JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
Grouping of Image Patterns Using Inceptionv3 For Face Shape Classification
Get PDFThe human face is an extraordinary part where nearly everybody is not quite the same as each other. One perspective that should be visible plainly is the shape. Face shape grouping can be used for amusement, security, or excellence. One technique that can be utilized in picture grouping is the InceptionV3 model. InceptionV3 is the structure of the Convolutional Neural Network (CNN) created by Google, which can tackle picture examination and item discovery issues. This engineering is utilized to order face shapes into five classes: Round, Heart, Square, Oblong, and Oval. At that point, the Google Pictures dataset goes through the pre-handling stage, and the Shrewd Edge Identifier is applied to each picture. Hair turns into a commotion. Consider recognizing the side of the face because it does not make any difference what the hairdo resembles. What is important is the side of the face. When there is a dataset of elongated class and heart class with a comparable hairdo, InceptionV3 will identify the component and expect the two pieces of information to come from a similar class. The exchange learning strategy is done in preparation for the last Layer of ImageNet's InceptionV3 model. This strategy puts the high precision level with an exactness of 93% preparation and testing between 88% - 98%. InceptionV3 could arrange upwards of 692 from 747 datasets or around 92.65%. The most reduced information class is the heart class, where out of 150 information, InceptionV3 can characterize upwards of 130 information
Case Study: Using Data Mining to Predict Student Performance Based on Demographic Attributes
Get PDFThis study predicts student performance at Universiti Pertahanan Nasional Malaysia (UPNM) based on their socio-demographic profile; it also determines how a prediction algorithm can be used to classify the student data for the most significant demographic attributes. The analytical pattern in academic results per batch has been identified using demographic attributes and the student's grades to improve short-term and long-term learning and teaching plans. Understanding the likely outcome of the education process based on predictions can help UPNM lecturers enhance the achievements of the subsequent batch of students by modifying the factors contributing to the prior success. This study identifies and predicts student performance using data mining and classification techniques such as decision trees, neural networks, and k-nearest neighbors. This frequently adopted method comprises data selection and preparation, cleansing, incorporating previous knowledge datasets, and interpreting precise solutions. This study presents the simplified output from each data mining method to facilitate a better understanding of the result and determine the best data mining method. The results show that the critical attributes influencing student performance are gender, age, and student status. The Neural Networks method has the lowest Root of the Mean of the Square of Errors (RMSE) for accuracy measurement. In contrast, the decision tree method has the highest RMSE, which indicates that the decision tree method has a lower performance accuracy. Moreover, the correlation coefficient for the k-nearest neighbor has been recorded as less than one
GoEkopz: An E-Koperasi and Marketplace Synergy of Koperasi MSMEs Model Platform - Case Study: Koperasi Giat, eKopz Startup, PPKM Community
Get PDFWith a population of 273.5 million in 2020, the people's economy is very important for Indonesia. The populist economic model has existed in Indonesia for a long time, locally known as “Koperasiâ€. However, Koperasi could not keep up with information and technology in the industrial 4.0 era. When other industrial models adopt information technology massively, Koperasi seems to struggle to shift from the conventional model. Slowly but surely, Koperasi become less popular, especially among Generation Z, who considers information technology part of their daily needs and lifestyle. The scope of Koperasi’sbusiness shrinks and becomes limited, along with their business capital. On the other hand, information technology provides opportunities for small and medium-sized entrepreneurs. The data shows that this sector contributes to 60.5% of the national GDP, absorbs 96.9% of the workforce, and provides 99.9% of total employment. Unfortunately, many Micro, Small, and Medium Enterprises (MSME) have limited capital and tend to prefer fintech services that offer easier accessibility than Koperasi. This paper aims to propose an E-Koperasi and the synergy of Koperasi and MSMEs model platform, specifically the digital marketplace platform. We design the platform requirements using the Software Requirement Specification with the User Acceptance Test. As a result, an application is developed as a model platform available for Koperasi and MSMEs. The platform is proposed to support the Government’s program to digitalize the Koperasi and the MSMEs and increase their competitiveness in Industrial 4.0
A New Face Region Recovery Algorithm based on Bicubic Interpolation
