Portal Jurnal Elektronik Universitas Kristen Satya Wacana (UKSW)
Not a member yet
2881 research outputs found
Sort by
Pengembangan Model Klasifikasi Kualitas Sarang Burung Walet Berbasis CNN dengan Transfer Learning MobileNetV2
No full textThe quality of edible bird’s nests is a crucial factor in determining their market value, thus requiring an accurate and automated classification system. This study aims to develop a quality classification model for edible bird’s nests using a Convolutional Neural Network (CNN) algorithm with a transfer learning approach based on the MobileNetV2 architecture. The dataset consists of 3,406 bird’s nest images collected directly from farmers, which were processed through background removal and aggressive augmentation to highlight the main object. The data were evenly split into 2,723 training images and 683 validation images, covering three quality classes: high, medium, and low. The model was trained in two stages: initial training with a frozen base and subsequent fine-tuning. Evaluation results showed an improvement in accuracy from 93% to 97% after fine-tuning, with average precision, recall, and F1-score values of 0.97. The confusion matrix indicated high classification accuracy. This study contributes to the development of an image-based classification model with high accuracy, offering potential for efficient and objective industrial sorting.Kualitas sarang burung walet merupakan faktor penting dalam menentukan nilai jualnya, sehingga dibutuhkan sistem klasifikasi yang akurat dan otomatis. Penelitian ini bertujuan untuk mengembangkan model klasifikasi mutu sarang burung walet menggunakan algoritma Convolutional Neural Network (CNN) dengan pendekatan transfer learning berbasis arsitektur MobileNetV2. Dataset terdiri dari 3.406 citra sarang burung walet yang diperoleh langsung dari petani, kemudian diproses melalui penghapusan latar belakang dan augmentasi agresif untuk menonjolkan objek utama. Data dibagi secara seimbang menjadi 2.723 citra pelatihan dan 683 citra validasi, mencakup tiga kelas: kualitas tinggi, sedang, dan rendah. Model dilatih dalam dua tahap, yaitu pelatihan awal frozen base dan fine-tuning. Hasil evaluasi menunjukkan peningkatan akurasi dari 93% menjadi 97% setelah fine-tuning, dengan nilai precision, recall, dan F1-score rata-rata sebesar 0,97. Confusion matrix memperlihatkan akurasi klasifikasi yang tinggi. Penelitian ini berkontribusi dalam pengembangan model klasifikasi mutu sarang burung walet berbasis citra, yang akurat dan berpotensi mendukung penyortiran industri secara efisien
Pendidikan Menurut Paulo Freire Sebagai Kerangka Analisis Darurat Literasi di NTT
Full text linkPenelitian ini mau menganalisis konteks yang terjadi di Indonesia, yaitu Darurat Literasi di NTT. Ini menggambarkan bagaimana situasi pendidikan dan kebijakan pemerintah tidak berjalan dengan baik dan tepat sasaran dan kompetensi pendidik yang tidak memadai. Dari sinilah, penelitian ini dibuat untuk dianalisis dengan menggunakan teori Pendidikan menurut Freire, yang menekankan pendidikan pembebasan, dialog, dan pendidikan gaya bank untuk membaca apa yang terjadi dan ini sebagai rujukan untuk membantu mengevaluasi, demi tercapainya pendidikan di Indonesia yang maju, dan berdaya juang. Metode yang dipakai ialah menggunakan penelitian jenis kualitatif dengan pendekatan studi literatur atau pustaka yang berfokus pada dokumen, artikel, buku, dan jurnal. Untuk data penelitian sendiri menggunakan sumber artikel terpercaya. Dari sinilah, bisa disimpulkan bahwa Imperatif moral adalah pemerintah dan masyarakat, serta melalui pembacaan teori Freire, sebagai responsif terkait adanya ketimpangan serta ketidaksesuaian di NTT dalam memberikan pengajaran dan implementasi kebijakan.This study attempts to analyze the context of the Literacy Emergency in NTT in Indonesia. This illustrates how the educational situation and system are not functioning well and on target, as well as the inadequate competence of educators. A large amount of data was collected and can be seen in Kompas, or regional media such as Suara Kupang. From here, this study was conducted to be analyzed using Freire's Educational Theory, which emphasizes liberation education, dialogue, and banking-style education to read what is happening and this as a reference to help evaluate, in order to achieve an advanced and fighting education in Indonesia. This study uses a qualitative approach with a type of literature study or library that focuses on analyzing articles, especially on news websites that describe the Literacy Emergency in NTT. From here, through the conclusions generated from reading and analyzing Freire's theory, as a response to the inequality and inconsistency in access to education in NTT, it is hoped that in the future the Literacy Emergency can be reduced
