JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
The Gamification of E-learning Environments for Learning Programming
Get PDFGamification is the most active methodology utilized in the E-learning environment for teaching-learning in computing; however, this does not restrict its use in other areas of knowledge. Gamification combines elements of play and its design techniques in a non-ludic context, achieving a motivation factor for the students. This systematic study aimed to collect and synthesize scientific evidence from the gamification field for learning programming through the E-learning environment. In order to do this, a systematic literature review was done, following the guidelines proposed by Petersen, which propose the definition of questions, search strategies, inclusion/exclusion criteria, and characterization. As a result of this process, eighty-one works were completely reviewed, analyzed, and categorized. The results revealed favorable learning among the students, the most used platforms and gamification elements, the most used languages and focuses of programming, and the education level, where gamification is most used to learn to program in an E-learning environment. These findings evidenced that gamification is a good active strategy for introducing beginning students to programming through an E-learning environment. Within this context, Learning programming through the use of gamification is a topic that is growing and taking force, and after what occurred during the pandemic, it is projected that there will continue to be more students who are focused on understanding its implementation and the impact it has on the different levels of education and the areas of knowledge
Performance Improvement of Deep Convolutional Networks for Aerial Imagery Segmentation of Natural Disaster-Affected Areas
Get PDFThis study proposes a framework for improving performance and exploring the application of Deep Convolutional Networks (DCN) using the best parameters and criteria to accurately produce aerial imagery semantic segmentation of natural disaster-affected areas. This study utilizes two models: U-Net and Pyramid Scene Parsing Network (PSPNet). Extensive study results show that the Grid Search algorithm can improve the performance of the two models used, whereas previous research has not used the Grid Search algorithm to improve performance in aerial imagery segmentation of natural disaster-affected areas. The Grid Search algorithm performs parameter tuning on DCN, data augmentation criteria tuning, and dataset criteria tuning for pre-training. The most optimal DCN model is shown by PSPNet (152) (bpc), using the best parameters and criteria, with a mean Intersection over Union (mIoU) of 83.34%, a significant mIoU increase of 43.09% compared to using only the default parameters and criteria (baselines). The validation results using the k-fold cross-validation method on the most optimal DCN model produced an average accuracy of 99.04%. PSPNet(152) (bpc) can detect and identify various objects with irregular shapes and sizes, can detect and identify various important objects affected by natural disasters such as flooded buildings and roads, and can detect and identify objects with small shapes such as vehicles and pools, which are the most challenging task for semantic segmentation network models. This study also shows that increasing the network layers in the PSPNet-(18, 34, 50, 101, 152) model, which uses the best parameters and criteria, improves the model's performance. The results of this study indicate the need to utilize a special dataset from aerial imagery originating from the Unmanned Aerial Vehicle (UAV) during the pre-training stage for transfer learning to improve DCN performance for further research
Storychart: A Character Interaction Chart for Visualizing the Activities Flow
Get PDFEvent-predicate-based storyline extraction results in a chronologically ordered activity journal. The extraction results contain complex human activities, so the activity journal requires a visualization model to describe actor interactions. This paper proposes a chart to visualize the activities' flow to describe the characters' interactions in an activity journal. This chart is called a storychart. Storycharts have an actor channel that can accept single entities or teams. The actor channel allows changing the type from single to a team or vice versa and moving members to other teams. The activity channel serves as a connector to accommodate interactions between actors. The activity channel provides a visual space for the elements of what, where, and when. Event predicates are the core of what. Therefore, the storychart visualizes the event predicate using glyphs to attract the reader’s attention. The main contribution of this paper is to introduce a team channel that can visualize the identity of team members and an activity channel that can visualize the details of events. We invited participants to discover the reader’s perception of the ease of team recognition and the integrity of the meaning of the narrative visualized by the storychart. Participants involved in the evaluation were filtered by literacy score. Evaluation of storychart reading showed that readers could easily distinguish teams from single actors, and storycharts could convey the story in the activity journal with little reduction in meaning
Wi-Fi Fingerprint for Indoor Keyless Entry Systems with Ensemble Learning Regression-Classification Model