Get PDFRecently, researchers focused on face image manipulation detection and localization techniques because of their importance in image security applications. The previous research has not highlighted the recovery of the face region after manipulation detection. This paper presents a new face region recovery algorithm (FRRA) to be included in the face image manipulation detection algorithms (FIMD). The proposed FRRA consists of two main algorithms: face data generation algorithm and face region restoration algorithm. Both algorithms start by detecting the face region using Multi-task Cascaded Neural Network followed by a face window selection process. In the face data generation algorithm, the recovery information is generated from the shirked face window using bicubic interpolation technique. In the face region restoration algorithm, the face region zoomed using bicubic interpolation technique. The proposed FRRA has been tested and compared with previous recovery methods for different color face images, and the results proved that the FRRA could recover the face region with better visual quality at the same data length compared to previous methods. The main contributions of this research are a) the suggestion of including a face region recovery algorithm to FIMD, b) the study of previous recovery data generation algorithms for color face images, and c) introducing a new algorithm for generating the recovery data based on bicubic interpolation. In the future, the proposed algorithm can be included in the recent FIMD algorithms to recover the face region, which can be very useful in practical applications, especially those used in data forensics systems
Economic Impact due Covid-19 Pandemic: Sentiment Analysis on Twitter Using Naïve Bayes Classifier and Support Vector Machine
Get PDFCovid-19 is an outbreak caused by severe acute respiratory syndrome. Covid-19 first appeared in Indonesia on March 2, 2020, with two confirmed cases and increased to 1285 cases in 30 provinces. One of the impacts of the Covid-19 pandemic is on the economic aspect, which has experienced a drastic decline in income. This study aims to classify public opinion to determine the level of public sentiment on the economic impact of the Covid-19 pandemic and to identify parameters that influence the accuracy of the sentiment analysis classification model. The methods used in this current research are Lexicon, Support Vector Machine (SVM), and Naive Bayes Classifier (NBC). First, Lexicon is used for scoring and labeling the preprocessed data. Second, SVM is used to classify the sentiment, then find the best accuracy using linear, radial, polynomial, and sigmoid kernels. Third, NBC is used to classify sentiment as a comparison method. The results indicated that 255 tweet data consisted of 44 positive tweets (17.25%), 46 neutral tweets (18.04%), and 165 negative tweets (64.71%). Therefore, it can be inferred that the economic impact on the Indonesian people due to the Covid-19 pandemic has a high negative sentiment value. In the performance, SVM yielded a better accuracy of 100%, precision, recall, and F-measure are 1. This study proves that selecting the kernel type and applying underfitting can improve the accuracy of SVM. Also, SVM can perform well on a small amount of training data
Feature Selection to Enhance DDoS Detection Using Hybrid N-Gram Heuristic Techniques
Get PDFVarious forms of distributed denial of service (DDoS) assault systems and servers, including traffic overload, request overload, and website breakdowns. Heuristic-based DDoS attack detection is a combination of anomaly-based and pattern-based methods, and it is one of three DDoS attack detection techniques available. The pattern-based method compares a sequence of data packets sent across a computer network using a set of criteria. However, it cannot identify modern assault types, and anomaly-based methods take advantage of the habits that occur in a system. However, this method is difficult to apply because the accuracy is still low, and the false positives are relatively high. Therefore, this study proposes feature selection based on Hybrid N-Gram Heuristic Techniques. The research starts with the conversion process, package extract, and hex payload analysis, focusing on the HTTP protocol. The results show the Hybrid N-Gram Heuristic-based feature selection for the CIC-2017 dataset with the SVM algorithm on the CSDPayload+N-Gram feature with a 4-Gram accuracy rate of 99.86%, MIB- Dataset 2016 with the 2016 algorithm. SVM and CSPayload feature +N-Gram with 100% accuracy for 4-Gram, H2N-Payload Dataset with SVM Algorithm, and CSDPayload+N-Gram feature with 100% accuracy for 4-Gram. As a comparison, the KNN algorithm for 4-Gram has an accuracy rate of 99.44%, and the Neural Network Algorithm has an accuracy rate of 100% for 4-Gram. Thus, the best algorithm for DDoS detection is SVM with Hybrid N-Gram (4-Gram)
Stock Price Movement Classification Using Ensembled Model of Long Short-Term Memory (LSTM) and Random Forest (RF)
