Nelly Khairani Daulay
Universitas Bina Insan

Published : 6 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 6 Documents
Search

ANALISIS SENTIMEN APLIKASI E-GOVERENMENT PADA GOOGLE PLAY MENGGUNKAN ALGORITMA NAÏVE BAYES Rini Sartina; Elmayati; Nelly Khairani Daulay
JURNAL ILMIAH BETRIK Vol. 14 No. 01 APRIL (2023): JURNAL ILMIAH BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : P3M Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/betrik.v14i01 APRIL.29

Abstract

E-gov is one of the innovations created by the government to be able to compete in the digital world. One government agency that has used e-gov is the Ministry of Spatial Planning/National Land Agency (ATR/BPN). in this study the authors used the Naïve Bayes Algorithm as a classification method to determine positive, negative, and also neutral sentiments. The labeling process shows that the touch my land application tends to get a neutral response which can be seen from the comparison of 4000 data, 2265 positive, 1453 negative and 211 neutral data. and the results of sentiment analysis testing using the Naïve Bayes Algorithm produce an accuracy of 81%, 78% precision, and 79% recall.
Densenet201 Feature Extraction With Soft Voting Ensemble For Accurate Rice Leaf Disease Classification Nelly Khairani Daulay; Novi Lestari; Rusdiyanto Rusdiyanto
Jurnal Media Computer Science Vol 5 No 3 (2026): Juli
Publisher : LPPJPHKI Universitas Dehasen Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37676/jmcs.v5i3.11876

Abstract

Rice leaf diseases are one of the major factors contributing to reduced agricultural productivity and economic losses for farmers. Manual disease identification generally requires expert knowledge and is often difficult to perform efficiently in field conditions. Therefore, this study aims to develop a rice leaf disease classification system by combining DenseNet201 as a feature extractor and a Voting Ensemble approach as the classifier. The dataset consisted of 1,470 rice leaf images categorized into five classes: Bacterial Leaf Blight, Brown Spot, Healthy Leaf, Leaf Blast, and Tungro. The dataset was divided using a stratified split strategy into 80% training data, 10% validation data, and 10% testing data. Image augmentation was applied only to the training set, increasing the number of training samples to 7,056 images. DenseNet201 was employed to extract image features into 1,920-dimensional feature vectors, which were subsequently classified using Logistic Regression, Support Vector Machine (SVM), Hard Voting, and Soft Voting. Experimental results showed that Logistic Regression achieved an accuracy of 95.24%, while SVM achieved 95.92%. Hard Voting obtained an accuracy of 95.24%, whereas Soft Voting achieved the best performance with an accuracy of 95.92%, precision of 95.75%, recall of 95.70%, F1-score of 95.71%, and ROC-AUC of 99.76%. Furthermore, the best-performing model was deployed in a Streamlit-based application for automatic rice leaf disease identification. The results demonstrate that the combination of DenseNet201 and Soft Voting provides an accurate and effective approach for rice leaf disease classification and has strong potential as an early disease detection tool in agriculture.
SISTEM PERAMALAN PENJUALAN KOPI BUBUK SELANGIT MENGGUNAKAN METODE WEIGHTED MOVING AVERAGE (WMA) MENGGUNAKAN DATA TIME SERIES BERBASIS FRAMEORK CI (CODEIGNITER) Glen Jupiter; Armanto Armanto; Nelly Khairani Daulay
Jurnal Komputer dan Teknologi Vol 5 No 2 (2026): JUKOMTEK JULI 2026
Publisher : Yayasan Pendidikan Cahaya Budaya Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64626/jukomtek.v5i2.706

Abstract

This study aims to forecast ground coffee sales using the Weighted Moving Average (WMA) method to support decision-making in business planning. The WMA method was selected because it assigns greater weight to recent historical data, thereby enabling a more responsive capture of changes in sales trends. The results indicate that ground coffee sales are projected to experience a stable upward trend in 2025. The model achieved a high level of accuracy, yielding a MAPE of 0.098%, an MAE of 54.463, and an RMSE of 70.683. Although discrepancies occurred in certain periods due to high sales volatility, the WMA method generally tracked actual data patterns effectively and produced realistic estimates. Consequently, the WMA method is a suitable tool for sales forecasting to support production planning, inventory control, and the formulation of more effective sales strategies.
ANALISIS PERBANDINGAN ALGORITMA DALAM MENEMUKAN POLA PEMBELIAN PRODUK PADA DATA PENJUALAN Antika dewi Asi; Budi Santoso; Nelly Khairani Daulay; Harma Oktafia Oktafia Lingga Wijaya
Jurnal Media Infotama Vol 22 No 1 (2026): April 2026
Publisher : UNIVED Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37676/jmi.v22i1.10951

