Deep learning is a part of machine learning method that uses artificial neural network (ANN). The type of learning in deep learning can be supervised, semi-supervised, and unsupervised [7] . CNN & RNN (Supervised) and RBM & Autoencoder (Unsupervised) are deep learning algorithms. All of the above algorithms have uses in their respective fields, depending on what we want to use them for. One of the most frequently used cases for deep learning is object detection and classification. The Convolutional Neural Network (CNN) algorithm is the most widely used algorithm for object detection cases, one of the reasons because it is supported by Google's Tensorflow framework, but it turns out that there is one object detection algorithm that has a higher level of accuracy and processing speed, namely You Only Look Once (YOLO) which can run on 2 frameworks (Darknet & Darkflow) and is supported by GPU. That's why here the author prefers to do object detection with the You Only Look Once (YOLO) method. The research data with the title Palembang Food Detection Object Using the YOLO (You Only Look Once) Algorithm is a sample photo of food from Google Image. There are 31 types of Palembang specialties, each type consists of approximately 50 to 70 images, so the total images used are from 31 types of Palembang foods, namely 1955 images with jpeg format for training data, and 31 images with jpeg format typical Palembang foods for test data.
Technology is growing with the times, relentlessly and constantly moving forward to make new breakthroughs in all aspects of life. The IEA report in the Study of Reading Literacy states that the ability of elementary school children in Indonesia is very low. Of the 31 countries studied, Indonesia was ranked 30th. Based on the explanation of the background above, the researchers formulated the problem in this study, namely how to design and build an Android-based educational game with unity using the Game development life cycle (GDLC). This research was conducted in August 2021 in 9/10 Ulu Village, Plaju District, Palembang with a sample of 10 children. The research method uses Action Research and application development methods use GDLC. The results of testing the game system with blackbox that can be operated smoothly and for the results of testing the feasibility of games with the results of 84.6% in the appropriate category, the results of development with Unity and GDLC succeeded in creating learning media based on android games quite well.
Purpose: This research aims to identify content that contains cyberbullying on Twitter. We also conducted a comparative study of several classification algorithms, namely NB, DT, LR, and SVM. The dataset we use comes from Twitter data which is then manually labeled and validated by language experts. This study used 1065 data with a label distribution, namely 638 data with a non-bullying label and 427 with a bullying label.Methods: The weighting process for each word uses the bag of word (BOW) method, which uses three weighting features. The three-word vector weighting features used include unigram, bigram, and trigram. The experiment was conducted with two scenarios, namely testing to find the best accuracy value with the three features. The following scenario looks at the overall comparison of the algorithm's performance against all the features used.Result: The experimental results show that for the measurement of accuracy weighting based on features and algorithms, the SVM classification algorithm outperformed other algorithms with a percentage of 76%. Then for the weighting based on the average recall, the DT classification algorithm outperformed the other algorithms by an average of 76%. Another test for measuring overall performance (F-measure) based on accuracy and precision, the SVM classification algorithm, managed to outperform other algorithms with an F-measure of 82%.Value: Based on several experiments conducted, the SVM classification algorithm can detect words containing cyberbullying on social media.
This paper describes the performance evaluation of fixed step closed loop power control algorithm in a novel wireless channel that is called High Altitude Platforms (HAPs). This new wireless delivery method is proposed as a complementary system in providing the next generation services in which the technology basically employs CDMA. In HAPs communication, the channel is predicted to have different characteristic com pared to that in terrestrial channel. In this work, HAPs channel is modeled to follow Ricean fading distribution whose K factor is obtained based on experimental measurement. Fixed step power control algorithm is then computer simulated under such a channel to evaluate its performance. The performance is presented in terms of step size of the power control, users elevation angle, feedback delay, and SIR estimation error. We found the performance of fixed step closed loop power control in HAPs channel increases with the increase of step size and elevation angle. Feedback delay has insignificant effect to the power control performance. SIR estimation error degrades the performance of the power control compared with the true SIR estimation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.