Get PDFStock investing is known worldwide as a passive income available for everyone. To increase the profit possibly gained, many researchers and investors brainstorm to gain a strategy with the most profit. Machine learning and deep learning are two of these approaches to predicting the stock's movement and deciding the strategy to gain as much as possible. To reach this goal, the researcher experiments with Random Forest (RF) and Long Short-Term Memory (LSTM) by trying them individually and merging them into an ensembled model. The researcher used RF to classify the results from LSTM models obtained throughout the Hyperparameter Optimization (HPO) process. This idea is implemented to lessen the time needed to train and optimize each LSTM model inside the ensembled model. Another anticipation done in this research to overcome the time needed to train the model is classifying the return for longer periods. The dataset used in this model is 45 stocks listed in LQ45 as of August 2021 This research results in showing that LSTM gives better results than RF model especially when using Bayesian Optimization as the HPO method, and that the ensembled model can return better precision in predicting stocks in comparison to the LSTM model itself. Future improvement can focus on the model structure, additional model types as the ensemble model estimator, improvement on the model efficiency, and datasets research to be used in predicting the stock movement predictio
Evaluation of the Visual Learning Application for Mathematics using Holography Display for the topic on Shape and Space
Get PDFMathematics is an important foundation in the life of an individual. The problems associated with learning Mathematics that are commonly encountered, are due to factors such as: abstract phenomena and concepts, low imagination, and lack of understanding of the concepts being studied. Thus, the purpose of this study is to help primary school children improve their ability to recognise 3D shapes that are abstract phenomena. This paper presents the Effectiveness Usability Evaluation of the Visual Learning Application for Mathematics using Holography Display for the topic on Shape and Space called MEL-VIS. This study was conducted on eighty (80) primary school students. The results of the study showed that learning about 3D shapes with the E-Visual MEL-VIS application prototype is more effective than traditional methods
Classification of Predicting Customer Ad Clicks Using Logistic Regression and k-Nearest Neighbors
Get PDFNowadays, conventional marketing techniques have changed to online (digital) marketing techniques requiring internet access. Online marketing techniques have many advantages, especially in terms of cost efficiency and fast information delivery to the public. Therefore, many companies are interested in online marketing and advertising on social media platforms and websites. However, one of the challenges for companies in online marketing is determining the right target consumers since if they target consumers who are not interested in buying the product, the advertising costs will be high. One use of online advertising is clicks on ads which is a marketing measurement of how many users click on the online ad. Thus, companies need a click prediction system to know the right target consumers. And different types of advertisers and search engines rely on modeling to predict ad clicks accurately. This paper constructs the customer ad clicks prediction model using the machine learning approach that becomes more sophisticated in effectively predicting the probability of a click. We propose two classification algorithms: the logistic regression (LR) classifier, which produces probabilistic outputs, and the k-nearest neighbors (k-NN) classifier, which produces non-probabilistic outputs. Furthermore, this study compares the two classification algorithms and determines the best algorithm based on their performance. We calculate the confusion matrix and several metrics: precision, recall, accuracy, F1-score, and AUC-ROC. The experiments show that the logistic regression algorithm performs best on a given dataset
3D Augmented Reality Marker-based Mobile Apps Design of Face Mask Layer
Get PDFThe outbreak of COVID-19 has spread rapidly across the globe, affecting how people interact, work and experience their daily lifestyles. The face mask is essential to personal protective equipment (PPE) even though COVID-19 is endemic today. A face mask is required to protect humans, and its description is important. Augmented reality is one new attractive technology for distributing information. This research presents a 3D augmented reality mobile application that visualizes a protective face mask and its material layers called 3DAR-FML. It was designed by utilizing 3D marker-based images on an Android smartphone platform. Ten types of protective face masks and each layer were designed in 3D images using Blender Animation and Unity 3D software. Vuforia Engine applications were used to build a personalized mobile Augmented Reality application. This initiative educates society and markets products in interactive ways that allow visualization of the 3D model of various face masks from android phones. Results present that successful mobile apps were developed. A survey shows that 70% of respondents agreed on the design app based on menu interactivity. This app also has been identified as a helper tool for The National Institute for Occupational Safety and Health (NIOSH) agency Malaysia for safety and health apps for the market and public. This product is significant in societies to gain information on face mask characteristics that help to contribute content based on the standard mask in Malaysia