This research aims to apply the localizing region-based active contour (LRAC) method to acquire the femur length in an ultrasound image automatically and to determine the effect of noise removal on the segmentation accuracy. The automatic femur length measurement system includes three main steps. The first step is the denoising process to reduce speckle noise in the ultrasound image. Afterwards, the LRAC method is applied to detect and segment a local region. The segmentation process with a certain number of iterations and a weight of the smoothing terms is started at the selected initial pixel. At the final step, the femur length is measured to estimate the gestational age. The experiment results show that the accuracy of the estimated gestational age increases significantly when the noise reduction technique is employed.

Evaluation of learning systems based on e-learning is very important to determine learning success. The purpose of this study is to obtain predictive results from evaluating students who follow e-learning based learning systems. The data used is the result of logs of student learning activities taken from the LMS. The data used in this study were 641 user logs of student activity. In predicting the evaluation results based on the learning system on e-learning we use a neural network method based on swarm particle optimization. Neural Network has a problem in optimizing very large data so using swarm particle optimization can solve this problem. From the data testing we have done, the results obtained by the Neural Network method get an accuracy value of 95.47%, and the results of the AUC value of 97.90%. The observation of variables C, ∊ and population of Neural Network and particle swarm optimization use the K-Fold Cross Validation method. Then the researchers tested several choices on the attributes used. By using the Neural Network method based on the swarm particle optimization attribute, there are 9 predictor variables so that as many as 6 attributes are used, namely sports, chat, discussion, messages, Quiz exercises and total logs. The results show an accuracy rate higher than 97.50%, and an AUC value of 98.20%. So the accuracy value increased by 2.03% and the AUC increased by 0.3%. With accuracy and AUC values, the Artificial Neural Network algorithm based on particle optimization is very well categorized.

No abstract

Perangkat lunak yang bermutu ditentukan oleh jumlah cacat yang ditemukan pada saat proses pengujian. Proses perbaikan perangkat lunak setelah terdistribusi memiliki resiko yang lebih tinggi. Beberapa metode telah diujikan untuk memprediksi cacat pada perangkat lunak. Secara umum dataset software metrics telah digunakan sebagai acuan. Dataset software metrics bersifat tidak seimbang sehingga berpengaruh terhadap tingkat akurasi pemrediksi cacat perangkat lunak.  Pada tahapan pra pemrosesan, digunakan metode Particle Swarm optimization (PSO) untuk mengatasi masalah polusi data serta metode Random Over Sampling (ROS) untuk menangani ketidak seimbangan kelas pada dataset. Metode yang diusulkan pada penelitian ini yaitu algoritma decision tree J48 yang dioptimalkan dengan teknik adaboost. Dataset software metrics yang digunakan pada penelitian ini bersumber pada  dataset PROMISE repository. Hasil penelitian menunjukan bahwa penggunaan teknik adaboost pada algoritma decision tree J48 layak digunakan sebagai metode untuk memprediksi cacat pada perangkat lunak dengan nilai akurasi mencapai 93,507% dan nilai AUC mencapai 0,935