Abstract

The development of information technology in the Industry 4.0 era has driven significant changes in how organizations utilize data to support strategic decision-making. Data utilization is no longer limited to transaction recording but has shifted toward data processing as a source of predictive information that plays an important role in competitive business management. The urgency of this research is reinforced by the relatively low level of adoption of data-driven analytical systems among Micro, Small, and Medium Enterprises (MSMEs) in Indonesia, including the outdoor equipment rental sector. In the modern business environment, decision-making can no longer rely solely on intuition but must be supported by data and predictive analysis to improve efficiency and competitiveness. Therefore, the development of a Smart Inventory Management system based on Business Intelligence, implementing the Apriori and FP-Growth algorithms at SAVANA Outdoor Store, is expected to provide automatic recommendations for inventory requirements based on real and representative historical transaction patterns. Based on the results of processing outdoor equipment rental data using the Apriori algorithm with a confidence value of 68%, several association rules were obtained, indicating a tendency of dissimilar borrowing patterns (mutually exclusive relationships) among certain types of equipment. Meanwhile, the processing results using the FP-Growth algorithm demonstrated better performance. This algorithm successfully generated a total of 21 association rules, with the top ten rules having confidence values ranging from 70% to 72%.
Two-stage Transfer Learning for Chilli Leaf Disease Classification using CNN and Vision Transformer Rudi Kurniawan; Samsuryadi Samsuryadi; Fatma Susilawati Mohamad; Bunga Intan; Nelly Khairani Daulay
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 10 No 5 (2026): October 2026
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v10i5.7674

Abstract

Chili cultivation is considerably affected by leaf diseases, making timely and reliable disease identification essential for maintaining agricultural productivity. While deep learning has demonstrated strong capabilities in plant disease recognition, many existing studies primarily focus on modifying network architectures rather than optimizing the training process. This study introduces a two-stage transfer learning framework that combines frozen-backbone adaptation with controlled fine-tuning to enhance classification performance. The experiments utilized a publicly available dataset comprising 12,000 augmented chili leaf images representing six disease categories. Several deep learning architectures were investigated, including DenseNet121, EfficientNetB0, MobileNetV2, ResNet152V2, and the Vision Transformer (ViT-B16). The experimental results indicate that the proposed two-stage training strategy improves classification performance consistently across the evaluated architectures. Among the models, ViT-B16 obtained the highest test accuracy of 99.61%, together with stable macro-averaged precision, recall, and F1-score values. Furthermore, the confusion matrix and ROC analysis revealed strong discrimination among the disease classes, with only limited errors occurring between visually similar categories. These findings highlight the importance of optimizing the training schedule when adapting pretrained models for plant disease recognition. Overall, the proposed framework provides an effective and scalable approach for chili leaf disease classification while offering an alternative to approaches that primarily rely on architectural modifications.
MODEL MACHINE LEARNING UNTUK KLASIFIKASI ANOMALI SERANGAN DENIAL OF SERVICE (DOS) PADA INFRASTRUKTUR JARINGAN MENGGUNAKAN DATASET CIC IIOT 2025 Tri Swastiningsih; Rudi Kurniawan; Elmayati Elmayati; Nelly Khairani Daulay
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 5 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/y01jc532

Abstract

Perkembangan Industrial Internet of Things (IIoT) meningkatkan efisiensi sistem industri, namun juga meningkatkan risiko serangan siber, khususnya Denial of Service (DoS) yang dapat mengganggu ketersediaan layanan jaringan. Penelitian ini bertujuan untuk membangun dan mengevaluasi model machine learning dalam mengklasifikasikan anomali serangan DoS pada infrastruktur IIoT menggunakan dataset CIC IIoT 2025. Dataset ini memuat trafik normal (benign) serta berbagai jenis serangan DoS seperti TCP SYN Flood, UDP Flood, ICMP Flood, dan MQTT Publish Flood. Permasalahan ketidakseimbangan data ditangani menggunakan metode Synthetic Minority Over-sampling Technique (SMOTE). Tiga algoritma supervised learning digunakan, yaitu Random Forest, XGBoost, dan Support Vector Machine (SVM). Evaluasi dilakukan menggunakan metrik accuracy, precision, recall, F1-score, confusion matrix, ROC-AUC, serta K-Fold Cross Validation. Hasil penelitian menunjukkan bahwa Random Forest dan XGBoost memperoleh akurasi tertinggi sebesar 97%, sedangkan SVM memperoleh akurasi 95%. Secara keseluruhan, menunjukkan performa paling stabil dalam mendeteksi berbagai variasi serangan DoS. Random Forest Penelitian ini berkontribusi dalam pengembangan sistem deteksi intrusi berbasis machine learning untuk meningkatkan keamanan jaringan IIoT.   Kata Kunci: IIoT, Denial of Service, Machine Learning, SMOTE.