Claim Missing Document
Check
Articles

Found 20 Documents
Search

Pengaruh Parameter Regularisasi (λ) terhadap Stabilitas Estimasi pada Regresi Ridge Linda Rassiyanti; Rohimatul Anwar
JURNAL RISET RUMPUN MATEMATIKA DAN ILMU PENGETAHUAN ALAM Vol. 5 No. 1 (2026): April : JURRIMIPA: Jurnal Riset Rumpun Matematika dan Ilmu Pengetahuan Alam
Publisher : Lembaga Pengembangan Kinerja Dosen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/jurrimipa.v5i1.8726

Abstract

Multicollinearity is one of the common issues in multiple linear regression that can lead to instability in the estimation of regression coefficients. This study aims to examine the impact of multicollinearity on regression models and to evaluate the use of Ridge Regression as an alternative estimation method. The study employs simulated data consisting of 1,000 observations, including one dependent variable and four independent variables designed to exhibit high correlation. The analysis begins with model estimation using the Ordinary Least Squares (OLS) method, followed by multicollinearity testing using the Variance Inflation Factor (VIF). The OLS results indicate that most independent variables significantly influence the dependent variable, with a coefficient of determination (R²) of 0.9863. However, the high VIF values reveal the presence of strong multicollinearity in the model. To address this issue, Ridge Regression is applied, with the optimal penalty parameter determined through cross-validation, yielding a lambda value of 4.201589. The results show that the regression coefficients in the Ridge model undergo shrinkage, resulting in greater stability compared to the OLS estimates. Model evaluation indicates that the Mean Squared Error (MSE) for the OLS model is 24.77, whereas the Ridge model produces an MSE of 29.72. Although the Ridge model exhibits a slightly higher MSE, it effectively mitigates the impact of multicollinearity and provides more stable parameter estimates.
Growth Model Study Using a Comparison of Gompertz, Logistic, and Weibull Models Indah Suciati; Vina Nurmadani; Yoga Aji Sukma; Linda Rassiyanti
Sciencestatistics: Journal of Statistics, Probability, and Its Application Vol. 3 No. 2 (2025): JULY
Publisher : Universitas Muhammadiyah Metro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24127/sciencestatistics.v3i2.9423

Abstract

Coronavirus Disease or COVID-19 has been a concern for the world, including Indonesia. The very rapid transmission of COVID-19 has had a wide impact on all communities around the world, especially Indonesia. To see the transmission of COVID-19 cases, which continues to increase rapidly, we can use a growth model. The growth model is a non-linear regression model that is used to describe growth behavior. These models can be exponential, sigmoidal, or S-shaped curves. The purpose of this study was to determine the growth curve model of positive COVID-19 cases in Indonesia using the Gompertz, Logistic, and Weibull models. After that, the model evaluation will be carried out using the coefficient of determination as a parameter, so that the best model will be obtained that can predict more accurately the growth of positive COVID-19 cases in Indonesia. The best model that can predict the growth of positive COVID-19 cases in Indonesia is the Gompertz model, with a coefficient of determination is 0.99064.
Perbandingan Estimator Robust Huber dan Tukey’s Biweight terhadap Berbagai Skema Pencilan dalam Regresi Linier Linda Rassiyanti; Indah Suciati; Vina Nurmadani; Yoga Aji Sukma
Sciencestatistics: Journal of Statistics, Probability, and Its Application Vol. 3 No. 2 (2025): JULY
Publisher : Universitas Muhammadiyah Metro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24127/sciencestatistics.v3i2.9630