Pengembangan Media Pembelajaran Puzzle Matematika “PUMA” Untuk Meningkatkan Hasil Belajar Siswa di Sekolah Dasar
Full text linkPecahan Senilai pada pembelajaran Sekolah Dasar perlu disertai dengan menggunakan alat penyaji yang menarik dan tidak monoton terutama pada latihan soal, hal tersebut yang menjadi landasan penelitian untuk mengembangkan media pembelajaran board game yang dapat digunakan pada materi Pecahan Senilai. Penelitian ini bertujuan untuk mengembangkan media yang valid, praktis, dan efektif sebagai sarana latihan soal pada materi Pecahan Senilai. Penelitian ini menggunakan model pengembangan ADDIE yang terdiri dari tahapan Analyze, Design, Development, Implementation, dan Evaluation. Media ini merupakan modifikasi dan pengembangan dari permainan puzzle. Media ini diberi nama PUMA “Puzzle Matematika”, dan sudah diujicobakan terhadap 37 siswa kelas 5 SD Kanisius Gendongan Salatiga. Instrumen pengumpulan data dalam penelitian ini berupa instrumen validasi aspek media dan aspek materi, instrumen kepraktisan, instrumen pretest dan posttest, dan instrument pendapat siswa. Media permainan board game puzzle dinyatakan valid dari aspek ahli media dengan skor 90,88% dan valid dari aspek materi dengan skor 93,5%, keduanya masuk dalam kategori sangat baik. Hasil dari kepraktisan memperoleh skor 94,59% masuk dalam kategori sangat baik. Hasil tersebut menunjukkan bahwa media pembelajaran puzzle board game PUMA merupakan media yang valid, praktis, dan efektif untuk mengajarkan materi Pecahan Senilai di sekolah dasar.Learning about Equivalent Fractions in elementary schools needs to be supported by engaging and varied instructional tools, particularly during practice activities. This consideration forms the basis of the present study, which aims to develop a board game–based learning medium for teaching Equivalent Fractions. The study seeks to produce a medium that is valid, practical, and effective as a tool for student practice. The research employed the ADDIE development model, consisting of the stages: Analysis, Design, Development, Implementation, and Evaluation. The developed medium, named PUMA (Puzzle Matematika), is a modification and enhancement of a puzzle game, and was tested on 37 fifth-grade students at SD Kanisius Gendongan, Salatiga. Data were collected using instruments for media and content validation, practicality assessment, pretest and posttest evaluations, and student feedback. The board game puzzle medium was rated valid by media experts with a score of 90.88% and by content experts with a score of 93.5%, both categorized as very good. The practicality assessment obtained a score of 94.59%, also categorized as very good. These results indicate that the PUMA board game puzzle is a valid, practical, and effective learning medium for teaching Equivalent Fractions in elementary schools. Based on the research findings, the “PUMA” puzzle media is appropriate for use as a learning medium for equivalent fractions at the elementary school level, as it is effective as a practice tool and supports active student engagement in the learning process
Hubungan Hubungan Antara Faktor Internal dan Eksternal dengan Efikasi Diri PEserta Didik Sd dalam Pengembangan P5
