan menguat pada 10 juni seharga Rp.14.191,-. Nilai MSE untuk data training kurs Riyal terhadap Rupiah sebesar 0,0009 dan data testing sebesar 0,6322 dengan arsitektur jaringan terbaik yaitu 7-16-1. Kemudian diperoleh hasil peramalan tanggal 02 Juni-10 Juni 2021 kurs Riyal terhadap Rupiah melemah pada 03 juni seharga Rp.3.830,- dan menguat pada 08 Juni seharga Rp.3.768,-.
Pendampingan Pengelolaan Data Sosial-Ekonomi Berbasis Partisipasi Masyarakat bersama Yayasan Sahabat Yatim Sulistijanti, Wellie; Nasikhin, Muh; Wijayaningrum, Taswati Nova; Jayus, Jayus; Cahyani, Desy Eki; Azizah, Dzahari Alikharimah
Abdimas Mandalika Vol 5, No 1 (2025): November
Publisher : Universitas Muhammadiyah Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31764/am.v5i1.34775

Abstract

Abstract:  A participatory approach in socio-economic data management is crucial to ensure targeted and sustainable aid programs. This study aims to assist Yayasan Sahabat Yatim in managing socio-economic data of residents in Dusun Sidawung, Desa Kaligading, Kecamatan Boja, Kabupaten Kendal, while enhancing community capacity in data collection and utilization. The method used was Participatory Action Research (PAR), involving residents in planning, action, and reflection stages. Data were collected through surveys of 50 households, in-depth interviews, and reflective discussions with residents and the foundation. Results indicate that most households have incomes below IDR 2,500,000 per month, with primary needs including business capital, healthcare access, and children’s education. The intervention successfully produced a more structured socio-economic database, increased community participation, and enabled the foundation to distribute aid more transparently and accountably. These findings confirm that a participatory approach effectively strengthens community empowerment and supports data-driven development.Abstrak: Pendekatan partisipatif dalam pengelolaan data sosial-ekonomi sangat penting untuk memastikan bahwa program bantuan dapat tepat sasaran dan berkelanjutan. Pengabdian ini bertujuan untuk memberikan pendampingan kepada Yayasan Sahabat Yatim dalam mengelola data sosial-ekonomi warga Dusun Sidawung, Desa Kaligading, Kecamatan Boja, Kabupaten Kendal, serta meningkatkan kapasitas warga dalam proses pengumpulan dan pemanfaatan data. Metode yang diterapkan adalah Participatory Action Research (PAR), yang melibatkan masyarakat dalam tahap perencanaan, pelaksanaan, dan refleksi. Data dikumpulkan melalui survei terhadap 50 kepala keluarga, wawancara mendalam, dan diskusi reflektif bersama warga serta mitra yayasan. Hasil pengabdian menunjukkan bahwa mayoritas keluarga memiliki pendapatan di bawah Rp 2.500.000 per bulan, dengan kebutuhan utama yang mencakup modal usaha, akses kesehatan, dan pendidikan anak. Pendampingan ini berhasil menghasilkan basis data sosial-ekonomi yang lebih terstruktur, meningkatkan keterlibatan warga, serta membantu yayasan dalam menyalurkan bantuan secara lebih transparan dan akuntabel. Temuan ini mengonfirmasi bahwa pendekatan partisipatif efektif dalam memperkuat pemberdayaan masyarakat dan mendukung pembangunan berbasis data.
Perbandingan MAPE Metode Arima dan FTS Chen pada Peramalan Harga Minyak Mentah Widuri di Indonesia Ikhtiyar, Zakaria Bani; Sulistijanti, Wellie; Sari, Silvia Novita; Mahiruna, Adiyah
Square : Journal of Mathematics and Mathematics Education Vol. 6 No. 2 (2024)
Publisher : UIN Walisongo Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21580/square.2024.6.2.23579

Abstract

The need for crude oil greatly affects economic activities on a micro and macro scale. Indonesia is one of the countries that produces crude oil, although the amount produced is not as large as countries in the Middle East. The global price of crude oil has a direct impact on Indonesia's rising fuel prices. To find out the world price of crude oil in the future, forecasting can be done. Widuri type crude oil is one of the crude oils that is a priority in Indonesia. This study compares the accuracy of forecasting the price of Widuri type crude oil with the MAPE accuracy calculation method and the forecasting method compared is the ARIMA method with the FTS Chen method.Keywords: forecasting, Oil, ARIMA, Fuzzy Time Series (FTS) Chen.
Peramalan Tingkat Penghunian Kamar Hotel Bintang Lima Provinsi Bali Menggunakan Metode ARIMA dan Fuzzy Time Series Lee Rahayu Dwiastuti Pujiningrum; Sulistijanti, Wellie
JURNAL SOSIAL EKONOMI DAN HUMANIORA Vol. 10 No. 2 (2024): JURNAL SOSIAL EKONOMI DAN HUMANIORA
Publisher : Universitas Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jseh.v10i2.520

