Garuda - Garba Rujukan Digital

Analisis dan Rancangan User Experience Website OAIL Menggunakan Metode Task Centered System Design (TCSD) Yulita, Winda; Algifari, Muhammad Habib; Rinaldi, Daniel; Praseptiawan, Mugi
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 5, No 2 (2021): EDISI SEPTEMBER
Publisher : STIKOM Tunas Bangsa Pematangsiantar

The increase in internet service users in obtaining information and knowledge has made many institutions or organizations begin to create and design websites. One of the institutions that started to build a website is UPT OAIL. In the process of forming the OAIL website, a good User experience (UX) design is needed because it affects user satisfaction in using the website. The analysis and design of user experience in this study uses the Task Centered System Design (TCSD) method. The TCSD method can identify user needs and task needs. The stages in the research are user identification and observation, user and organization requirements analysis, design as scenario and walkthrough evaluate. In this study, identification and observation were carried out by interviewing UPT OAIL staff as well as prospective users. The test is based on the usability method using USE Questionare on the ease of use dimension. The results of the study obtained that the interpretation score for testing the ease of use dimension was 94.2% with the conclusion that the average user or respondent chose the design of each page and the features made were very easy to use.

Pendeteksian Jumlah Bangunan Berbasis Citra Menggunakan Metode Deep Learning Bagaskara, Radhinka; Rizkita, Alya Khairunnisa; Fernandes, Rivaldo; Yulita, Winda
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 6, No 1 (2022): EDISI MARET
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v6i1.428

Counting residential houses is one of the problems faced when determining population density level in Indonesia, therefore it’s required to find a method that’s able to solve said problem. Deep learning method is capable of creating a prediction model for detecting the number of buildings from an image. The deep learning prediction model is created with MobileNetv2 application. The prediction model is trained by using a dataset from Kaggle. The prediction model is tested using satellite photos taken from Way Kandis-Sukarame, Bandar Lampung. The result is a deep learning prediction model with accuracy of 91.30% for SenseFly and 10.34% for Way Kandis dataset. The research can be further improved by using a better training and testing dataset

Automatic Short Answer Assessment Using The Cosine Similarity Method Bahri, Samsu; Praseptiawan, Mugi; Yulita, Winda
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 7, No 1 (2023): EDISI MARET
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v7i1.608

In the learning process, most exams to assess learning achievement have been carried out by providing questions in the form of short answers or essay questions. The variety of answers given by students makes a teacher have to focus on reading them. This process of assigning grades is difficult to guarantee quality if done manually. Moreover, each class is mastered by a different teacher, which can cause inequality in the grades obtained by students due to the influence of differences in teacher experience. Therefore the automated answer assessment research was developed. The automatic short answer assessment is designed to automatically assess and evaluate students' answers based on a trained set of answer documents. The automated grading system uses the cosine similarity method to determine the degree of similarity of a student's answer to the teacher's answer. While the word weighting used is the Term Frequency-Inverse Document Frequency (TF-IDF) method. The data used is a question totaling 5 questions with each question answered by 30 students, while the students' answers are assessed by experts to determine the real value. This study was evaluated by Mean Absolute Error (MAE) with the resulting value of 0.22.

Improved human image density detection with comparison of YOLOv8 depth level architecture and drop-out implementation Yulita, Winda; Ramadhani, Uri Arta; Mufidah, Zunanik; Atmajaya, Gde KM; Bagaskara, Radhinka; Kesuma, Rahman Indra; Aprilianda, Mohamad Meazza
Journal of Soft Computing Exploration Vol. 6 No. 1 (2025): March 2025
Publisher : SHM Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joscex.v6i1.556

