Claim Missing Document
Check
Articles

Found 20 Documents
Search

Diabetes risk prediction using logistic regression model Rassiyanti, Linda; Farid, Fajri; Pitri, Rizka
Desimal: Jurnal Matematika Vol. 8 No. 1 (2025): Desimal: Jurnal Matematika
Publisher : Universitas Islam Negeri Raden Intan Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/djm.v8i1.26493

Abstract

This study aims to analyze the factors that contribute to diabetes using the logistic regression method. The data used in this study include variables of number of pregnancies, glucose levels, blood pressure, skin thickness, insulin levels, body mass index (BMI), family history of diabetes, and age. The logistic regression model was applied to determine the effect of each variable on the likelihood of a person having diabetes. Evaluation of model performance was carried out using the ROC (Receiver Operating Characteristic) curve, and the results obtained showed an AUC value of 0.8391, which indicated a very good classification ability of the model. The results of the analysis showed that the number of pregnancies, glucose levels, blood pressure, BMI, and family history of diabetes had a significant effect on the risk of diabetes.
Multi-objective bees algorithm for portfolio diversification Farid, Fajri; Linda Rassiyanti; Rohmi Dyah Astuti; Ade Lailani
Desimal: Jurnal Matematika Vol. 8 No. 2 (2025): Desimal: Jurnal Matematika
Publisher : Universitas Islam Negeri Raden Intan Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/mttkw723

Abstract

Portfolio diversification is the practice of spreading investments across different types of stocks or sectors to reduce overall risk. The basic principle is that the poor performance of one stock asset can be offset by the satisfactory performance of another stock asset. This study uses the Bees Algorithm for portfolio optimization problems, aiming to discover the combination of stock proportions in a portfolio that maximizes stock returns and minimizes risk. Then, the Sharpe ratio value is calculated and compared with conventional methods. The expected return, risk, and Sharpe ratio values for the portfolio generated using the Bees algorithm are 0.178007%, 2.353956%, and 0.0663484322, respectively. According to the results, the Bees Algorithm had better results and performance than conventional methods. As a result, the Bees Algorithm outperforms conventional approaches.
Penerapan Metode Least Significant Bit untuk Penyembunyian Pesan Rahasia dalam Gambar dengan Optimasi Ukuran File Yuliana; Rohmi Dyah Astuti; Ade Laelani; Linda Rassiyanti; Yusni Puspha Lestari; Ronal
Jurnal ICT: Information Communication & Technology Vol. 25 No. 1 (2025): JICT-IKMI, July, 2025
Publisher : LPPM STMIK IKMI Cirebon

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

Abstract

Steganografi merupakan teknik untuk menyembunyikan pesan rahasia dalam media digital guna menjaga kerahasiaan informasi. Penelitian ini menerapkan metode penyisipan pesan menggunakan algoritma Least Significant Bit (LSB), di mana pesan rahasia disisipkan langsung ke dalam bit-bit paling tidak signifikan dari piksel citra grayscale. Setelah proses penyisipan, citra stego disimpan dalam tiga format berbeda—PNG, WebP, dan ZIP—untuk mengevaluasi dampak kompresi terhadap integritas pesan dan kualitas citra. Evaluasi dilakukan berdasarkan empat parameter: ukuran file, degradasi kualitas citra (PSNR), kesamaan struktur visual (SSIM), dan keberhasilan ekstraksi pesan. Hasil menunjukkan bahwa format PNG mampu mempertahankan kualitas citra dan integritas pesan secara optimal (PSNR 75,58 dB, SSIM 1,0000). Sebaliknya, kompresi lossy pada WebP mengganggu bit pesan sehingga menyebabkan pesan rusak. Format ZIP terbukti dapat mempertahankan file stego secara utuh. Penelitian ini menunjukan bahwa steganografi berbasis Least Significant Bit tetap efektif bila dikombinasikan dengan format gambar lossless. Format lossy seperti WebP tidak disarankan karena berisiko merusak data.
Analisis Regresi Kernel Gaussian untuk Memprediksi Indeks Pembangunan Manusia (IPM) Berdasarkan Faktor Sosial-Ekonomi Provinsi di Indonesia Rohimatul Anwar; Linda Rassiyanti; Rizka Pitri
JURNAL RISET RUMPUN MATEMATIKA DAN ILMU PENGETAHUAN ALAM Vol. 4 No. 3 (2025): Desember : 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.v4i3.7017

