Garuda - Garba Rujukan Digital

Sentiment Analysis Of The Shopee Marketplace On Twitter Using The Naive Bayes Classifier Method Natalia, Nila; Astikarani, Ester Krisdianti; Adi Khairul, Muhammad; Nafis Sjamsuddin, Irfan
Journal of Applied Information System and Informatic (JAISI) Vol 3, No 2 (2025): November 2025
Publisher : Deparment Information System, Siliwangi University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/jaisi.v3i2.17077

Shopee is the number one most downloaded marketplace application on the App Store and Play Store. In its promotion, Shopee provides discounts on shipping costs, price discounts, and cashback for each transaction; however, not all of its users are satisfied with the service. There are criticisms and suggestions, one of which is conveyed via social media, Twitter. Sentiment analysis was conducted to extract information related to Shopee user reviews on Twitter. The stages of the research carried out followed the Cross Industry Standard Process for Data mining (CRISP-DM) method, namely Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. Data collection was carried out by scraping, and all classification processes were carried out using the RapidMiner tool. The data obtained tends to contain negative sentiment rather than positive; most reviews are made by buyers and discuss promos. Sentiment classification is carried out by applying the Naive Bayes Classifier and TF-IDF as feature extraction. Testing using 10-fold cross-validation and a Confusion Matrix resulted in an accuracy value of 84.20%, a precision value of 87.21%, and a recall value of 84.20%.

SISTEM INFORMASI PENGADUAN KAMPUS MENGGUNAKAN METODE ANALYTICAL HIERARCHY PROCESS (AHP) BERBASIS WEB (STUDI KASUS POLITEKNIK SUKABUMI) Waruwu, Nover Lindra; Natalia, Nila
SEMNASTERA (Seminar Nasional Teknologi dan Riset Terapan) Vol 7 (2025)
Publisher : SEMNASTERA (Seminar Nasional Teknologi dan Riset Terapan)

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

Sistem informasi pengaduan kampus adalah sistem yang memberikan infromasi kepada pihak kampus dalam menangani pengaduan dari mahasiswa namun, di Politeknik Sukabumi proses pengaduan masih dilakukan secara manual, yang mengakibatkan mahasiswa sering mengalami kendala dalam mengungkapkan keluhan dan aspirasi mereka. keadaan ini menimbulkan keterlambatan dalam penanganan. dalam menyelesaikan permasalahan maka tugas akhir ini bertujuan untuk merancang dan menciptakan sistem informasi pengaduan kampus yang berbasis web dengan menggunakan metode Analytical Hierarchy Process (AHP) sebagai pendukung keputusan. aplikasi ini juga dirancang menggunakan framework laravel, dengan MySQL sebagai basis data. metode AHP diterapkan untuk mengidentifikasi prioritas dalam menentukan pengaduan dengan mempertimbangkan beberapa kriteria, termasuk urgensi , dampak , waktu penyelesaian. hasil pengujian black box testing menunjukkan semua skenario fungsional pada sistem telah berjalan valid dan sesuai harapan. Sementara itu, implementasi metode AHP menunjukkan tingkat akurasi perhitungan yang sangat tinggi, berkisar antara 95,93% hingga 99,94% jika dibandingkan dengan perhitungan manual. sistem yang dihasilkan berhasil menyediakan fitur penting seperti notifikasi real-time dan pelacakan status.

Cloud Computing Integration in the Digital Transformation of MSMEs: A Case Study on the Retail Sector Nila Natalia; Errysa Subekthi; Febri Dolis Herdiani; Foezi Arisandi SJ
Technologia Journal Vol. 2 No. 4 (2025): Technologia Journal-November
Publisher : Pt. Anagata Sembagi Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62872/jkbsjt34

This study aims to analyze the role of cloud computing integration in supporting the digital transformation process of Micro, Small, and Medium Enterprises (MSMEs) in the retail sector in Indonesia. Digital transformation has become a strategic necessity in facing the dynamics of global competition and changes in technology-based consumer behavior. This study uses a qualitative method with a case study approach on several retail MSMEs that have adopted cloud computing services. Data were obtained through in-depth interviews, field observations, and documentation, then analyzed descriptively through the stages of data reduction, data presentation, and conclusion drawing. The results show that the implementation of cloud computing can improve operational efficiency, strengthen data management, and accelerate the decision-making process based on real-time information. MSMEs that adopt this technology experience a reduction in IT infrastructure costs of up to 40% and an increase in employee productivity of up to 30%. However, challenges remain such as limited digital literacy, uneven internet infrastructure, and concerns about data security. This study confirms that cloud computing functions not only as a technological solution, but also as a catalyst for changing work culture and business management systems towards sustainable digitalization.