Abstract

Regresi linier secara umum menggunakan pendekatan Ordinary Least Squares (OLS) namun sering kali mengalami gangguan ketika data mengandung pencilan (outlier), yang dapat menyebabkan estimasi parameter menjadi bias dan tidak akurat. Regresi robust dikembangkan untuk mengatasi kelemahan OLS dengan menurunkan sensitivitas terhadap pencilan. Terdapat dua fungsi kerugian yang sering digunakan dalam regresi robust, yaitu Huber Loss dan Tukey’s Biweight Loss. Penelitian ini bertujuan untuk membandingkan performa dua metode regresi robust, yaitu Huber Loss dan Tukey’s Biweight, dalam menghadapi berbagai skema pencilan. Data simulasi dibangkitkan dengan parameter intersep dan slope masing-masing sebesar 3 dan 2, kemudian ditambahkan pencilan secara sistematis pada variabel X, Y, maupun keduanya, dengan proporsi 10%, 20%, dan 30%. Hasil analisis menunjukkan bahwa Tukey’s Biweight memberikan estimasi parameter yang lebih stabil pada kondisi pencilan ekstrem, terutama saat pencilan terjadi pada variabel Y atau kombinasi X dan Y. Sedangkan, Huber Loss cenderung menghasilkan Mean Squared Error (MSE) yang lebih rendah dalam beberapa kondisi, mencerminkan adanya trade-off antara bias dan variansi. Dengan demikian, Tukey’s Biweight lebih cocok untuk pencilan ekstrem, sedangkan Huber Loss lebih efisien dalam kondisi pencilan ringan hingga sedang. Linear regression, commonly estimated using the Ordinary Least Squares (OLS) method, is known for its sensitivity to outliers, which can lead to biased and inefficient parameter estimates. Robust regression was developed to overcome the weaknesses of OLS by reducing sensitivity to outliers. Two commonly used loss functions in robust regression are Huber Loss and Tukey’s Biweight Loss. This study aims to compare the performance of these two robust regression methods—Huber Loss and Tukey’s Biweight—in handling various outlier scenarios. Simulated data were generated with intercept and slope parameters set at 3 and 2, respectively, and outliers were systematically introduced to the X variable, the Y variable, or both, in proportions of 10%, 20%, and 30%. The analysis results indicate that Tukey’s Biweight provides more stable parameter estimates under extreme outlier conditions, especially when outliers occur in the Y variable or in both X and Y. Meanwhile, Huber Loss tends to yield lower Mean Squared Error (MSE) in certain conditions, reflecting a classic trade-off between bias and variance. Therefore, Tukey’s Biweight is more suitable for extreme outliers, whereas Huber Loss is more efficient under mild to moderate outlier conditions.
Modeling and Predicting Indonesia’s Inflation Using the ARIMA Model Linda Rassiyanti; Rohmi Dyah Astuti; Yuliana
Sciencestatistics: Journal of Statistics, Probability, and Its Application Vol. 4 No. 1 (2026): JANUARY
Publisher : Universitas Muhammadiyah Metro

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

Abstract

Inflation is one of the most important macroeconomic indicators used to evaluate the stability and performance of a country's economy. This study aims to model and predict Indonesia’s monthly inflation rate using the Autoregressive Integrated Moving Average (ARIMA) approach. The dataset consists of monthly inflation observations from January 2010 to December 2025 obtained from Bank Indonesia. The analysis begins with testing the stationarity of the series using the Augmented Dickey–Fuller (ADF) test, followed by model identification through the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) plots. Several candidate models are estimated, including ARIMA (0,1,1), ARIMA (1,1,0), and ARIMA (1,1,1). Model comparison based on the Akaike Information Criterion (AIC) indicates that the ARIMA (0,1,1) model provides the lowest AIC value and is therefore selected as the most appropriate model. The forecasting results suggest that Indonesia’s inflation rate is expected to remain relatively stable at around 3.63% over the next six periods. However, the prediction intervals become wider as the forecasting horizon increases, reflecting growing uncertainty in longer-term predictions.
Klasifikasi Multikelas Varietas Kacang Kering Menggunakan Metode Hybrid SVM Berbasis DAG Dinda Nababan; Christyan Tamaro Nadeak; Linda Rassiyanti; Fajri Farid
MDP Student Conference Vol 5 No 2 (2026): The 5th MDP Student Conference 2026
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/mdp-sc.v5i2.14094

Abstract

This study analyzes the performance of three conventional SVM strategies, namely One-vs-One (OvO), One-vs-Rest (OvR), and Directed Acyclic Graph OvO (DAG-OvO), compared with the hybrid approach Directed Acyclic Graph Rest-vs-Rest (DAG-RvR) in the context of multiclass classification using the Dry Bean Dataset. All models are evaluated based on accuracy and macro metrics to measure the consistency of predictions between classes. The results show that both conventional and hybrid methods achieve the same high level of accuracy, namely 0.92, with Precision, Recall, and F1-score Macro values ​​that were also identical between approaches. The main difference between the approaches lies in computational efficiency. OvO and DAG-OvO show the fastest training time, while DAG-RvR is the most efficient method in the inference stage. These findings confirm that the hybrid DAG-RvR structure can accelerate the prediction process without compromising accuracy, making it worthy of consideration for applications that require fast inference.
Klasifikasi Multikelas Support Vector Machine dengan Hibrida Directed Acyclic Graph One Vs One dan Rest Vs Rest pada Klasifikasi Tingkat Obesitas Daffa Ahmad Naufal; Christyan Tamaro Nadeak; Linda Rassiyanti; Fajri Farid
MDP Student Conference Vol 5 No 2 (2026): The 5th MDP Student Conference 2026
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/mdp-sc.v5i2.14097