Abstract

The Human Development Index (HDI) functions as a key indicator for assessing the level of welfare and overall quality of life of the population within a specific region. This study aims to examine the socio-economic factors influencing HDI at the provincial level in Indonesia using a Gaussian kernel regression approach. A nonparametric method is employed due to its flexibility in capturing nonlinear relationships between the response and predictor variables without the need to assume a specific functional form. The analysis utilizes secondary data, including education, poverty, per capita expenditure, expected years of schooling, open unemployment rate, and gross regional domestic product for each Indonesian province. The findings from this study indicate that educational factors, particularly mean years of schooling and expected years of schooling, exert the most significant impact on HDI improvement. The estimated Gaussian kernel regression model demonstrates a coefficient of determination of 0.9954 and a residual standard error of 0.3468, reflecting a very high predictive accuracy and relatively low error. These results suggest that Gaussian kernel regression is an effective nonparametric approach for analyzing human development in Indonesia.
Hubungan Sebab-Akibat Kecerdasan Interpersonal Terhadap Hasil Belajar Peserta Didik: SEM-PLS Pitri, Rizka; Irawan, Miranda Dimas; Rassiyanti, Linda
Lattice Journal : Journal of Mathematics Education and Applied Vol. 4 No. 2 (2024): Desember 2024
Publisher : Universitas Islam Negeri Sjech M. Djamil Djambek Bukittinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30983/lattice.v4i2.9069

Abstract

Interpersonal intelligence is a person's ability to work together and understand between individuals with others. The utilization of interpersonal intelligence in mathematics learning will have an impact on improving the quality of learning outcomes. However, currently the social interaction that occurs in educational units is still low. Thus, resulting in low quality of learning outcomes. Based on this, an analysis is needed to measure the causal relationship between interpersonal intelligence and learning outcomes using Structural Equation Modeling (SEM)-PLS. This study used 75 samples of grade X students at SMAN Martapura. The purpose of this research is to see the cause-and-effect relationship between interpersonal intelligence and learning outcomes. The method used was SEM-PLS method. The instruments used in the study consisted of essay questions and interpersonal intelligence questionnaire. This study shows that there is a significant causal relationship between interpersonal intelligence and students' learning outcomes in mathematics.   Kecerdasan interpersonal merupakan suatu kemampuan seseorang dalam bekerja sama dan memahami antar individual dengan yang lainnya. Pemanfaatan kecerdasan interpersonal dalam pembelajaran matematika akan berdampak pada peningkatan kualitas hasil belajar. Namun, saat ini interaksi sosial yang terjadi di satuan pendidikan masih rendah. Sehingga, mengakibatkan rendahnya kualitas hasil belajar. Berdasarkan hal tersebut, diperlukan suatu analisis untuk mengukur hubungan sebab-akibat antara kecerdasan interpersonal terhadap hasil belajar menggunakan Structural Equation Modeling (SEM)-PLS. Penelitian ini menggunakan 75 sampel siswa kelas X di SMAN Martapura. Tujuan penelitian ini adalah melihat hubungan sebab-akibat antara kecerdasan interpersonal terhadap hasil belajar. Metode yang digunakan adalah metode SEM-PLS. Instrumen yang digunakan dalam penelitian terdiri dari soal essay dan angket kecerdasan interpersonal. Penelitian ini menunjukkan bahwa terdapat hubungan sebab-akibat yang signifikan antara kecerdasan interpersonal terhadap hasil belajar peserta didik di bidang matematika.
Analisis Komparasi Model Peramalan Prophet Dan Arima Dalam Memprediksi Harga Saham Penutupan PT ANTM Anwar, Rohimatul; Rassiyanti, Linda
Lattice Journal : Journal of Mathematics Education and Applied Vol. 5 No. 1 (2025): Juni 2025
Publisher : Universitas Islam Negeri Sjech M. Djamil Djambek Bukittinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30983/lattice.v5i1.9478