Abstract

This research focuses on forecasting the Occupancy Rate of five-star hotels in the Province of Bali. The increase in Occupancy Rate directly impacts the increase in the number of hotel rooms, which is a key factor for Bali Province as a leading tourist destination in Indonesia and globally. The data used spans from November 2021 to November 2023, with the methods employed being ARIMA with a model of (1,1,1) and Fuzzy Time Series Lee. After conducting the research, the accuracy of the forecast is evaluated through Mean Absolute Percentage Error (MAPE). The MAPE for ARIMA (1,1,1) is found to be 71.4413, while the Fuzzy Time Series Lee method yields a MAPE of 1.1778. With a significantly lower MAPE, the Fuzzy Time Series Lee method is considered more accurate in forecasting the Occupancy Rate of five-star hotels in the Province of Bali for the December 2023 period, with a forecast result of 70.4033%.
Strategies Of Religious Extension In Strengthening Moderate Understanding For The Prevention Of Radicalism And Identity Conflict Nasihin, Muh; Sulistijanti, Wellie; Setiawan, Rahmat; Suprayitno, Muhammad Robeth; Putri, Meia Rizqi Talitha; Cahyani, Desy Eki
Intiqad: Jurnal Agama dan Pendidikan Islam Vol 17, No 2 (2025)
Publisher : UMSU

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30596/26677

Abstract

This study aims to examine the strategies implemented by religious instructors in preventing radicalism and identity friction in four regions in Central Java—Semarang City, Demak Regency, Semarang Regency, and Kendal Regency. Using a qualitative approach with a case study design, data were collected through in-depth interviews and focus group discussions with religious instructors, lecturers, and community leaders to provide in-depth context for local dynamics. The results reveal three complementary clusters of strategies: (1) socializing the value of religious moderation through sermons/lectures, religious study groups, and targeted use of social media; (2) strengthening religious and national literacy through non-formal education such as training, workshops, and interactive forums; and (3) multi-sector collaboration with local governments, religious organizations, and community leaders. However, the implementation of these strategies faces recurring challenges, including limited resources, uneven policy support, and security risks when interacting with intolerant groups. Research findings indicate that the role of religious instructors as key local actors in building social resilience against radicalism is highly strategic, but its effectiveness depends on the integration of educational, collaborative, and contextual approaches supported by sustainable policies and resources.
PENERAPAN VECM DALAM MENGIDENTIFIKASI PENGARUH CURAH HUJAN DAN LUAS LAHAN TERHADAP PRODUKSI KOPI DI SUMATERA SELATAN Sangnandha, Habill Putra; Sulistijanti, Wellie
EPSILON: JURNAL MATEMATIKA MURNI DAN TERAPAN Vol 19, No 2 (2025)
Publisher : Mathematics Study Program, Faculty of Mathematics and Natural Sciences, Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20527/epsilon.v19i2.15392

Abstract

Coffee production in South Sumatra Province plays an important role in Indonesia’s economy, both as a source of foreign exchange and as a livelihood for farmers. During the 2020–2023 period, coffee production exhibited fluctuations that reflected instability and were suspected to be influenced by environmental factors such as rainfall and land area. This study aims to analyze the influence of rainfall and land area on coffee production using the Vector Error Correction Model (VECM), which is capable of examining both short-term and long-term relationships among cointegrated time series variables. The data used consist of monthly records of coffee production, land area, and rainfall obtained from the Central Bureau of Statistics for the 2020–2023 period. The analysis was conducted through a series of statistical tests, including stationarity testing, cointegration, determination of optimal lag, VECM estimation, Granger causality test, as well as impulse response function (IRF) and variance decomposition (VD). The results reveal the existence of a long-term relationship among the variables, where rainfall significantly affects coffee production, while land area does not show a meaningful effect. The VD analysis also emphasizes that rainfall’s contribution to production variation increases up to 10% in the long term, while the applied model is validated through the Portmanteau test. These findings confirm that climatic factors, particularly rainfall, play an essential role in maintaining the stability and sustainability of coffee production in South Sumatra.