No full textThis study aims to examine the simultaneous relationship between internal and external factors and the self-efficacy of elementary school students in developing the Pancasila Student Profile Strengthening Project (P5). This study uses a correlational research design with a quantitative approach. The research subjects consisted of 125 fourth-grade students from five public elementary schools in Malang City, selected using cluster random sampling. Data collection techniques were conducted through interviews and the completion of validated questionnaires. Data analysis was performed using prerequisite tests and hypothesis testing through multiple linear regression analysis with the help of IBM SPSS for Windows software. The research findings show that internal and external factors together have a significant relationship with student self-efficacy, as indicated by a significance value (<0.001) < 0.05. The Adjusted R Square value of 0.257 indicates that the simultaneous contribution of internal and external factors to self-efficacy is 25.7%. These results confirm that the synergy between individual readiness and environmental support is very important in building students' confidence to complete the P5 project optimally.Studi korelasi ini bertujuan untuk mengetahui bagaimana variabel eksternal (dukungan teman sebaya, dukungan guru, dan lingkungan kelas) dan variabel internal (konsep diri akademik, motivasi belajar, dan pengalaman) mempengaruhi rasa efikasi diri siswa sekolah dasar dalam pengembangan keterampilan P5 mereka. Temuan survei terhadap siswa kelas empat dievaluasi menggunakan berbagai teknik regresi linier. Temuan penelitian ini menunjukkan hubungan yang kuat dan positif antara efikasi diri dan faktor internal maupun eksternal. Nilai R kuadrat yang disesuaikan dari pemeriksaan regresi linier berganda adalah 0,257, sedangkan nilai sig. berada di bawah 0,001, keduanya di bawah 0,05. Menurut temuan ini, efikasi diri siswa dipengaruhi oleh keterlibatan mereka dengan lingkungan kelas, bukan hanya oleh bakat bawaan atau motivasi intrinsik mereka saja
Segmentasi pelanggan pada penjualan perusahaan air minum XYZ dengan analisis RFM dan K-Means clustering
No full textA suitable and targeted marketing plan is required because of the intense competition in the retail drinking water sector. Customer segmentation using RFM (Recency, Frequency, and Monetary) analysis is one of the techniques employed. Additionally, K-Means clustering, a clustering technique based on machine learning, is employed. This study's goal is to present the findings in the form of graphs that can be used to examine consumer trends according to their attributes. With a value of 10286, the Calinski Harabaz index is a suitable metric to move on to the segmentation step in this study, which also tests three metrics using the clustering method. An ideal cluster is created for every cluster evaluation by dividing the Calinski Harabaz index into three more manageable clusters. This contrasts with other evaluation metrics that only yield two clusters. For instance, when XYZ drinking water sales transaction data was distributed, it was discovered that, out of the total drinking water sales, woodsale had 422 customers, diamond had 1061 customers, and star diamond had 2005 customers. The management of the XYZ drinking water company and other marketing fields are expected to encounter more intense competition as a result of the study's findings
Pengembangan sistem pendukung keputusan berbasis AI untuk evaluasi ide konten digital pra-produksi
No full textThe rapid growth of short-form social media platforms has increased the complexity of decision-making during the digital content planning stage. Content creators are required to evaluate the feasibility of content ideas and determine platform suitability prior to production, while most existing tools primarily focus on post-publication analytics. This study aims to design an Artificial Intelligence (AI)-enabled Decision Support System (DSS) to evaluate digital content ideas in the pre-production stage. Adopting a Design Science Research approach, the study develops a conceptual design artifact that integrates intrinsic content idea characteristics with cognitive and affective response modeling grounded in the Stimulus–Organism–Response (S-O-R) framework, alongside platform affordance mapping. The proposed artifact operationalizes a reflective evaluation mechanism that generates platform recommendation scores and idea enhancement suggestions without claiming deterministic or predictive performance modeling. Evaluation was conducted qualitatively through practitioner assessment to examine perceived usefulness, clarity of recommendations, and decision support contribution. The findings indicate that the developed artifact provides a structured reflective framework for early-stage content evaluation. Theoretically, this study extends the application of the S-O-R framework by operationalizing it as a design logic for a pre-production DSS artifact. Practically, the proposed system has the potential to support more systematic decision-making prior to content production