Abstract

This research is focused on analyzing how well different multiclass Support Vector Machine (SVM) classification methods can predict obesity levels. It also presents a new hybrid Directed Acyclic Graph Rest-vs-Rest (DAG-RvR) method as a better option. The study utilizes a dataset called the Obesity Risk Prediction Cleaned, which has information on seven different obesity categories. The methods being assessed include One-vs-One (OvO), One-vs-Rest (OvR), DAG-One-vs-One (DAG-OvO), and the new DAG-RvR method. For fine-tuning the parameters, GridSearchCV and the RBF kernel were used. The findings reveal that DAG-RvR achieves an accuracy of 0.91, which is similar to OvO and DAG-OvO, but it trains much quicker, taking just 0.3422 seconds. Even though its precision, recall, and F1-score are a bit lower than the pairwise methods, DAG-RvR still maintains reliable multiclass performance. In summary, this method strikes a good balance between achieving high accuracy and being efficient in computations.
Klasifikasi Varietas Beras Menggunakan Hybrid SVM Berbasis DAG–OVO dan RVR Marleta Cornelia Leander; Christyan Tamaro Nadeak; Linda Rassiyanti; Fajri Farid
MDP Student Conference Vol 5 No 2 (2026): The 5th MDP Student Conference 2026
Publisher : Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/mdp-sc.v5i2.14108

Abstract

This research proposes a hybrid Support Vector Machine (SVM) strategy for multiclass rice variety classification by combining Directed Acyclic Graph Rest-vs-Rest (DAG-RvR) with K-Means clustering. Five rice varieties were analyzed using 16 morphological and texture features extracted from the Rice Image Dataset. Three conventional SVM methods—One-vs-One (OvO), One-vs-Rest (OvR), and DAG-OvO—were evaluated as baselines. Two hybrid schemes were then developed: DAG-RvR K-Means–OvO and DAG-RvR K-Means–K-Means. Experimental results show that all methods achieve high accuracy of approximately 99%, indicating strong feature separability among rice varieties. However, the proposed DAG-RvR K-Means–OvO provides the most efficient performance, achieving the fastest training time while maintaining competitive testing speed and the highest accuracy of 0.99040. The findings demonstrate that integrating K-Means–based class partitioning with pairwise SVM classification improves computational efficiency without reducing predictive performance, making the hybrid approach suitable for fast and accurate multiclass classification tasks.
ANALISIS HUBUNGAN ANTARA ANGKA TIDAK SEKOLAH (ATS), TINGKAT KEMISKINAN, DAN ANGKA MELEK HURUF (AMH) DENGAN BIPLOT Linda Rassiyanti; Ayu Sofia; Rizka Pitri
MATHunesa: Jurnal Ilmiah Matematika Vol. 13 No. 02 (2025)
Publisher : Universitas Negeri Surabaya

