Article Per Year (5 Year)
p-Index From 2020 - 2025
2.768
P-Index
This Author published in this journals
All Journal Jurnal Statistika Universitas Muhammadiyah Semarang Jurnal Edukasi dan Penelitian Informatika (JEPIN) BAREKENG: Jurnal Ilmu Matematika dan Terapan JTAM (Jurnal Teori dan Aplikasi Matematika) Variance : Journal of Statistics and Its Applications Community Development Journal: Jurnal Pengabdian Masyarakat Journal of Data Insights Prosiding Seminar Nasional Unimus Parameter: Jurnal Matematika, Statistika dan Terapannya LogicLink: Journal of Artificial Intelligence and Multimedia in Informatics Emerging Statistics and Data Science Journal
Fauzi, Fatkhurokhman
Unknown Affiliation
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Energy Engineering Health Professions Mathematics Mechanical Engineering Nursing Physics Public Health Transportation

Published : 20 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 20 Documents
Search

 1   2 
Peramalan Harga Emas Menggunakan Pendekatan Long-Short Term Memory (LSTM) Fauzi, Fatkhurokhman; Aulia, Syifa; Syaifullah, Ahmad Reyhan; Utami, Tiani Wahyu
JEPIN (Jurnal Edukasi dan Penelitian Informatika) Vol 10, No 2 (2024): Volume 10 No 2
Publisher : Program Studi Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/jp.v10i2.78332

Abstract

Inventasi menjadi salah satu pilihan masyarakat untuk mengelola kelebihan dana agar nilainya meningkat dikemudian hari. Emas menjadi salah satu komoditi yang sering dijadikan instrumen investasi favorit. Harga emas yang fluktuatif menimbulkan efek kerugian bagi investor. Peramalan harga emas dimasa yang akan datang menjadi penting untuk meminimalisir resiko kerugian. Pendekatan machine learning lebih baik dibandingkan inferensial seperti Autoregressive Integrated Moving Average (ARIMA) dalam meramalkan data yang fluktuatif. Metode dengan pendekatan machine learning seperti Long-Short Term Memory (LSTM) memiliki performasi yang baik pada data yang fluktuatif. Metode LSTM digunakan untuk untuk meramalkan harga emas. Penelitian ini membagi data training dan data testing sebesar 80% dan 20%. Metode evaluasi model Mean Absolute Percentage Error (MAPE) digunakan untuk melihat kebaikan model. Penelitian ini menerapkan enam scenario tunning parameter. Parameter terbaik metode LSTM yaitu learning rate 0,01, neuron 10, dan Epoch 100 dengan nilai MAPE sebesar 3,499%. Hasil MAPE pada data training dan data testing tidak menunjukan terjadinya overfitting atau underfitting pada metode LSTM terbaik. Hasil peramalan tiga puluh periode cenderung fluktuatif, terjadi kenaikan yang signifikan pada periode ke dua puluh empat ke dua puluh lima.
STOCK PRICE FORECASTING OF PT. BANK CENTRAL ASIA USING HYBRID AUTOREGRESSIVE INTEGRATED MOVING AVERAGE-NEURAL NETWORK (ARIMA-NN) METHOD Azizah, Apipah Nur; Fauzi, Fatkhurokhman; Arum, Prizka Rismawati
Jurnal Statistika Universitas Muhammadiyah Semarang Vol 12, No 1 (2024): Jurnal Statistika Universitass Muhammadiyah Semarang
Publisher : Department Statistics, Faculty Mathematics and Natural Science, UNIMUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26714/jsunimus.12.1.2024.48-59

Abstract

PT. Bank Central Asia is a private company that has superior shares in the Lq45 category but has share prices that fluctuate every period. So forecasting is needed to predict stock prices in the next period. These fluctuations can cause linear and nonlinear relationships in historical stock price data. This research uses the Hybrid ARIMA-NN approach, where the ARIMA model is able to overcome data non-stationarity while the Neural Network is used to capture nonlinear patterns that cannot be explained by the ARIMA model by using the residuals as NN input, the hybrid model can increase forecasting accuracy. The data used is weekly data on closing stock prices for the period January 2019 to June 2024. Prediction measurements use Mean Absolute Percentage Error. The research results show that forecasting with Hybrid ARIMA(2,1,2)-NN(1-5-1) obtained a MAPE value of 3.99% smaller than the ARIMA(2,1,2) a MAPE value of 4.13%, that the accuracy of the forecasting model is good.
Analisis Sentimen Opini Masyarakat Terhadap Stunting Pasca Debat Cawapres Pertama 2024 Dengan Algoritma Bootstrap Aggregating Naïve Bayes Yuliardi, Fahrul Raditiar; Fauzi, Fatkhurokhman; Utami, Tiani Wahyu
Prosiding Seminar Nasional Unimus Vol 7 (2024): Transformasi Teknologi Menuju Indonesia Sehat dan Pencapaian Sustainable Development G
Publisher : Universitas Muhammadiyah Semarang

