Claim Missing Document
Check
Articles

Sistem Pendukung Keputusan Penentuan Area Pembayaran Pajak Menggunakan Simple Additive Weighting (SAW) di Kota Cimahi Aranda, Sepri; Witanti, Wina; Melina, M
Prosiding Seminar Riset Mahasiswa Vol 1, No 1: Maret 2023
Publisher : Universitas Islam Sultan Agung

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

Abstract

Pungutan wajib daerah (pajak) memegang kunci penting terutama bagi  penghasilan  daerah  di setiap kota. Disini pemerintah  Kota Cimahi punya  tiga kelurahan, dan lima belas kelurahan masih berusha dalammenggali kesanggupan  keuangan  dalam mendanai  pengembangan daerah tersebut. Penerapan pembayaran pemungutan wajib pajak di kota Cimahi terkhusus  di kabupaten yang belum  optimal dalam menelusuri  daerah mana yang membayar iuran daerah dan siapa yang membayar pungutan wajib  di setiap kabupaten. Di penelitian  sistem pendukung keputusan penentuan daerah perpajakan Kota Cimahi dibantu menggunakan Simple Additive  Weighting (SAW). SAW juga termasuk kedalam   suatu metode yang mampu  membantu mencari suatu jalan pintas berlandaskan tipe/kriteria yang ada untuk dipetakan setiap  kelurahan  Kota Cimahi  yang  membayar  pungutan wajibnya.  Lalu proses  pemetaan  pemungutan wajib  pajak  tersebut, maka terciptalah sistem pendukung keputusan berbasis WebGIS yang dapat merincikan suatu wilaya dikota Cimahi, mengetahui pendapatan dan kepatuhan  pungutan iuranwajib di setiap tingkat daerah, kemudian melihat wilayah kota Cimahi yang mana, miliki pemungutan iuran wajib (pajak) terendah yang menperhitungan Sample Additive Weighting (SAW). Hasil penelitian ini dapat digunakan sebagai awal pengambilan keputusan bagi Badan Perencanaan Pembangunan  Daerah  (BAPPEDA )dikota  Cimahi  bisa  mengeperbaharui  serta  mampu  menelaaah lokasi yang  menguntungkan bagi  pemungutan iuran wajib  (pajak) daerahdisetiap  wilayah  Kota  Cimahi.
Prediksi Kinerja Akademik Siswa Bimbingan Belajar Menggunakan Algoritma Extreme Gradient Boosting (XGBoost) Alfarizi, Muhammad Bayu Ardi; Witanti, Wina; Komarudin, Agus
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7387

Abstract

Improving the quality of education has become a primary focus in addressing the increasingly complex challenges of the educational landscape. One promising approach to support data-driven decision-making is the prediction of students' academic performance using machine learning algorithms. This study aims to develop a classification model for predicting students' academic performance by leveraging the Extreme Gradient Boosting (XGBoost) algorithm. The dataset used was obtained from SMPN 1 Gunung Halu and includes both academic and non-academic attributes of students. Five key features were selected: initial grades, midterm grades, final grades, student behavior, and attendance. Data preprocessing involved feature selection, handling missing values, transforming categorical variables using label encoding, and balancing the classes using the SMOTE method. The XGBoost model was then trained using an 80:20 data split and hyperparameter tuning was performed using Grid Search. Evaluation results showed that the model achieved an accuracy of 84% with balanced F1-scores across all classes. The model outperformed other algorithms such as Bagging and Random Forest. With its strong accuracy and stability, the XGBoost model has the potential to serve as a tool for identifying students who require academic intervention. This study makes a significant contribution to the development of AI-based education systems and provides a foundation for the application of machine learning in improving the quality of secondary-level learning.
Prediksi Penyakit Kanker Payudara Menggunakan Algoritma Synthetic Minority Oversampling Technique dan Categorical Boosting Classifier Mandala, Muhamad Bintang; Witanti, Wina; Komarudin, Agus
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7403

