Search

Pandemi corona telah mengubah proses pembelajaran dari yang semula tatap muka secara langsung (offline) menjadi pembelajaran secara daring (online). Pembelajaran daring ini menyebabkan kesulitan dalam pemantauan perilaku siswa oleh guru karena berkurangnya interaksi secara langsung. Lebih dari itu, siswa seringkali merasa terisolasi sehingga jika dibiarkan, situasi ini akan menyebabkan kegagalan dalam prestasi belajarnya. Permasalahan ini mendorong banyak dilakukannya penelitian tentang pemodelan yang berkaitan dengan perilaku siswa. Namun, para peneliti sebelumnya tidak banyak yang fokus pada peningkatan kinerja model atau sistem yang dibangun, padahal kinerja model ini sangat berpengaruh terhadap kualitas hasil pemetaan perilaku siswa. Untuk itu, makalah ini berfokus pada peningkatan kinerja pengklasteran perilaku siswa ketika berinteraksi dengan sistem e-Learning. Peningkatan kinerja dilakukan dengan reduksi dimensi pada data siswa dengan Principal Component Analysis (PCA). Selanjutnya, dua teknik inisialisasi titik pusat klaster dieksplorasi untuk mendapatkan hasil yang optimal, yaitu: random dan K-means++. Untuk pengukuran kualitas klaster, makalah ini menggunakan silhouette index. Hasil pengujian menunjukkan bahwa klaster dengan kualitas tertinggi dicapai oleh penerapan PCA dengan tujuh komponen dan banyaknya klaster tiga sampai empat untuk semua teknik inisialisasi titik pusat. Klaster yang berkualitas ini dapat membantu guru dalam memonitor perilaku siswa pada pembelajaran secara daring.

The corona pandemic has changed the learning method from conventional to a more flexible one, such as through the internet. Consequently, students may have less direct interaction with teachers. This condition has made it difficult for teachers to monitor the students' behavior. This research works on this problem by focussing on the clustering of students' behavior using the DBSCAN, which is a density-based algorithm. Noises generated in this process can be considered students who do the uncommon behavior when taking the e-Learning system. Further, we evaluate the resulted clusters using the silhouette index to find their quality. The experimental result shows that the DBSCAN can differentiate clusters containing noises. By taking the silhouette index, the Manhattan distance parameter is superior to that of Euclidean.

Violence detection research is still quite a challenge for researchers and a considerable amount of effort. Before the video can be processed for classification, feature extraction is an important process to obtain important information. Determination of feature extraction and classification algorithms is an important factor for accurate classification results. This study uses deep transfer learning for feature extraction and combining it with the Support Vector Machine (SVM) classifier. The deep transfer learning algorithm in this study is a pre-trained model of Visual Geometry Group Network-16 (VGGNet-16). The video data was extracted using VGGNet-16 and then classified using SVM. Tests were carried out with 5-fold cross-validation with a variety of linear kernel, RBF, and Polynomial functions. The results were also compared with the Principal Component Analysis (PCA) feature extraction algorithm combining with SVM also. The results showed that the combination of deep transfer learning with SVM linear kernel functions resulted in higher accuracy compared to RBF and Polynomial kernel functions, and also compared to PCA combined with SVM.

This study aims to describe the learning outcomes and students’ responses after joining in the Foundation of Mathematics courses using virtual learning. The research subjects were Mathematics students who programed the Foundation of Mathematics which consisted of 31 people. The method of data collection was carried out by test methods and questionnaires. This learning was said to be successful if the learning outcomes of students at least 70% get A. A-, B+, or B grades or at least 70% the scores obtained by students was more than or equal to 70, and students’ responses after joining in Foundation of Mathematics courses using virtual learning is positive category (the percentage of students positive responses was more than 70%). Based on the results of data analysis it can be stated that 90.03% of the final grades of students joining in the Foundation of Mathematics courses are B, B+, A- or A, and the percentage of positive responses of students after joining in Foundation of Mathematics courses was 80.0%. Thus it can be concluded that Foundation of Mathematics courses using virtual learning has been successful.

Plant recognition based on digital leaf image has received as particular attention in computer vision and intelligence system, due its important implication in automatic plant identification. Plant species have the unique leaf characteristics such as the shape, texture, margin, and colour, which different each other. This study presents a novel method for automation plant recognition using Generalized Relevance Learning Vector Quantization (GRLVQ). GRLVQ is a competitive based learning algorithm which is integrating features extraction and classification phases. The experimental result shows that GRLVQ has better performance than the predecessor algorithm.

Animal tracker is an important phase in animal behavior analysis. It leads to understanding how, when, and why the animal use the environmental resources, how, where, and when they interact with each other, with other species, and with their environment. Understanding the animal behavior is providing the link to population distribution which is essential for predicting the human-caused environmental change and guidance for conservation strategies. Tracking and detecting the animal is time and cost consuming. Machine Learning can relieve this burden by detecting animal automatically. Complex-Valued Neural Network is a method of Machine Learning that is challenging and interesting to be explored. This study applied of Complex-Valued Neural Network (CVNN) for animal tracking, especially in detecting the animal species. The experiment results present that CVNN is robust to recognition the animal automatically.

Gamelan is a set of musical instruments from Indonesia and played together as an orchestra. In gamelan playing, one of instruments will act as a conductor so the sound of this instrument must be separated from the others.