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

Abstract

Tujuan dari penelitian ini adalah untuk memperdalam analisis tentang hubungan antara pendidikan, literasi, dan kemiskinan. Metode penelitian yang digunakan pada penelitian ini adalah metode analisis biplot yang merupakan teknik statistik deskriptif yang memungkinkan penyajian visual simultan dari objek pengamatan dan variabel dalam ruang dua dimensi. Data yang digunakan dalam penelitian ini adalah Rata-Rata Lama Sekolah (RLS), Angka Melek Huruf (AMH), dan Persentase Orang Miskin. Tiap variabel dilakukan analisis desktiptif untuk mengetahui sebaran datanya, lalu selanjutnya dilakukan analisis biplot, dan menarik kesimpulan. Hasil penelitian menunjukkan bahwa RLS dan AMH memiliki hubungan yang kuat dengan tingkat kemiskinan (persentase orang miskin). Tingkat pendidikan yang lebih tinggi secara umum berkorelasi dengan tingkat kemiskinan yang lebih rendah. Provinsi DKI Jakarta dan Kep. Riau memiliki keberhasilan dalam taraf pendidikan yang tinggi dan kemiskinan rendah, sedangkan Provinsi Papua Pegunungan dan Nusa Tenggara Timur masih memerlukan perhatian untuk mengurangi kemiskinan dan meningkatkan taraf pendidikan. Provinsi Papua Pegunungan dan Papua Tengah terlihat jauh dari kelompok lainnya, menunjukkan karakteristik yang unik, seperti kemiskinan sangat tinggi dan pendidikan sangat rendah.
Workshop Pembuatan Sistem Monitoring Jaringan Sederhana Menggunakan Python bagi Siswa SMKN 4 Bandar Lampung M. Syamsuddin Wisnubroto; Yuliana Yuliana; Linda Rassiyanti; Ade Lailani; Fajri Farid; Christyan Tamaro Nadeak; Fitri Nurjanah; Indah Suciati; Rian Kurnia; Yusni Puspha Lestari; Dewi Indra Setiawan
BERDAYA: Jurnal Pendidikan dan Pengabdian Kepada Masyarakat Vol. 8 No. 2 (2026)
Publisher : LPMP Imperium

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36407/berdaya.v8i2.1810

Abstract

This community service program was conducted to strengthen vocational students' competencies in network monitoring using Python. The partner school's main challenge was that networking lessons were still centered on hardware-oriented tools such as Mikrotik and Cisco, while software-based monitoring skills had not been introduced systematically. The program took the form of a workshop at SMKN 4 Bandar Lampung on 24 September 2025 and combined short lectures, demonstrations, guided practice, and mini projects. The training module covered basic networking concepts, connectivity and server status, bandwidth and latency, Python fundamentals, and the use of requests, psutil, socket, subprocess, and pandas to build a simple network monitoring system. Evaluation was conducted descriptively using pre-tests, post-tests, and practical assessment. The results showed that the mean pre-test accuracy of 46% from 32 participants increased to 64% in the post-test completed, representing an 18 percentage-point gain. All students also completed the assigned Python-based monitoring practice successfully. The outputs included a training module, poster, short video, and press release to support sustainability and dissemination.
Optimalisasi Pembelajaran Digital melalui Pelatihan Google Colab untuk Meningkatkan Kompetensi Mahasiswa Ayu Sofia; Tiara Yulita; Dwi Mahrani; Linda Rassiyanti; Rohmi Dyah Astuti; Fajri Farid; Yuliana Yuliana; Ade Lailani
Jurnal Pengabdian Masyarakat Bangsa Vol. 4 No. 8 (2026): Oktober
Publisher : Amirul Bangun Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59837/jpmba.v4i8.5328

Abstract

Perkembangan teknologi digital menuntut mahasiswa memiliki kompetensi yang memadai dalam bidang pemrograman dan analisis data. Namun, pada Program Studi Sains Data UIN Raden Intan Lampung yang baru berdiri dan mulai menerima mahasiswa pada Tahun Akademik 2024/2025, kurikulum yang berjalan belum memuat mata kuliah khusus komputasi cloud, di samping adanya kendala keterbatasan akses perangkat lunak berlisensi dan variasi spesifikasi perangkat keras mahasiswa. Kegiatan Pengabdian kepada Masyarakat (PKM) ini bertujuan untuk meningkatkan kompetensi mahasiswa Program Studi Sains Data UIN Raden Intan Lampung melalui pelatihan penggunaan Google Colab sebagai platform pembelajaran berbasis cloud. Metode pelaksanaan menggunakan pendekatan pelatihan partisipatif (participatory training) melalui hands-on workshop yang mencakup tahap persiapan, pelaksanaan workshop interaktif, serta evaluasi menggunakan instrumen pre-test dan post-test. Materi yang diberikan mencakup pengenalan Google Colab, dasar-dasar pemrograman Python, pengolahan data dengan Pandas, serta visualisasi data. Hasil evaluasi menunjukkan adanya peningkatan kompetensi peserta setelah mengikuti pelatihan. Hasil uji Paired Samples T-Test menunjukkan bahwa pelatihan memberikan pengaruh yang signifikan terhadap peningkatan kompetensi peserta. Dengan demikian, pelatihan Google Colab efektif dalam mendukung optimalisasi pembelajaran digital serta meningkatkan kesiapan mahasiswa dalam menghadapi kebutuhan dunia kerja berbasis teknologi.