Abstract

Breast cancer remains one of the leading causes of mortality worldwide, with high prevalence rates among women in Indonesia. Accurate and efficient diagnostic models are essential to support early detection and reduce mortality. This study aims to develop a predictive model for breast cancer classification using the CatBoost algorithm, a gradient boosting method known for its ability to natively handle categorical features and reduce overfitting through ordered boosting. The dataset used consists of diagnostic features of breast tumors, which were preprocessed by checking completeness and transforming numerical attributes into categorical bins to capture value distribution more effectively. To address class imbalance between benign and malignant cases, the SMOTE (Synthetic Minority Over-sampling Technique) method was applied, resulting in a balanced training set. Optimal hyperparameters for the CatBoost model were obtained using Bayesian optimization, with key parameters including depth, learning rate, and L2 regularization. The model was then trained and evaluated using recall, accuracy, and F1-score metrics, with a confusion matrix used to assess prediction quality. The results demonstrate that CatBoost achieved high performance with a recall of 1,0, accuracy of 98,6%, and F1-score of 0,99, outperforming or matching other benchmark models such as SVM, Neural Network, and XGBoost. These findings highlight the reliability and effectiveness of CatBoost in supporting medical decision-making for breast cancer diagnosis.
PRIMARY QUERY ANALYSIS ON SQL DATABASE RESTRUCTURING IN GEOGRAPHIC INFORMATION SYSTEMS Ilyas, Ridwan; Witanti, Wina; Syarafina, Fildzah
JIKO (Jurnal Informatika dan Komputer) Vol 7, No 2 (2024)
Publisher : Universitas Khairun

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33387/jiko.v7i2.8565

Abstract

Database restructuring is a crucial process aimed at enhancing data management and access efficiency by modifying the existing data structure. This research focuses on improving a Geographic Information System (GIS) for taxation by migrating and restructuring an inefficient and redundant database. The study conducts a comparative performance evaluation of the old and restructured databases using benchmarking tests with varying numbers of threads and ramp-ups. The results reveal a significant increase in average throughput (24.60%) following the restructuring, indicating a substantial improvement in the database's data processing capacity. However, there is also an average increase in response time (21.65%), suggesting a trade-off between enhanced throughput and slower response times. This increase in response time indicates that while the system can handle more data, it requires more time to process each query. Overall, the restructured database demonstrates enhanced performance and efficiency, though further optimization is necessary to achieve consistent throughput across different workloads and to mitigate the increased response times
Analisis Sentimen Ulasan Aplikasi CapCut Menggunakan Model RoBERTa Dengan Fitur Ekstraksi Word2vec Budiman, Firman Nur; Witanti, Wina; Nurul Sabrina, Puspita
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2480

Abstract

To improve the accuracy of sentiment classification in CapCut app reviews, this study tested a hybrid model built from a combination of RoBERTa and Word2Vec. A total of 5,000 reviews from the Google Play Store were used as a dataset, which was then processed through data cleaning, tokenization, and stopword removal stages. Next, the EDA oversampling technique was used to address the issue of class distribution imbalance. The proposed model architecture works by combining the concatenation of vector features from Word2Vec for local word meaning representation and RoBERTa for overall sentence context understanding. Model evaluation showed an accuracy of 80%, a higher result compared to the 79% accuracy obtained by the single RoBERTa baseline model. This study concludes that combining contextual and semantic feature representations effectively results in better sentiment classification performance.
Pola Pembelian Konsumen Supermarket Menggunakan Algoritma ECLAT Dan Fp-Growth Fahrezi Ahmad, Rafly Fikri; Witanti, Wina; Ramadhan, Edvin
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2482

Abstract

This study aims to uncover consumer purchasing patterns in supermarkets to support more targeted sales strategies. The primary focus is on identifying products that are frequently bought together and their relationship with contextual factors such as payment methods, seasons, and discount status. The main challenge lies in handling transactional data that is highly diverse (high cardinality) and sparsely co-occurring, necessitating an approach capable of generating relevant association patterns. To address this, the study implements an integrated approach combining the ECLAT and FP-Growth algorithms in Market Basket Analysis. ECLAT is employed to filter items with low frequency through a TID-List structure, resulting in a more focused set of transactional data for FP-Growth processing. FP-Growth is then used to identify frequently co-occurring product and attribute combinations and to form association rules based on support, confidence, and lift values. The research data comprises 10,000 transactions with 13 attributes, focusing on Product, Payment_Method, Discount_Applied, Season, and City. As a result, ECLAT successfully filtered 81 products and 101 frequently occurring contextual attributes. FP-Growth subsequently discovered 407 itemset patterns, with 13 valid patterns forming association rules between products and contextual attributes. Additionally, three-item patterns were found for watch products associated with discounts and seasons. The contribution of this study lies in demonstrating that the integration of ECLAT and FP-Growth can serve as an effective method for discovering consumer shopping patterns based on transactional context, thereby supporting data-driven business decision-making.
Desain Antarmuka E-Commerce Tema Natal untuk Layanan Virtual Office Menggunakan Design Thinking Adhinata Kusuma, Rendy; Witanti, Wina; Yuniarti, Rezki
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2492