No full textKeyless entry systems have made human life more comfortable because the possibility of someone leaving a key is non-existent. However, there is a research opportunity to use Wi-Fi fingerprint localization in an indoor keyless entry system. The use of Wi-Fi as a solution is low cost because of the already hdeployed Wi-Fi access points (AP) around buildings. This study uses ensemble learning to utilize Wi-Fi fingerprints for keyless entry systems in indoor environments. The first step of this research is to obtain the UJIIndoorLoc Data Set, an open-source Wi-Fi fingerprint dataset. The next step is to design and implement a keyless entry system based on Wi-Fi Fingerprint. We compared four ensemble learning methods: Random Forest, AdaBoost, gradient boosting, and XGBoost. We use several metrics to compare the four methods, namely r2, the area under the receiver operating curve (AUROC), and the equal error rate (EER). The test results show that for localization based on ensemble regression, XGBoost has the best performance compared to the other three ensemble methods for both longitude and latitude predictions. The r2values are 0.94 and 0.98, respectively. For authentication based on ensemble classification, the gradient-boosting model performance increases as the number of weak learners increases. The optimum number of weak learners is 60. With AUROC = 0.99, gradient boosting outperforms random forest, AdaBoost, and XGBoost on authentication. Finally, our authentication method has EER = 0.01, which outperforms other state-of-the-art keyless entry systems. This research focuses on local building map datasets for future work
Preliminary study: Readiness of WLAN Infrastructure at Malaysian Higher Education Institutes to support Smart Campus Initiative
Get PDFSmart campus initiative enables higher education to enhance services, decision-making, and campus sustainability. The initiatives are being actively implemented globally by higher education, including in Malaysia. The recent COVID-19 pandemic has underscored the need for the education sector to explore a digital revolution. The adaptation of digital technologies has improved many aspects, including the teaching and learning experiences and administration tasks, which results in more efficient task handling. This study investigates the readiness of the WLAN infrastructure at Malaysian Public Higher Education Institutes (HEIs) in implementing smart campus initiatives and measures readiness based on the availability of WLAN Infrastructure, WLAN logical architecture and WLAN populated coverage area. This study administered a questionnaire to 19 respondents, all of whom are IT personnel from Malaysian public HEIs to gather preliminary data on the readiness of WLAN infrastructure at Malaysian Public HEI to support the adaptation of smart campus initiatives in their teaching and learning activities. This study is a preliminary study concerning the readiness of WLAN infrastructure at Malaysian Public HEI in adapting smart campus initiatives. The findings show that, even though WLAN service is available at all Malaysian Public HEI, it is essential to enhance the adopted logical architecture and WLAN coverage to prepare HEI to become smart campuses. The findings of this study can provide the fundamental guidelines for the Ministry of Higher Education in determining the baseline of WLAN infrastructure required by Malaysian HEI to support smart campus initiatives
Fake News Detection in Indonesian Popular News Portal Using Machine Learning For Visual Impairment
No full textIt has become a necessity for people to communicate with each other to complete their needs. The exchange of information conveyed in communication often cannot be directly assessed, especially online news. They just get news and are unable to filter out inappropriate stuff. The media website conveys a great deal of information. Popular news websites are one source for keeping up with the newest news. It requires a significant amount of work to deliver news on prominent websites and to choose content that is not incorrect. To crawl the web and analyse enormous data, massive computer power is required, and solutions to lower the process's space and temporal complexity must be created.Data mining is seen to be a solution to the aforementioned difficulties since it extracts particular information based on defined attributes. This research investigated a model to determine the content of false news information in Indonesian popular news. Firstly, preprocessing process from dataset that collected from keaggle. Secondly, we try use classification methods to determined which the optimal method to classify fake news. Thirdly, we use another public dataset for testing method. Furthermore, five machine learning classifiers are compared: Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree Classifier (DTC), Gradient Boosting Classifier (GBC), and Random Forest (RF). These classifications are utilized independently before being compared based on receiver operating characteristic curves and accuracy. The experimental result shows that DTC has the lowest accuracy of 75.33% and SVM has the highest accuracy of 83.55%.Â
Skew Correction and Image Cleaning Handwriting Recognition Using a Convolutional Neural Network
Get PDFHandwriting recognition is a study of Optical Character Recognition (OCR) which has a high level of complexity. In addition, everyone has a unique and inconsistent handwriting style in writing characters upright, affecting recognition success. However, proper pre-processing and classification algorithms affect the success of pattern recognition systems. This paper proposes a pre-processing method for handwriting image recognition using a convolutional neural network (CNN). This study uses public datasets for training and private datasets for testing. This pre-processing consists of three processes: image cleaning, skew correction, and segmentation. These three processes aim to clean the image from unnecessary ink streaks. In addition, to make angle corrections to characters in italics in their writing. The model testing process uses image test data of handwriting that are not straight. There are three images based on the inclination angle: less than 45 degrees, equal to 45 degrees, and more than 45 degrees. Picture cleaning removes unnecessary strokes (noise) from the image using a layer mask, whereas skew correction changes the handwriting to an upright posture based on the detected angle. The pre-processing model we propose worked optimally on handwriting with a skew angle of fewer than 45 degrees and 45 degrees. Our proposed model generally works well for handwriting with fewer than 45 degrees skew with an accuracy of 88,96%. Research with a similar scope can continue to improve optimization with a focus on algorithms related to analysis layout studies. Besides that, it can focus more on automation in the segmentation process of each character
