cover
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota bandung,
Jawa barat
INDONESIA
Kubik
ISSN : -     EISSN : -     DOI : -
Core Subject : Education,
Arjuna Subject : -
Articles 131 Documents
Implementation of BiLSTM to Predict World Crude Oil Prices Sari, Firda Yunita; Ulinnuha, Nurissaidah
KUBIK Vol 10 No 1 (2025): IN PRESS
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

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

Abstract

The main source of energy worldwide is crude oil, which is used by almost all countries as an energy source. Crude oil plays a key role in driving the global economy, especially in the industrial and transportation sectors. Along with technological developments, crude oil price predictions can be made more sophisticated using artificial intelligence-based methods, one of which is the Bidirectional Long Short-Term Memory (BiLSTM) method which is a development of the Long Short-Term Memory (LSTM) method by combining past and future information when processing sequential data, BiLSTM uses forward and backward LSTM simultaneously to increase accuracy. The study used world crude oil price data for 1 year. There are 57 tests with several parameters such as data division, number of neurons, batch size, and activation function. After testing with the BiLSTM method for 57 scenarios, there is the smallest MAPE value of 0.09% at a data division of 90:10, number of neurons 100, batch size of value 4, and ReLu activation function. The resulting prediction model is highly accurate based on the MAPE criterion value.
Penyelesaian Masalah Transportasi Untuk Mencari Solusi Optimal Dengan Pendekatan Minimum Spanning Tree (Mst) Menggunakan Algoritma Kruskal Dan Algoritma Prim Dili, Yusufiani Nurlinawati; Wulan, Elis Ratna; Ilahi, Fadilah
KUBIK Vol 6 No 1 (2021): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v6i1.13907

Abstract

Penelitian ini membahas tentang penyelesaian masalah transportasi dengan pendekatan Minimum Spanning Tree (MST) menggunakan algoritma Kruskal dan algoritma Prim untuk mencari solusi optimal. Algoritma Kruskal dan algoritma Prim merupakan algoritma dalam teori graf untuk mencari Minimum Spanning Tree (MST). Langkah algoritma Kruskal yaitu mengurutkan biaya dari yang terkecil hingga terbesar. Selanjutnya, pilih biaya yang paling terkecil. Kemudian, lakukan perhitungan dengan melihat sumber persediaan dan permintaan di setiap tujuan sampai semuanya terpenuhi, sehingga terlihat bentuk Minimum Spanning Tree (MST) dari algoritma Kruskal. Sedangkan langkah algoritma Prim yaitu dengan memilih sembarang titik atau sumber. Selanjutnya, pilih active edge dengan biaya terkecil. Kemudian, lakukan perhitungan dengan melihat sumber persediaan dan permintaan di setiap tujuan sampai semuanya terpenuhi, sehingga terlihat bentuk Minimum Spanning Tree (MST) dari algoritma Prim. Bentuk dari Minimum Spanning Tree (MST) menghasilkan solusi yang optimal. Dari hasil penelitian ini, pendekatan Minimum Spanning Tree (MST) dengan algoritma Prim yang lebih unggul. 
Penerapan Perangkat Lunak RStudio untuk Penaksiran Parameter Model Spatial Autoregressive Salsabil, Tsuroyya; Kusuma, Dianne Amor; Ruchjana, Budi Nurani
KUBIK Vol 8 No 1 (2023): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v8i1.30037

Abstract

Research and analysis that are not only based on time (temporal) but also on space (spatial) require tools in the form of software to ensure that the data analysis and processing yield good, fast, and accurate results. One of the software tools that can be used for this purpose is RStudio software. The advantages of RStudio include being open-source software (OSS), which can be used freely without cost, and it has many packages and functions that can facilitate data processing. One of the spatial-based analyses is spatial data analysis. The structure within RStudio allows users to call functions related to spatial data analysis, perform computations with sparse matrices (matrices with many zero values), such as spatial weight matrices, estimation of spatial model parameters, and so on. This research examines the application of RStudio software in estimating the parameters of a first-order Spatial Autoregressive (SAR) model using the Maximum Likelihood Estimation (MLE) method on the data of the designation of Intangible Cultural Heritage (ICH) in Indonesia. Based on the results of applying RStudio software, a first-order SAR model with a Queen contiguity weight matrix for the categories of Traditional Customs, Rituals, and Celebrations (TCRC) and Performing Arts (PA) with the minimum Akaike Information Criterion (AIC) value and maximum pseudo- value was obtained for predicting the designation data of ICH in Indonesia. The application of RStudio software to the first-order SAR model for the designation data of ICH in Indonesia speeds up and simplifies calculations, making it suitable as a recommendation for relevant agencies such as the Department of Culture, Tourism, Youth, and Sports (Disbudparpora). 
Bilangan Rainbow Connection Pada Graf Buku Medika, Gema Hista; Tomi, Zebbil Billian; Wulan, Elis Ratna
KUBIK Vol 9 No 1 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i1.30406