Abstract

PT Aneka Tambang Tbk (ANTM) is a major mining company in Indonesia whose shares are actively traded on the Indonesia Stock Exchange. Its stock price is influenced by both internal and external factors. Time series forecasting methods, such as ARIMA and Prophet, are used to predict future stock movements. ARIMA is known for its flexibility and high prediction accuracy, while Prophet is capable of handling missing values and shifting trends, making it suitable for complex financial data. This study aims to compare the performance of ARIMA and Prophet models in forecasting ANTM stock prices. The dataset consists of monthly closing stock prices from January 2016 to May 2025. The models are evaluated using Mean Absolute Percentage Error (MAPE), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). The results show that the ARIMA model performs better to predict PT ANTM’s traded than Prophet model, with lower MAPE, AIC, and BIC values of 7.88, 433.24, and 423.63, respectively.   PT Aneka Tambang Tbk (ANTM) adalah perusahaan pertambangan besar di Indonesia yang sahamnya aktif diperdagangkan di Bursa Efek Indonesia. Harga sahamnya dipengaruhi oleh faktor internal dan eksternal.  Memprediksi pergerakan harga saham di masa depan bagi PT ANTM dapat digunakan metode peramalan deret waktu seperti ARIMA dan Prophet. ARIMA dikenal fleksibel dan mampu memberikan hasil prediksi yang akurat. Sementara itu, Prophet unggul dalam menangani data yang memiliki nilai hilang dan tren yang berubah, menjadikannya cocok untuk data pasar keuangan yang dinamis. Penelitian ini bertujuan membandingkan performa model ARIMA dan Prophet dalam memprediksi harga saham ANTM. Data yang digunakan adalah harga penutupan saham bulanan selama periode Januari 2016 hingga Mei 2025. Perbandingan dilakukan berdasarkan nilai Mean Absolute Percentage Error (MAPE), Akaike Information Criterion (AIC), dan Bayesian Information Criterion (BIC). Hasil analisis menunjukkan bahwa model ARIMA memiliki performa lebih baik dengan nilai MAPE 7.88, AIC 433.24, dan BIC 423.63, yang lebih rendah dibandingkan dengan model Prophet.
Application of the DAG-SVM for multi-class mobile phone price classification Sihombing, Natanael Oktavianus Partahan; Christyan Tamaro Nadeak; Linda Rassiyanti; Fajri Farid
Desimal: Jurnal Matematika Vol. 8 No. 3 (2025): Desimal: Jurnal Matematika
Publisher : Universitas Islam Negeri Raden Intan Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/djm.v8i3.202529450

Abstract

This study investigated the application of multiclass Support Vector Machine (SVM) strategies for smartphone price range classification using the Mobile Price Classification dataset (N = 2,000). The aim was to assess whether the Directed Acyclic Graph SVM (DAG-SVM) could provide improvements in predictive performance or computational efficiency compared with the conventional One-vs-One (OvO) and One-vs-Rest (OvR) approaches. The dataset’s twenty features were standardized using Z-score normalization and split into training and testing sets with an 80:20 ratio. All models were implemented using a linear kernel and evaluated based on accuracy, macro-precision, macro-recall, macro-F1, and execution time. The results showed that both OvO and DAG-SVM achieved the highest performance, with an accuracy and macro-F1 score of 96.25%, while OvR performed substantially lower. Despite the theoretical efficiency of DAG-SVM, its Python-based sequential elimination process led to slower prediction time than OvO. This study contributed empirical evidence that execution time can diverge from theoretical expectations in practical implementations and demonstrated the importance of computational efficiency analysis when comparing multiclass SVM architectures for mobile price classification.
Enhancing multiclass SVM classification using a hybrid directed acyclic graph and rest-vs-rest strategy Nadeak, Christyan Tamaro; Farid, Fajri; Rassiyanti, Linda; Siahaan, Arielva Simon; Putri, Lutfia Aisyah
Desimal: Jurnal Matematika Vol. 8 No. 3 (2025): Desimal: Jurnal Matematika
Publisher : Universitas Islam Negeri Raden Intan Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/djm.v8i3.202529469