Sistem pendukung keputusan pemilihan lulusan terbaik untuk jurusan teknik informatik di Instituto Profissional de Canossa (IPDC) menggunakan metode multi attribute utility theory
No full textBest graduate selection is crucial for academic achievement and contributes to the accreditation value of the institution. Instituto Profissional de Canossa (IPDC) is a higher education institution founded by the Canossian Sisters in Timor-Leste. To improve the effectiveness of assessment and decision-making processes, an information system is needed to assist in selecting the best graduates based on multiple criteria. This research develops a web-based Decision Support System (DSS) using the Multi-Attribute Utility Theory (MAUT) method. MAUT is a multi-criteria decision-making method that evaluates alternatives based on their utility scores across several criteria. The study uses four main criteria: attendance, academic performance, ethics, and discipline. The system is implemented as a web application for universal access. The MAUT calculation results provide valid and accurate recommendations for the best graduates. System testing showed that the application successfully ranked candidates based on defined weights and criteria, providing objective and consistent selection results
Meningkatkan Student Engagement dan Hasil Belajar Menggunakan Media Pembelajaran Pemrograman Berbasis Game
Full text linkComputer programming is an essential skill in the digital era, yet many students struggle due to teacher-centered lecture methods that rely heavily on one-way explanations and passive notetaking. This study aims to develop a game-based programming learning media to enhance student engagement and learning outcomes. Using Design-based Research (DBR) with the ADDIE model, a web-based media integrating game elements such as live coding, leaderboards, and level-based challenges was designed and tested. Participants included 276 students from Wijaya Kusuma Surabaya University, selected via purposive sampling. Data were collected through knowledge tests, engagement questionnaires, and user satisfaction surveys, then analyzed using descriptive statistics and paired sample t-tests. Results indicated significant improvements in student engagement (89% of students achieved high engagement) and learning outcomes (average score increase of 19.94 points). These findings demonstrate the effectiveness of game-based media in creating interactive and enjoyable learning experiences. The study contributes to the development of innovative educational methods in technology-based learning.Pemrograman komputer merupakan keterampilan esensial di era digital, namun banyak mahasiswa mengalami kesulitan akibat metode pembelajaran yang monoton. Penelitian ini bertujuan mengembangkan media pembelajaran pemrograman berbasis game untuk meningkatkan student engagement dan hasil belajar. Metode Design-based Research (DBR) dengan pendekatan ADDIE digunakan untuk merancang, mengembangkan, dan menguji media berbasis web yang mengintegrasikan elemen permainan seperti live coding, leaderboard, dan tantangan berbasis level. Partisipan terdiri dari 276 mahasiswa Universitas Wijaya Kusuma Surabaya yang dipilih melalui purposive sampling. Data dikumpulkan melalui tes pengetahuan, kuesioner student engagement, dan kuesioner kepuasan pengguna, kemudian dianalisis secara statistik deskriptif dan paired sample t-test. Hasil menunjukkan peningkatan signifikan student engagement (89% mahasiswa mencapai kategori tinggi) dan hasil belajar (rata-rata kenaikan skor 19,94 poin). Temuan ini membuktikan efektivitas media berbasis game dalam menciptakan pembelajaran yang interaktif dan menyenangkan. Penelitian ini berkontribusi pada pengembangan metode pembelajaran inovatif di bidang pendidikan teknologi
Pertumbuhan Tanaman Duku Belum Menghasilkan terhadap Aplikasi Pupuk Kompos Kotoran Kambing dan Pupuk NPK