Implementation of DBSCAN Clustering and Random Forest Algorithm for Mapping and Predicting Shooting Incidents in New York Rangkuti, Azka Niaji; Arifin, Samoedra Cakra; Putra, Muhammad Ramadansyah Kurnia; Natalia, Nila
Journal Collabits Vol 3, No 1 (2026)
Publisher : Journal Collabits

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/collabits.v3i1.37587

Shooting incidents in crowded, heavily populated areas of cities cause serious threats to public safety and social security. New York State, which includes large metropolitan areas and suburban regions, experiences complex spatial and temporal crime patterns that are difficult to identify using traditional crime analysis methods that rely only on descriptive statistics and manual hot spot identification. This study proposes a data-driven quantitative approach to mapping and predicting shooting incidents by integrating spatial clustering and machine learning techniques. Density-based clustering methods are applied to the geographic coordinates of shooting incidents to identify areas with high incident concentrations while filtering out isolated events as noise. The resulting spatial clusters are then interpreted as hotspot locations and used as reference labels for a supervised classification model. A Random Forest algorithm is then used to predict hotspot and non-hotspot locations using spatial and temporal features, including geographic position and time of occurrence. The model is evaluated using standard classification performance measures, including accuracy, precision, recall, F1 score, and confusion matrix analysis.

Development of a Sustainable Fashion Mobile Application Prototype Using Rule-Based Filtering Approach Nila Natalia; Zahra Ramadhani Putri; Irfan Nafis Sjamsuddin
Journal of Applied Information System and Informatic (JAISI) Vol 4, No 1 (2026): MEI 2026
Publisher : Deparment Information System, Siliwangi University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/jaisi.v4i1.18276

Abstract— Rapid growth in the fast fashion industry has increased consumer behavior, especially among college students. This growth led to a surge in textile waste. This study aims to design and develop a mobile application prototype named "ReHue" as a digital solution. The app promotes sustainable fashion through a smart wardrobe management and donation system. This study uses a quantitative, applied approach. It focuses on UI/UX design in Figma and prototyping with React Native. Materials analyzed include literature on the circular economy, rule-based filtering algorithms, and textile waste management. The proposed system applies a rule-based filtering approach. It gives personalized clothing recommendations based on real-time weather and the user's color profile. The system uses a threshold-based tracking mechanism. It triggers notifications when clothing has been unworn for more than 210 days, encouraging users to donate through a gamified feature. Experimental results from logic scenario testing show the system can execute recommendation rules and trigger alerts as intended. By offering monetary rewards in the donation flow, the system incentivizes users to reduce textile waste. This research suggests ReHue can help transform impulse-buying habits into ecological actions. Future research should integrate real-time API connectivity and advanced machine learning models to improve recommendations and user engagement across wider reverse logistics networks. Keywords— Sustainable fashion; Smart wardrobe; Rule-based filtering; Mobile application; Textile waste

Designing User Interface and User Experience for Habit Tracker Application for Android Mobile Devices Nila Natalia
Journal Collabits Vol. 2 No. 2 (2025)
Publisher : Journal Collabits

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/collabits.v2i2.35091

Positive habit formation is one of the main challenges in everyday life that requires consistency and continuous motivation. This study aims to design and develop an Android-based habit tracker application with a focus on optimizing the User Interface (UI) and User Experience (UX) to increase user engagement and effectiveness in building positive habits. The research methodology uses the User-Centered Design (UCD) approach which includes the stages of user needs analysis, wireframe and mockup design, prototyping, and usability testing evaluation. Data collection was carried out through a survey of 150 respondents to identify user needs and preferences for the habit tracker application. The UI/UX design process integrates the principles of Material Design Guidelines and the psychology of habit formation theory to create an intuitive and motivating interface. The main features of the application include habit tracking with progress visualization, reward and gamification systems, reminder notifications, and analytics dashboards. The results of the usability testing evaluation showed a user satisfaction level of 87% with a System Usability Scale (SUS) score of 82.5, which is included in the "excellent" category. A/B testing on various UI elements showed an increase in user retention rate of 34% and completion rate of 28% compared to conventional design. The application successfully implemented a consistent responsive design across different screen sizes with an average loading time of 2.3 seconds. This research contributes to the development of user-friendly mobile applications for habit tracking, provides insights into the importance of a psychological approach in UI/UX design, and serves as a reference for the development of similar applications in the future.

Implementation of DBSCAN Clustering and Random Forest Algorithm for Mapping and Predicting Shooting Incidents in New York Azka Niaji Rangkuti; Samoedra Cakra Arifin; Muhammad Ramadansyah Kurnia Putra; Nila Natalia
Journal Collabits Vol. 3 No. 1 (2026)
Publisher : Journal Collabits

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/collabits.v3i1.37587

Shooting incidents in crowded, heavily populated areas of cities cause serious threats to public safety and social security. New York State, which includes large metropolitan areas and suburban regions, experiences complex spatial and temporal crime patterns that are difficult to identify using traditional crime analysis methods that rely only on descriptive statistics and manual hot spot identification. This study proposes a data-driven quantitative approach to mapping and predicting shooting incidents by integrating spatial clustering and machine learning techniques. Density-based clustering methods are applied to the geographic coordinates of shooting incidents to identify areas with high incident concentrations while filtering out isolated events as noise. The resulting spatial clusters are then interpreted as hotspot locations and used as reference labels for a supervised classification model. A Random Forest algorithm is then used to predict hotspot and non-hotspot locations using spatial and temporal features, including geographic position and time of occurrence. The model is evaluated using standard classification performance measures, including accuracy, precision, recall, F1 score, and confusion matrix analysis.

Title

Found 27 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 27 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 27 Documents
Search