Abstract

Penelitian ini berfokus pada penentuan bilangan rainbow connection pada graf buku, terutama graf buku dengan bentuk segitiga, segi empat, serta segi lima. Graf buku segitiga, disimbolkan sebagai BT_n, dihasilkan dari penggabungan sisi pada graf lingkaran C_3. Graf buku segi empat, disimbolkan sebagai BE_n, dihasilkan dari penggabungan sisi pada graf lingkaran C_4. Sedangkan graf buku segi lima, disimbolkan sebagai BL_n, dihasilkan dari penggabungan sisi pada graf lingkaran C_5. Bilangan rainbow connection pada sebuah graf terhubung G ditulis rc(G), yang didefinisikan sebagai jumlah warna minimum yang dibutuhkan untuk menjadikan graf G menjadi terhubung secara rainbow. Penelitian ini merupakan penelitian deskriptif kualitatif, melalui teknik studi pustaka dan analisis data non-statistik. Adapun hasil penelitian menyatakan bahwa bilangan rainbow connection pada Graf Buku Segitiga yaitu 1 jika n=1, 2 jika n=2, dan 3 untuk n≥3. Bilangan rainbow connection pada Graf Buku Segi Empat adalah 2 jika n=1, 3 jika n=2, dan 4 untuk n≥3. Sedangkan bilangan rainbow connection pada Graf Buku Segi Lima yaitu 3 jika n=1, 4 jika n=2, dan 6 untuk n≥3.
SEIHR-SEI Mathematical Model of Zika Virus Transmission with Vector Control Shiddiqie, Ichwal Afrizan; Khumaeroh, Mia Siti; Zulkarnaen, Diny; Diana, Arista Fitri
KUBIK Vol 9 No 2 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i2.30948

Abstract

Zika virus (ZIKV) is transmitted by Aedes Aegypti mosquito, which is recognized as a vector for viruses causing dengue fever and chikungunya. This study uses SEIHR‐SEI mathematical model to analyze the dynamics of Zika virus transmission. In this model, human population (host) is classified into five compartments: Susceptible Humans (Sh), Exposed Humans (Eh), Infected Humans (Ih), Hospitalized Humans (Hh) and Recovered Humans (Rh). Meanwhile, the mosquito population (vector) is divide into three compartments: Susceptible Vectors (Sv), Exposed Vectors (Ev), and Infected Vectors (Iv). Stability analysis is conducted using Routh‐Hurwitz criteria for assessing local stability and Lyapunov function for evaluating global stability. Moreover, Basic Reproduction Number (R0), which represents the average number of new infections produced by one infected individual in a susceptible population, is derived by using the Next Generation Matrix (NGM) method. The result shows that the equilibrium point for disease‐free conditions is globally asymptotic stable when R0 < 1, meanwhile the equilibrium point for endemic conditions is stable when R0 > 1. The simulation result using endemic data and sensitivity analysis of three parameters, including contact rate between susceptible humans and infected humans (c), hospitalization rate of infected individuals (τ ), and mosquito control rate (ω), reveals that c and ω exert a more significant effect on changes in R0 compared to τ . Therefore, minimizing contact with infected individuals or implementing vector control is more effective than isolating or hospitalizing infected patients.
Comparative Study of Parameter Estimation Methods in Pharmacokinetic Model with Oral Administration: Simulations of Theophylline Drug Concentration Zulkarnaen, Diny
KUBIK Vol 9 No 1 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i1.31233

Abstract

Parameter estimation for the elimination and absorption rate constants is performed in a pharmacokinetic model, where a drug is administered orally. Some methods have been introduced to estimate these parameters but without comparison which one gives better estimates. Here, two different methods are used for comparison in estimating the absorption rate constant: the Wagner-Nelson and residual methods. The Wagner-Nelson method requiring fewer data sets while the residual method uses all available data sets for estimation. For the elimination rate constant estimate, we use only the least square error method. Simulations are conducted using sample data points of Theophylline drug concentration that varies over time to estimate the parameters. These parameter values are then utilized to approximate the drug concentration over time, using both methods. These approximations are then compared with the actual data sets to see and calculate the error values so that the best method can be determined. The comparison shows that the residual method provides better approximation since this method utilizes the entire sample data points, while the Wagner-Nelson uses only the data in the beginning time, that is when the absorption process is dominant.
Peramalan Menggunakan Model Holt-Winters Exponential Smoothing Multiplikatif dengan Optimasi Parameter Menggunakan Particle Swarm Optimization (PSO) Martina, Annisa; Dewi, Apianti Setia; Awalluddin, Asep Solih
KUBIK Vol 9 No 2 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i2.31368