Abstract

Baros Information Technology Creative (BITC) is a virtual office service provider based in Cimahi that requires adaptive, engaging, and seasonally contextual user interface (UI) and user experience (UX) support to enhance user appeal and engagement. This study aims to design the UI/UX of BITC’s virtual office e-commerce website with a Christmas theme using the Design Thinking methodology. This approach was selected for its iterative and user-centered nature, enabling in-depth exploration of user needs and the development of contextually relevant design solutions. The design process was carried out through five stages—empathize, define, ideate, prototype, and test—involving five respondents selected through purposive sampling. User requirements were gathered via interviews and surveys, then translated into design concepts visualized through interactive prototypes. Usability testing was conducted using the System Usability Scale (SUS), adapted to suit the seasonal theme context. The test results yielded an average SUS score of 82, categorized as “good” with an “excellent” level of acceptance, indicating that the resulting design is user-friendly, visually relevant, and supports seasonal promotional strategies. This study reinforces previous findings that user-centered design approaches are effective in improving interface quality and user experience, while also offering a novel contribution through the application of seasonal themes within the context of professional digital services.
Optimasi Prediksi Penjualan Retail Online Menggunakan LightGBM dan Hyperparameter Tuning Nailendra, Septian Yudha; Witanti, Wina; Abdillah, Gunawan
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2551

Abstract

This research aims to develop and optimize a daily sales prediction model based on time series data using the Light Gradient Boosting Machine (LightGBM) algorithm on online retail data from the Olist marketplace. The research process began with merging and aggregating e-commerce transaction data into a daily format, followed by outlier handling using the Interquartile Range (IQR) capping method, and feature engineering to add temporal and historical information such as prev_day_sales and day_of_week. The dataset was then split into training and testing sets using a time-based split approach. A baseline model was trained with default parameters and subsequently optimized through hyperparameter tuning using GridSearchCV with TimeSeriesSplit cross-validation. Evaluation was conducted using MAE, RMSE, and R² metrics. The results show that the tuned model improved prediction accuracy, with MAE reduced by 5.34%, RMSE decreased by 8.34%, and R² increased by 0.76%. The one-day-ahead daily sales prediction reached R$ 1,676.86 and closely followed the actual sales pattern. This study demonstrates that a systematic approach involving data preprocessing, feature engineering, and parameter tuning can produce a more accurate, stable, and practical sales prediction model to support decision-making in the e-commerce sector. Theoretically, this research contributes to strengthening the understanding of the effectiveness of the LightGBM algorithm in daily time series modeling, particularly through the integration of temporal feature engineering and systematic parameter tuning strategies. These findings underscore the importance of a comprehensive approach in building accurate sales prediction models.
Klasterisasi Gaya Belajar Mahasiswa Berbasis VARK dengan Algoritma DBSCAN untuk Personalisasi E-Learning Maulana, Iqbal; Witanti, Wina; Melina
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2980

Abstract

The incompatibility between e-learning systems and students' learning styles remains a major challenge in improving the effectiveness of learning in Indonesian universities. This study aims to classify the learning styles of students at Jenderal Achmad Yani University using the VARK (Visual, Auditory, Read/Write, Kinesthetic) model, enriched with the Kano method. Data were collected from 1,000 students through the VARK-Kano questionnaire and analyzed using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. The clustering process was carried out by determining the optimal parameters using the k-distance plot, and the validity of the clusters was assessed using the Silhouette Score. The results showed that DBSCAN could form representative clusters of student learning styles and effectively detect data noise. This study contributed to the development of a cluster-based adaptive e-learning framework that could be implemented in Indonesian universities. These findings could serve as a basis for designing adaptive learning strategies that are more suited to student characteristics, thereby increasing the effectiveness of e-learning and learning motivation.