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

Abstract

Stunting adalah kondisi gagal tumbuh pada anak-anak yang disebabkan oleh kekurangan gizi, kurangnyapengetahuan mengenai stunting, serta faktor ekonomi dan akses terhadap makanan bergizi. Debat calonwakil presiden pertama tahun 2024 menarik perhatian publik dan menjadi topik yang banyak dibicarakandi media sosial, termasuk X. Penelitian ini menggunakan metode Bootstrap Aggregating (Bagging) padaalgoritma Naïve Bayes untuk menganalisis sentimen opini masyarakat yang diekspresikan melalui tweetterkait stunting. Metode ini dipilih karena kemampuannya dalam meningkatkan akurasi klasifikasi denganmengatasi masalah ketidakseimbangan kelas dalam data sentimen. Hasil confusion matrix menunjukkanbahwa metode yang digunakan mampu mengklasifikasikan sentimen dengan akurasi sebesar 96%,precision 95%, recall 98%, dan F1-score 96%. Metode ini menunjukkan bahwa Bootstrap AggregatingNaïve Bayes efektif dalam mengklasifikasikan sentimen masyarakat mengenai stunting, terutama dalamkonteks respons terhadap pernyataan dan diskusi yang muncul setelah debat cawapres. Mayoritas sentimenyang dianalisis adalah positif, hal ini menunjukan dukungan dan harapan masyarakat untuk meningkatkankesehatan dan kesejahteraan anak-anak.  Penelitian ini memberikan wawasan tentang bagaimana debatcawapres 2024 dapat mempengaruhi persepsi masyarakat terhadap kebijakan pemerintah Indonesia. Kata Kunci : Analisis Sentimen, Bootstrap Aggregating, Naive Bayes, Debat Calon Wakil Presiden,Stunting.
Metode Bidirectional Long Short-Term Memory (BiLSTM) Untuk Memprediksi Harga Saham BBRI Dengan Optimasi Nesterov Adaptive Moment (Nadam) Permatasari, Shella Heidy; Nur, Indah Manfaati; Fauzi, Fatkhurokhman
Prosiding Seminar Nasional Unimus Vol 7 (2024): Transformasi Teknologi Menuju Indonesia Sehat dan Pencapaian Sustainable Development G
Publisher : Universitas Muhammadiyah Semarang