Abstract

This study proposes a modified Directed Acyclic Graph Support Vector Machine (DAG-SVM) using a Rest-vs-Rest (RvR) strategy to address the multiclass classification problem in the Hepatitis C dataset from Kaggle, which contains four diagnostic categories with a highly imbalanced class distribution, with class sample sizes of 540, 24, 21, and 30, respectively. The aim of this study is to examine how hierarchical decision structures interact with extreme class imbalance in SVM-based multiclass classification. The method is implemented through three fixed hierarchical decision schemes {0,1} vs. {2,3}, {0,2} vs. {1,3}, and {0,3} vs. {1,2} which restructure the decision flow of conventional DAG-SVM. Experimental evaluation shows that although the proposed schemes achieve relatively high overall accuracy (0.91–0.93), the precision, recall, and F1-scores for minority classes remain extremely low. These findings offer a new empirical insight into how class imbalance propagates through the DAG hierarchy, leading to early elimination of minority classes, and highlight the need for imbalance-handling techniques such as resampling, cost-sensitive learning, or synthetic data generation. The contribution of this work lies in demonstrating the limitations of DAG-RvR under severe imbalance and providing a structured evaluation that can guide future improvements for more reliable multiclass recognition.
ANALISIS KLASIFIKASI KUALITAS HIDUP MANUSIA ANTAR KABUPATEN/KOTA DI INDONESIA MENGGUNAKAN ALGORITMA CATBOOST CLASSIFIER DAN SHAP VALUES Ayu Sofia; Linda Rassiyanti
MATHunesa: Jurnal Ilmiah Matematika Vol. 14 No. 01 (2026)
Publisher : Universitas Negeri Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/mathunesa.v14n1.p430-444

Abstract

Human development must go hand in hand with improving quality of life, as reflected by HDI and its influencing factors. Classifying quality of life based on HDI into developed, developing, and underdeveloped areas offers insights into the human development performance of each district/city in Indonesia. The CatBoost Classifier and SHAP values help build an accurate model while interpreting variable influences. This study analyzes human quality of life classification across districts/cities based on HDI and related factors. The CatBoost model achieved 92.23% accuracy, with the best performance in the developing class, while the underdeveloped class showed low accuracy due to data imbalance. SHAP analysis revealed that average years of schooling, per capita expenditure, and region type were key variables in the developed and developing classes, while island location and sanitation access dominated in the underdeveloped class. These findings highlight the importance of education, economic welfare, and basic infrastructure in shaping quality of life. This research also supports actuarial social risk planning, particularly in designing data- and region-based social security systems.
Empowering the Future: AI-Based Website Development Training to Boost High School Students' Creativity and Digital Skills Luluk Muthoharoh; Yuliana Yuliana; Linda Rassiyanti; Ade Lailani; Ayu Sofia; Tiara Yulita; Dharu Cahyoaji Sasongko; Khairunnisa Maharani; Aditya Rahman
Smart Society Vol. 5 No. 1 (2025): Smart Society
Publisher : FOUNDAE (Foundation of Advanced Education)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58524/smartsociety.v5i1.658

Abstract

In the rapidly evolving digital era, Artificial Intelligence (AI) has become one of the key technologies playing a significant role in various fields, including website development. AI-based website development training is a strategic step to enhance the skills and creativity of the younger generation in the era of digitalization. This activity was conducted at SMA Al Huda, South Lampung, with the aim of equipping students with an understanding of basic concepts, steps for creating websites, and how AI can be utilized to enhance creativity in the digital world. The methods used included interactive presentations, demonstrations, hands-on practice, and discussions. The activity also assessed students' knowledge before and after the training through an interactive quiz using Quizizz. The results of the study showed a significant improvement in students' understanding of AI-based website development. In conclusion, this training program was effective in enhancing students' skills and creativity in the digitalization world, particularly in creating AI-based websites