Energy inefficiency due to Air Conditioners (AC) running in empty rooms contribute to unnecessary energy consumption and increased CO₂ emissions. This study explores how different depth levels of the YOLOv8 architecture and dropout regularization can enhance human density detection for smarter AC control systems. By evaluating model accuracy through Mean Average Precision (mAP50-95), we provide quantitative insights into how these modifications improve detection performance. Our dataset consists of 1363 images taken in an office environment at ITERA under varying lighting conditions and different human presence densities. The results show that the YOLOv8m model performs best, achieving an mAP50-95 score of 0.814 in training and 0.813 in validation, outperforming other YOLOv8 variants. Furthermore, applying dropout regularization improves model generalization, increasing mAP50-95 from 0.552 to 0.6 and effectively reducing overfitting. This study highlights the balance between architectural depth and dropout regularization in YOLOv8, demonstrating its effectiveness in energy-efficient smart buildings. The findings support the potential of deep learning-based human density detection in improving energy conservation strategies, making it a valuable solution for intelligent automation systems.

Prediksi Penyakit Daun Pisang Menggunakan Metode LSTM (Long Short-Term Memory) Ba’its, Alfian Kafilah; Bagaskara, Radhinka; Setiawan, Andika; Yulita, Winda; Harmiansyah, Harmiansyah; Listiani, Amalia; Untoro, Meida Cahyo; Drantantiyas, Nike Dwi Grevika; Faisal, Amir; Anggraini, Leslie; Febrianto, Andre; Aprilianda, Mohamad Meazza; Fitrawan, Mhd. Kadar
Jurnal Teknik Informatika UNIKA Santo Thomas Vol 10 No. 1 : Tahun 2025
Publisher : LPPM UNIKA Santo Thomas

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

Dalam sektor pertanian, tanaman yang memiliki peran signifikan dalam skala global adalah pisang, yaitu buah yang mudah didapatkan, dapat tumbuh dimana saja, memiliki gizi yang tinggi, serta memiliki nilai ekonomi & budaya yang tinggi. Pisang mempunyai kontribusi yang signifikan terhadap pendapatan nasional Indonesia, terutama di Provinsi Lampung sebagai penghasil pisang nasional terbesar. Tetapi, proses produksi pisang seringkali mengalami kendala, salah satunya karena faktor serangan penyakit Black Sigatoka. Penyakit tersebut memberikan kerugian pada tanaman pisang, seperti daun yang meranggas, panen tertunda, bakal buah rontok, dan kualitas buah yang rendah, dan dapat menyebar melalui aliran udara atau percikan air hujan. Tingkat keparahan penyakit Black Sigatoka perlu diprediksi agar penyakit tersebut dapat dikontrol dan dapat dicegah sedini mungkin. Model yang digunakan untuk memprediksi permasalahan ini dalam jangka panjang adalah model Long Short-Term Memory (LSTM), salah satu jenis dari arsitektur Recurrent Neural Network (RNN), yang mempunyai kinerja yang baik dan mempunyai model yang prediktif. Aplikasi LSTM diterapkan terhadap dataset pohon pisang yang terdampak penyakit Black Sigatoka. Hasil dari model LSTM dalam melakukan prediksi penyakit Black Sigatoka menghasilkan model dengan nilai error yang kecil, dengan nilai MAE dan MAPE masing-masing sebesar 0.084 dan 5.7%

Analisis Hubungan dan Prediksi Depresi Mahasiswa Berdasarkan Faktor Akademik dan Gender Verdiana, Miranti; Dwi Nugroho, Eko; Anggraini, Leslie; Bagaskara, Radhinka; Yulita, Winda; Afriansyah, Aidil; Habib Algifari, Muhammad
Jurnal Teknik Informatika UNIKA Santo Thomas Vol 10 No. 1 : Tahun 2025
Publisher : LPPM UNIKA Santo Thomas

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

This study aims to analyze the level of depression among university students by examining gender and several academic indicators. The dataset includes responses from 27,901 students across various regions, with variables covering age, gender, academic pressure, study satisfaction, work/study hours, CGPA, and depression status. The analytical methods applied in this study include the chi-square test to eval_uate the association between gender and depression status, point-biserial correlation to examine relationships between numeric variables and depression, and logistic regression to develop a prediction model. The chi-square test results revealed no significant relationship between gender and depression (p = 0.774), indicating that depression affects both genders. In contrast, academic pressure exhibited the strongest correlation with depression status (r = 0.47), followed by work/study hours (r = 0.209) and study satisfaction (r = -0.168). The Logistic Regression model constructed using the four most relevant variables demonstrated satisfactory performance, achieving 75.5% accuracy and 82.1% recall in identifying students experiencing depression. These findings highlight the critical role of academic-related factors—particularly academic pressure—in influencing students’ mental health. Therefore, targeted academic support strategies are essential to mitigate depression risks in higher education environments. Keywords— Student Depression, Academic Pressure, Gender, Logistic Regression, Mental Health Prediction