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

Abstract

Prediksi harga saham merupakan salah satu bidang yang sangat menantang dan memiliki dampak signifikandalam dunia keuangan. Penelitian ini menggunakan metode Bidirectional Long Short-Term Memory(BiLSTM) yang dioptimasi dengan Nesterov Adaptive Moment (Nadam) untuk memprediksi harga sahamharian PT Bank Rakyat Indonesia (BBRI). Metode BiLSTM merupakan variasi dari metode Long ShortTerm Memory (LSTM) yang memecahkan ketergantungan jangka pajang LSTM dengan RNN (RecurrentNeural Network). BiLSTM ini memiliki kemampuan untuk menangkap pola temporal dari masa lalu danmasa depan sehingga efektif dalam analisis deret waktu. Sedangkan, optimasi Nadam digunakan untukmeningkatkan kecepatan konvergensi dan akurasi prediksi dengan memanfaatkan kelebihan darimomentum Nesterov dan adaptivitas Adam. Metode BiLSTM yang dioptimalkan dengan optimasi Nadamtersebut menghasilkan model prediksi harga saham harian terbaik dengan konfigurasi optimal 30 neuronper lapisan tersembunyi, batch size 256, dan 500 epoch. Konfigurasi ini efektif dalam menangkap polafluktuasi harga saham harian dengan MSE 0,000415. Pada evaluasi kinerja model dengan MAPE, diperolehnilai MAPE 1,7511% pada data training serta 1,5432% pada data testing, yang menunjukkan bahwa modeltersebut akurat dengan tingkat kesalahan prediksi di bawah 10%. Hasil prediksi menunjukkan kenaikanstabil setiap hari dengan harga saham terendah 4496,028 pada 1 Juni 2024 dan tertinggi 4819,317 pada 29Juni 2024. Kata Kunci : BiLSTM, Nesterov Adam, Prediksi, Saham
Pelatihan Pertanian Modern untuk Meningkatkan Regenerasi Petani Muda di Desa Sidorejo Demak Fadillah, Muhammad Reza; Fauzi, Fatkhurokhman; Haris, M. Al; Multiyaningrum, Riska; Pandiriyan, Muhammad Tegar; Amrullah, Setiawan; Putri, Melfia Verahma; Nur, Rachmat Kahfiwan; Ramadhan, Abimanyu Arya; Khikman, Muhammad Alvaro; Widiyanti, Karin Dita; Watur, Annisa Cahyaningrum; Fabiola, Gwenda; Syaharani, Nabbila Dyah; Putra, Septian Malik; Ninu, Maria Febronia; Barlian, Seftia Amelia Rizki
Prosiding Seminar Nasional Unimus Vol 7 (2024): Transformasi Teknologi Menuju Indonesia Sehat dan Pencapaian Sustainable Development G
Publisher : Universitas Muhammadiyah Semarang

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

Abstract

Desa Sidorejo, Kecamatan Karangawen, Kabupaten Demak, merupakan daerah agraris dengan mayoritaspenduduk bermata pencaharian sebagai petani. Dalam upaya meningkatkan regenerasi petani muda danmemperkenalkan teknologi pertanian modern, Tim Pengabdian Himpunan Mahasiswa StatistikaUniversitas Muhammadiyah Semarang melaksanakan pelatihan penggunaan alat transplanter dan powerthresher. Kegiatan ini diawali dengan koordinasi bersama pemerintah desa dan Dinas Pertanian setempat,diikuti dengan pembuatan materi pelatihan yang divalidasi oleh Balai Penyuluhan Pertanian. Sebanyak 20petani muda berusia 15 hingga 35 tahun dilibatkan dalam pelatihan ini. Hasil evaluasi menunjukkanpeningkatan signifikan dalam pengetahuan dan keterampilan peserta, meskipun tantangan keberlanjutanprogram tetap perlu diperhatikan. Pelatihan ini diharapkan dapat mendorong pertanian berkelanjutan dankesejahteraan ekonomi masyarakat. Kata Kunci : Pemberdayaan Masyarakat, Pertanian Modern, Regenerasi Petani Muda, Transplanter, PowerThresher.
Segmentasi Provinsi Berdasarkan Indikator Sanitasi Total Berbasis Masyarakat di Indonesia: Pendekatan Hierarchical Clustering Agglomerative Syahrani, Nabbila Dyah; Rahmawati, Gita; Izzah, Nasyiatul; Sanmas, Safril Ahmadi; Rahmah, Alfidha; Fauzi, Fatkhurokhman
Prosiding Seminar Nasional Unimus Vol 7 (2024): Transformasi Teknologi Menuju Indonesia Sehat dan Pencapaian Sustainable Development G
Publisher : Universitas Muhammadiyah Semarang