Implementation of Convolutional Neural Network and Long Short-Term Memory Algorithms in Human Activity Recognition Based on Visual Processing Video
Get PDFHuman Activity Recognition (HAR) is an interesting research topic, especially in identifying human movement actions focusing on video-based security surveillance. Symptom of an illness from a movement. The use of HAR in this research is the key to better understanding the various semantics contained in the video to find out the pattern of a human movement, especially in sports movements. In this study, a combination of the CNN and LSTM method algorithms was applied by using several variations of the model parameter values on the dropout layer and batch size to convert the pattern in the video into image form to produce a HAR model. Data processing at the convolution layer is used to extract spatial features in the frame. The extraction results are fed to the LSTM layer on each network for modeling the temporal sequence of human movement. In this way, the network on the model will learn spatiotemporal features directly in end-to-end data training tests to produce a robust model. The test data used are 10 sports activities obtained from related research from the University of Central Florida (UCF). The results showed that the performance was quite good, although there were still errors in the classification of sports activities because they had similarities in the movements of the activities carried out. The classification results show a loss value of 0.4 and an accuracy of 0.94. In further research, what needs to be corrected is the loss value which is still high so that several times the test results show an error in the classification of sports activities that have similarities in the movements of the activities
Player's Affective States as Meta AI Design on Augmented Reality Games
Get PDFGames are considered one of the most popular entertainment forms worldwide. The interaction in the game environment makes the players addicted to playing the game. One technique to build an addicting game is utilizing the player's emotions using Meta Artificial Intelligence (AI). The player's emotions can be utilized by adjusting the game difficulty. Most of the game offers static and steady difficulty development throughout the game. This research proposes a Meta AI game design using the player's affective states. We argue that a dynamic difficulty development throughout the game will increase the player's game experiences. The player's facial expressions are utilized to extract the player's affective state information. To recognize the player's facial expressions, a Facial Expressions Recognition (FER) model was trained using VGG-16 architecture and The Indonesian Mixed Emotion Dataset (IMED) dataset in addition to a self-collected dataset. The emotions recognition model (from player's facial expressions) achieved the best validation accuracy of 99.98%. The model was implemented in the proposed Meta AI game design. The Meta AI game design proposed in this game was implemented in several game scenarios to be compared and evaluated. The proposed Meta AI game design was evaluated by 31 respondents using Game Experiences Questionnaire (GEQ). Overall, the results show that the game with Meta AI and Augmented Reality implemented significantly improved the Game Experiences Questionnaire (GEQ) score and the player's overall satisfaction compared to the other game scenarios
Classification of Sugarcane Area Using Landsat 8 and Random Forest based on Phenology Knowledge
Get PDFIndonesia is one of the largest countries globally with an area for planting sugarcane. The current problem is that determining the planting area of sugarcane is still done conventionally; this is very limited and wastes time. Thus, knowing the sugarcane planting area becomes essential for policymaking through Remote Sensing technology. However, the challenge of Remote Sensing is the limited data due to weather and the spectral variability of other plants. So, it is necessary to classify based on phenological knowledge. The study aims to classify sugarcane areas based on phenological knowledge using Remote Sensing and Machine Learning. This application finished on the cloud platform Google Earth Engine (GEE) through Landsat 8 satellite imagery data. The knowledge of sugarcane phenology was built based on the Normalized Difference Vegetation Index (NDVI) spectral value and built with the harmonic model. In addition, classification is accomplished by object-oriented (OO) methods for segmentation classification. Object-oriented is solved by the Simple Non-Iterative Clustering (SNIC) algorithm for spatial cluster identification, the Gray-Level Co-occurrence Matrix (GLCM) for texture extraction, and the Random Forest algorithm for Land Use-Land Cover (LULC) classification. The results of the accuracy analysis using the confusion matrix and the classification of sugar cane areas based on phenological knowledge obtained the best results with an overall accuracy of 95.9% with a Kappa coefficient of 0.92. It can be concluded that a classification approach with knowledge of plant phenology can help better classify the availability of land for plantations in the future