Prediksi Penyakit Daun Pisang Menggunakan Metode LSTM (Long Short-Term Memory) Ba’its, Alfian Kafilah; Bagaskara, Radhinka; Setiawan, Andika; Yulita, Winda; Harmiansyah, Harmiansyah; Listiani, Amalia; Untoro, Meida Cahyo; Drantantiyas, Nike Dwi Grevika; Faisal, Amir; Anggraini, Leslie; Febrianto, Andre; Aprilianda, Mohamad Meazza; Fitrawan, Mhd. Kadar
Jurnal Teknik Informatika UNIKA Santo Thomas Vol 10 No. 1 : Tahun 2025
Publisher : LPPM UNIKA Santo Thomas

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

Dalam sektor pertanian, tanaman yang memiliki peran signifikan dalam skala global adalah pisang, yaitu buah yang mudah didapatkan, dapat tumbuh dimana saja, memiliki gizi yang tinggi, serta memiliki nilai ekonomi & budaya yang tinggi. Pisang mempunyai kontribusi yang signifikan terhadap pendapatan nasional Indonesia, terutama di Provinsi Lampung sebagai penghasil pisang nasional terbesar. Tetapi, proses produksi pisang seringkali mengalami kendala, salah satunya karena faktor serangan penyakit Black Sigatoka. Penyakit tersebut memberikan kerugian pada tanaman pisang, seperti daun yang meranggas, panen tertunda, bakal buah rontok, dan kualitas buah yang rendah, dan dapat menyebar melalui aliran udara atau percikan air hujan. Tingkat keparahan penyakit Black Sigatoka perlu diprediksi agar penyakit tersebut dapat dikontrol dan dapat dicegah sedini mungkin. Model yang digunakan untuk memprediksi permasalahan ini dalam jangka panjang adalah model Long Short-Term Memory (LSTM), salah satu jenis dari arsitektur Recurrent Neural Network (RNN), yang mempunyai kinerja yang baik dan mempunyai model yang prediktif. Aplikasi LSTM diterapkan terhadap dataset pohon pisang yang terdampak penyakit Black Sigatoka. Hasil dari model LSTM dalam melakukan prediksi penyakit Black Sigatoka menghasilkan model dengan nilai error yang kecil, dengan nilai MAE dan MAPE masing-masing sebesar 0.084 dan 5.7%

Analisis Hubungan dan Prediksi Depresi Mahasiswa Berdasarkan Faktor Akademik dan Gender Verdiana, Miranti; Dwi Nugroho, Eko; Anggraini, Leslie; Bagaskara, Radhinka; Yulita, Winda; Afriansyah, Aidil; Habib Algifari, Muhammad
Jurnal Teknik Informatika UNIKA Santo Thomas Vol 10 No. 1 : Tahun 2025
Publisher : LPPM UNIKA Santo Thomas

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

This study aims to analyze the level of depression among university students by examining gender and several academic indicators. The dataset includes responses from 27,901 students across various regions, with variables covering age, gender, academic pressure, study satisfaction, work/study hours, CGPA, and depression status. The analytical methods applied in this study include the chi-square test to eval_uate the association between gender and depression status, point-biserial correlation to examine relationships between numeric variables and depression, and logistic regression to develop a prediction model. The chi-square test results revealed no significant relationship between gender and depression (p = 0.774), indicating that depression affects both genders. In contrast, academic pressure exhibited the strongest correlation with depression status (r = 0.47), followed by work/study hours (r = 0.209) and study satisfaction (r = -0.168). The Logistic Regression model constructed using the four most relevant variables demonstrated satisfactory performance, achieving 75.5% accuracy and 82.1% recall in identifying students experiencing depression. These findings highlight the critical role of academic-related factors—particularly academic pressure—in influencing students’ mental health. Therefore, targeted academic support strategies are essential to mitigate depression risks in higher education environments. Keywords— Student Depression, Academic Pressure, Gender, Logistic Regression, Mental Health Prediction