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

Abstract

Sanitasi yang buruk menjadi salah satu faktor penyebab beban penyakit global, dengan sekitar 1,4 jutakematian setiap tahunnya disebabkan oleh faktor lingkungan, termasuk sanitasi yang tidak aman. Padatahun 2023, sebanyak 4,2 miliar orang di dunia tidak memiliki akses sanitasi yang dikelola dengan aman,yang dapat menyebabkan berbagai masalah kesehatan seperti malnutrisi dan kematian. Di Indonesia,program Sanitasi Total Berbasis Masyarakat (STBM) dikembangkan oleh Kementerian Kesehatan untukmeningkatkan perilaku higienis melalui pemberdayaan masyarakat. STBM terdiri dari lima pilar: StopBuang Air Besar Sembarangan (SBS), Cuci Tangan Pakai Sabun (CTPS), Pengelolaan Sampah RumahTangga (PSRT), Pengelolaan Air Minum dan Makanan Rumah Tangga (PAMMRT), serta Pengelolaan AirLimbah Domestik Rumah Tangga (PALDRT). Namun, implementasi program ini di berbagai daerah diIndonesia masih bervariasi, dengan beberapa daerah belum menerapkan STBM secara optimal. Oleh karenaitu, diperlukan pemetaan daerah-daerah yang membutuhkan intervensi lebih lanjut. Penelitian inimenggunakan metode Hierarchical Clustering untuk mengelompokkan provinsi-provinsi di Indonesiaberdasarkan data capaian STBM tahun 2023. Penelitian ini menghasilkan empat cluster denganmenggunakan Agglomerative Average Linkage sebagai metode terbaik. Hasil clustering menunjukkanCluster 1 memiliki kondisi sanitasi terburuk, Cluster 2 memiliki pencapaian terbaik, sementara Cluster 3dan Cluster 4 berada di antara keduanya, dengan beberapa aspek memerlukan perbaikan. Diperoleh nilaievaluasi hasil clustering menggunakan silhouette coefficient sebesar 0,5027.  Kata Kunci : Hierarchical Clustering, STBM, Akses Sanitasi
FORECASTING NICKEL PRICES WITH THE AUTOMATIC CLUSTERING FUZZY TIME SERIES MARKOV APPROACH Haris, M. Al; Sari, Wulan; Fauzi, Fatkhurokhman; Sam'an, Muhammad
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 19 No 2 (2025): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol19iss2pp1237-1250

Abstract

Nickel was a critical raw material used in a wide range of industries. The price movement of nickel tends to fluctuate and remain uncertain due to market conditions varying over time. Therefore, forecasting nickel prices was essential to understanding future price movements. In this study, we applied the automatic clustering fuzzy time series Markov chain method. The automatic clustering algorithm generates multiple intervals and fuzzy relations. Subsequently, forecasting was based on these fuzzy relations and a Markov chain transition probability matrix involving three stages to enhance forecast accuracy. We use monthly closing futures nickel price data from January 2009 to May 2024. The accuracy of the forecasting model was measured using the mean absolute percentage error (MAPE). The analysis showed that implementing the automatic clustering fuzzy time series Markov chain method results in excellent forecasting accuracy, with a MAPE value of 1.76% (equivalent to 98.24% accuracy). The predicted nickel price for June 2024 was US$ 19,608.5.
STATISTICAL DOWNSCALING USING REGRESSION NONPARAMETRIC OF FOURIER SERIES-POLYNOMIAL LOCAL OF CLIMATE CHANGE Utami, Tiani Wahyu; Fauzi, Fatkhurokhman; Yuliyanto, Eko
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 17 No 3 (2023): BAREKENG: Journal of Mathematics and Its Applications
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol17iss3pp1411-1418

Abstract

Indonesia is a tropical country that is vulnerable to the impacts of climate change. Climate change causes an effect on the level of comfort (heat stress) that can affect the level of human immunity, one of the indices to calculate the level of human comfort (heat stress) is the Thermal Humidity Index (THI). Climate change scenarios modeled in Earth System Models (ESMs). ESM has a coarse resolution and is subject to considerable bias. This research is using secondary data. The data source used in this study comes from the Coupled Model Intercomparison Project (CMIP5). This research will focus on projected heat stress which is calculated based on THI with the temperature and humidity variables. Therefore, in this research to reduce the bias correction method used Statistical Downscaling (SD) and nonparametric regression. The results of the bias correction using the Statistical Downscaling (SD) method and Nonparametric Regression Fourier-Polynomial Local Series in this study the R-square value for Relative Humidity yields 95% and for Temperature yields 94%. The projection of climate change based on the value of the Temperature Humidity Index (THI) in Indonesia in the category of 50% of the population of Indonesians feeling comfortable conditions occurred in 2006-2059. Then the population of citizens in Indonesia felt uncomfortable conditions occurred in 2060 to 2100 with a THI value of 27.0730°C - 27.7800°C.
COMPARISON OF RANDOM FOREST AND NAÏVE BAYES CLASSIFIER METHODS IN SENTIMENT ANALYSIS ON CLIMATE CHANGE ISSUE Fauzi, Fatkhurokhman; Setiayani, Wiwik; Utami, Tiani Wahyu; Yuliyanto, Eko; Harmoko, Iis Widya
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 17 No 3 (2023): BAREKENG: Journal of Mathematics and Its Applications
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol17iss3pp1439-1448