Abstract

Peramalan merupakan suatu metode yang digunakan untuk memprediksi kejadian di masa yang akan datang berdasarkan pada sekumpulan data historis. Adanya pola trend dan musiman multiplikatif pada suatu data memerlukan metode peramalan yang sesuai agar peramalan yang dilakukan dapat memberikan hasil yang baik. Model Holt-Winters Exponential Smoothing Multiplikatif merupakan salah satu yang dapat mengatasi data dengan pola tersebut. Pada Holt-Winters Exponential Smoothing Multiplikatif terdapat nilai parameter pemulusan yang biasanya diperoleh dengan menggunakan trial and error. Namun, metode trial and error dirasa kurang efektif karena akan ada banyak kombinasi parameter pemulusan yang mungkin memberikan hasil peramalan optimal. Oleh karena itu, metode Particle Swarm Optimization digunakan untuk mengoptimasi parameter pemulusan Holt-Winters Exponential Smoothing Multiplikatif dengan nilai MAPE sebagai nilai akurasi peramalan. Dari hasil simulasi yang dilakukan diperoleh bahwa metode Particle Swarm Optimization memiliki kinerja yang baik dalam memperoleh nilai parameter pemulusan Holt-Winters Exponential Smoothing Multiplikatif dalam meramalkan data jumlah kunjungan wisatawan mancanegara ke Indonesia.
Panel Data Analysis of Two Level Mixed Linear Models for Factors Affecting The Health Index in West Java Awalluddin, Asep Solih; Khumaeroh, Mia Siti; Amalia, H.; Wahyuni, Inge
KUBIK Vol 9 No 1 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i1.31369

Abstract

The purpose of this study is to construct a multilevel mixed linear model for panel data by estimating parameters and testing the hypothesis of fit of the model with case studies in determining the prediction of the health index for the marginal and conditional models on the factors that influence the prediction of the health index in West Java for 2016 data. -2021, with time (year) and region (district and city) variables as factors involved in the model. Multilevel mixed linear model is the development of a mixed linear model that can be used to analyze correlated panel data. Parameter estimation uses the Maximum Likelihood (ML) method to estimate fixed effect parameters and Restricted Maximum Likelihood (REML) to estimate covariance parameters. The results obtained by the health index prediction model in West Java, both for the marginal and conditional prediction models and goodness of fit model.
Peramalan Menggunakan Metode MS-AR, MS-Regression dan MS-VAR pada Model Perubahan Struktur Martina, Annisa; Jajuli, Marsa Aufa; Cahyandari, Rini
KUBIK Vol 9 No 1 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i1.31371

Abstract

Aktivitas ekspor dan impor memiliki peranan penting pada perdagangan internasional yang dapat meningkatkan pembangunan ekonomi di dalam negeri. Ekspor di Indonesia terdiri dari dua sektor yaitu migas dan non migas. Adanya perubahan regime dalam data migas menyebabkan volume ekspor cenderung fluktuatif dan memiliki perubahan struktur. Model perubahan struktur dapat diselesaikan dengan menggunakan metode Markov Switching (MS) yang terdiri dari MS Autoregressive, MS Regression dan MS Vector Autoregressive. Pada penelitian ini menggunakan data bulanan volume ekspor migas Indonesia selama 5 tahun (2018 sd 2022). Hasil penelitian menunjukkan bahwa terjadi perubahan struktur pada waktu ke-12 (Desember 2018) dengan 2 regime. Setelah dilakukan pemodelan dan uji diagnostik, didapatkan model terbaik untuk volume ekspor migas Indonesia adalah MS(2) menggunakan metode MS Regression dengan nilai MAPE 12.49% dan nilai peluang perpindahan regime 94%, yang menyatakan bahwa ekspor migas yang dilakukan oleh negara Indonesia memberikan dampak yang cukup baik untuk perekonomian Indonesia, terutama dalam hal perdagangan internasional.
Pure Multiplication Module over Dedekind Domain Nurhalimah, Luthfia; Kusniyanti, Elvira; Irawati, Irawati
KUBIK Vol 9 No 1 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Jurusan Matematika, Fakultas Sains dan Teknologi, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i1.33207

Abstract

Various studies have explored the fascinating characteristics of modules over discrete valuation domain. One notable finding is that the multiplication module is regarded as indecomposable within a discrete valuation domain. Based on this distinctive property, a categorization of weak and pure multiplication modules over discrete valuation domain is established. A notable property of a discrete valuation ring is its role as the localization of a Dedekind domain. With this connection, there has been a classification of weak multiplication modules over the Dedekind domain. In this article, we examine the characteristics of the discrete valuation domain and the properties of pure multiplication modules over the discrete valuation domain, which collectively contribute to the properties of pure multiplication modules over the Dedekind domain.              

Page 11 of 14 | Total Record : 131