No full textDuku production and productivity in Jambi Province remain low because the existing plants are old, and cultivation technology is not precise. Efforts to overcome this problem include rejuvenation using superior seeds obtained through propagation of shoot grafting. Intensification can be done through fertilization. One fertilizer option is goat manure compost combined with NPK. The purpose of this research is to develop a technology package to accelerate duku growth in ultisols. This research was conducted in the Faculty of Agriculture’s experimental garden for 4 months on 1-year-old duku plants. The design used in this study was a randomized block design (RAK) with one factor, namely the dose of goat compost fertilizer consisting of five levels, namely 100% NPK (306g); 5 kg Goat Compost + 75% NPK (229.5g); 10 kg Goat Compost + 50% NPK Fertilizer (153g); 15 kg Goat Compost + 25% NPK (76.5g); 20 kg Goat Compost Fertilizer. Observations were made on the variables of increase in plant height, increase in stem diameter, increase in the number of leaf tillers, bud break time, and increase in the number of branches. The results showed that goat compost combined with NPK can substitute for NPK fertilizer. The application of 15 kg of goat manure compost with 25% NPK content showed the best combination for supporting the growth of immature Duku plants.Produksi dan Produktivitas tanaman duku di Propinsi Jambi masih rendah, hal ini disebabkan tanaman yang ada sudah tua, dan kurang tepatnya penerapan teknologi budidaya. Upaya untuk mengatasi masalah ini, dengan peremajaan dengan menggunakan bibit unggul yang diperoleh melalui perbanyakan sambung pucuk. Intensifikasi dapat dilakukan melalui pemupukan. Salah satu pupuk yang dapat digunakan adalah pupuk kompos kotoran kambing yang dikombinasi dengan NPK. Tujuan penelitian ini adalah diperolehnya suatu paket teknologi yang dapat mempercapat pertumbuhan tanaman duku dilahan ultisol. Penelitian ini dilakukan di kebun percobaan Fakultas Pertanian selama 4 bulan pada tanaman duku yang sudah berumur 1 tahun. Rancangan yang digunakan dalam penelitian ini adalah Rancangan acak Kelompok (RAK) dengan 1 faktor yaitu takaran Pupuk Kompos Kotoran kambing yang terdiri dari 5 taraf yaitu NPK 100% ( 306g ); 5 kg Pupuk Kompos Kotoran Kambing+NPK 75% ( 229,5g ) 10 kg Pupuk Kompos Kotoran Kambing + NPK 50% ( 153g );15 kg Pupuk Kompos Kotoran Kambing + NPK 25 % ( 76,5g ); 20 kg Pupuk Kompos otoran Kambing . Pengamatan dilakukan terhadap peubah pertambahan tinggi tanaman, pertambahan diameter batang, pertambahan jumlah anakan daun, waktu pecah tunas dan pertambahan jumlah cabang. Hasil penelitian menunjukkan bahwa pemberian pupuk kompos kambing yang dikombinasi dengan NPK dapat menstubtitusi pemakaian pupuk NPK. Pemberian kompos kotoran kambing 15 kg dengan NPK 76,5 g menunjukkan kombinasi terbaik
Penerapan load balancing web server berbasis K-NN dengan open vSwitch dalam Jaringan software defined network
No full textSoftware Defined Network is a popular computer network concept today because of the ease of managing network traffic with the control plane. Massive internet usage makes web server services on SDN networks overloaded. There are many load balancing concepts to overcome this problem, one of which is implementing the K-NN algorithm. This study aims to maximize the performance of the K-NN algorithm on SDN networks by optimizing the K value using Grid Search Cross Validation, and adding server status selection logic based on the smallest disk if the server status calculated by K-NN has the same. All implementations of the load balancing concept in this study were created virtually using Open vSwitch and virtualbox. Testing was carried out using CPU, MEMORY, and DISK parameters sent by the server with the help of the psutils component. JMeter software was used for testing by sending data using the POST method. The data type is text/plain with a data size of 1MB, testing was carried out in stages with threads 100, 200, 300, 400. The test results showed that the performance of the K-NN algorithm was running optimally. There was no significant difference in the distribution of the load to the server, this made the optimization and addition of logic successful