Ideal Temperature Classification of Meeting Rooms Using You Only Look Once Architecture Version 8 and Multilayer Perceptron Based on Human Density Image Data Ridwan, Naufal Taufiq; Yulita, Winda; Kesuma, Rahman Indra; Ramadhani, Uri Arta; Bagaskara, Radhinka
Indonesian Journal of Artificial Intelligence and Data Mining Vol 8, No 2 (2025): July 2025
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v8i2.34230

Indonesia, located along the equator, experiences a tropical climate that results in high indoor temperatures. Elevated temperatures can affect health, making air conditioning (AC) necessary to regulate indoor environments. However, improper use of AC systems, such as leaving them on even when a room is unoccupied, can lead to significant energy waste. This research focuses on the efficient use of AC systems through the integration of sensors and cameras, combining two distinct technologies. The first technology is object detection using You Only Look Once (YOLOv8), which was chosen for its superior performance in terms of speed, accuracy, and computational efficiency. The second is the classification of optimal AC temperatures using the Multilayer Perceptron (MLP) algorithm, selected for its high performance in accuracy, sensitivity, and speed. In addition, the study takes into account human density in the room to optimize temperature regulation. The integration of object detection and temperature classification technologies enables the system to operate in real time and automatically adjust temperature settings based on dynamic room conditions. The research successfully implemented YOLOv8 for object detection and Multilayer Perceptron for optimal room temperature classification. Test results showed precision, recall, and F1-score values of 0.82, 0.92, and 0.86, respectively.

Sentiment Analysis of the Minister of Education and Culture using Vader and RBF, Polynomial, Linier Kernels SVM Based on Binary Particle Swarm Optimization Sinaga, Rutlima; Ashari, Ilham Firman; Yulita, Winda
Sistemasi: Jurnal Sistem Informasi Vol 13, No 3 (2024): Sistemasi: Jurnal Sistem Informasi
Publisher : Program Studi Sistem Informasi Fakultas Teknik dan Ilmu Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v13i3.2186

Comments from social media can be analyzed further. Social media is used to interact from one person to another, as well as with the government. This Issue Was Raised Because Of Debate And Public Opinion From The Community, Institutions And Ngos Regarding Ministerial Regulation No. 30 Of 2021 Concerning Prevention And Handling Of Sexual Violence. In The Higher Education Environment, Therefore In This Research We Want To Examine What Is The Main Root Of The Problem Using A Methodical Approach Using Natural Language Processing. The pre-processing applied is case folding, tokenization, elimination of stop words, stemming using literature. The model implementing PSO failed to improve accuracy on all kernels. Best performance before applying PSO to twitter dataset using linear kernel. This study conducted sentiment analysis regarding the issuance of ministerial regulation no. 30 of 2021. The data obtained was then preprocessed. The performance measured is accuracy and f1-macro in the model without PSO and accuracy in the model using accuracy. The model to be formed uses linear kernels, RBF and polynomials of order 1 and order 2. Sentence analysis is a field that analyzes sentiment, attitudes and emotions of entities and their attributes in text form. The aim of this research is to compare the performance of the Support Vector Machine classification algorithm without Particle Swarm Optimization feature selection and the performance of the Support Vector Machine classification algorithm using Particle Swarm Optimization feature selection. The data obtained is then pre-processed. The data set was automatically labeled using VADER (Valence Dictionary for Sentiment Reasoning). The kernels that succeeded in increasing accuracy were the RBF kernel and polynomials on the Twitter dataset. Keywords: SVM, Vader, PSO, Sentiment Analysis, Government Policy

Title

Found 14 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 14 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 14 Documents
Search