Abstract

The last decade was recorded as a decade with a bad record on the issue of disasters in the world due to climate change. Measuring public opinion is one of the steps to mitigate the impact of climate change. Twitter is a popular social media for channeling opinions. Twitter provides a great source of data for understanding public opinion and the perceived risk of an issue. In recent decades, when discussing climate change, there are those who agree and those who oppose it. Sentiment analysis is a branch of learning in the realm of text mining that is used as a solution to see opinions on a problem, one of which is climate change. In this study, we will try to analyze opinions on climate change issues using the Random Forest and Naïve Bayes classifier methods. Data were obtained from Twitter for the period January 2022-June 2022. The training data used in this research is 80%:20%. There are slightly more positive statements than negative ones. The results obtained with the Naïve Bayes classifier method are an accuracy of 76.25%, an F-1 score of 78%, and a recall of 80%. While the results of the random forest method are 70.6% accuracy, 69% F-1 score, and 63% recall. The Nive Bayes method is better than the Random Forest method for classifying climate change opinions with an accuracy of 76.25%.
CRYPTOCURRENCY PRICE PREDICTION: A HYBRID LONG SHORT-TERM MEMORY MODEL WITH GENERALIZED AUTOREGRESSIVE CONDITIONAL HETEROSCEDASTICITY Nur, Indah Manfaati; Nugrahanto, Rifqi; Fauzi, Fatkhurokhman
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 17 No 3 (2023): BAREKENG: Journal of Mathematics and Its Applications
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol17iss3pp1575-1584

Abstract

Cryptocurrency is a virtual payment instrument currently popular as an investment alternative. One type of cryptocurrency widely used as an investment is Bitcoin due to its high-profit potential and risk due to unstable exchange rate fluctuations. This high exchange rate fluctuation makes trading transactions in the crypto market speculative and highly volatile. To overcome this volatility factor, this research used the Generalized Autoregressive Conditional Heteroscedasticity forecasting method to describe the heteroscedasticity factor, as well as a Recurrent Neural Network (RNN) with long-short-term memory that has feedback in modeling sequential data for time series analysis. The two methods are combined to overcome the dependency of time series data in the long term and the heteroscedastic effect of the volatility of price changes. The results of the GARCH-LSTM hybrid model in this study show a Mean Absolute Percentage Error (MAPE) value of 15.69%. The accuracy value is obtained from the division of training data by 80% and testing data by 20%, with the number of neurons as many as three and epochs of 100 using the Adam optimizer. The MAPE accuracy results show a good prediction in predicting the value.
Co-Authors Alwan Fadlurohman Amri, Saeful Amrullah, Setiawan Anawai Basman, Aprilla Arafat, Amaliah Sholeha Ardiansyah, Muhammad Rifqy Aulia, Syifa Azizah, Apipah Nur Barlian, Seftia Amelia Rizki Burhanuddin Izzul Salam Eko Yuliyanto, Eko Eny Winaryati Erlinda, Relly Fabiola, Gwenda Fadillah, Muhammad Reza Haris, M Al Haris, M. Al Iis Widya Harmoko, Iis Widya Indah Manfaati Nur Izzah, Nasyiatul Junaidi, Muhammad Rifki Khikman, Muhammad Alvaro Laila Khoirun Nisa Lestari, Febi Anggun M. Al Haris M. Al Haris Multiyaningrum, Riska Ninu, Maria Febronia Nugrahanto, Rifqi Nur, Rachmat Kahfiwan Pandiriyan, Muhammad Tegar Permatasari, Shella Heidy Prizka Rismawati Arum Putra, Septian Malik Putri, Agata Dwi Putri Putri, Melfia Verahma Rahmah, Alfidha Rahman , Budiono Rahman, Budiono Rahmawati, Gita Ramadhan, Abimanyu Arya Sam'an, Muhammad Sanmas, Safril Ahmadi Sarah, Albertus Dion Setiayani, Wiwik Syaharani, Nabbila Dyah Syahrani, Nabbila Dyah Syaifullah, Ahmad Reyhan Tari Fitri Ningsih Tiani Wahyu Utami Watur, Annisa Cahyaningrum Widiyanti, Karin Dita Wulan Sari, Wulan Yuliardi, Fahrul Raditiar