JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
Enhance Document Contextual Using Attention-LSTM to Eliminate Sparse Data Matrix for E-Commerce Recommender System
Get PDFE-commerce has been the most important service in the last two decades. E-commerce services influence the growth of the economic impact worldwide. A recommender system is an essential mechanism for calculating product information for e-commerce users. The successfulness of recommender system adoption influences the target revenue of an e-commerce company. Collaborative filtering (CF) is the most popular algorithm for creating a recommender system. CF applied a matrix factorization mechanism to calculate the relationship between user and product using rating variable as intersection value between user and product. However, the number of ratings is very sparse, where the number of ratings is less than 4%. Product Document is the product side information representation. The document aims to advance the effectiveness of matrix factorization performance. This research considers to the enhancement of document context using LSTM with an attention mechanism to capture a contextual understanding of product review and incorporate matrix factorization based on probabilistic matrix factorization (PMF) to produce rating prediction. This study employs a real dataset using MovieLens dataset ML.1M and Amazon information video (AIV) to observe our ATT-PMF model. Movielens dataset represents of number sparse rating that only contains below 4% (ML.1M). Our experiment report shows that ATT-PMF outperforms more than 2% on average than previous work. Moreover, our model is also suitable to implement on huge datasets. For further research, enhancement of product document context will be a good factor in eliminating sparse data problems in big data problems
Designing and ERP System: A Sustainability Approach
Get PDFTechnology has grown up massively, which is numerous aspects. Whether the system is limited or open source, had to adapt to certain circumstances. For instance, decades ago, people mainly used to have a paper to work on reporting or to make notes. These days, people started to migrate the work stuff into digitalization. Furthermore, environmental pollution has been triggered by emissions from vehicles. Moreover, due to a lack of knowledge and resources to solve this issue, the manufacturing process does not apply the sustainability process. This matter has led to an impenetrable environmental issue in the manufacturing area for years. Therefore, developing a sales module dashboard system that is configured on Sales Order (SO) transactions and customer data could assist the problem-solving process. The reason for this matter is to highlight the Key Performance Indicator (KPI) at the sustainability that has been added by the open-source Enterprise Resources Planning (ERP) System and display the indicator of data visualization application. Furthermore, the development of the ERP system is aligned with the purpose of the Quickstart methodology that was applied in this research. To sum up, the sales module dashboard system is designed to assist the new user in this implementation in classifying the material, process, or products that require consideration to be maintained. Moreover, to reduce the number of raw materials or shipment processes that did not apply the eco-friendlines
Enhanced Technology for Logistics Courier Delivery Using RFID Label to Minimize Processing Time
Get PDFCourier services have become a sector that has experienced a growth spurt during the Covid-19 pandemic. The soaring growth of courier services is due to e-commerce in Indonesia. Increased people's digital activities show this during the pandemic, including online or online shopping. Data from the Indonesian Ministry of Finance shows that purchase transactions via e-commerce increased 18.1 percent to 98.3 million, with a total transaction value of 9.9 percent to Rp20.7 trillion. Fast and efficient delivery and pick-up of goods is the core operation of courier services. The biggest challenge for courier service providers is how to compete with other companies that offer the same type of service. Service users are increasingly demanding the security and reliability of delivery services so that they can meet the expectations of service users. The expectations of service users used as targets for company achievement are (1) reliability (on time, accuracy, integrity), (2) convenience (collecting units, delivery coverage, operating hours), (3) services, and (4) cost. Based on the activities in courier services, the potential for errors or inefficiencies in processing time is in the pre-delivery activities. In the pre-delivery activity is also the initial activity used to input the data base, collect goods, distribute goods and so on. This research proposes that RFID Label technology can overcome errors and increase process time efficiency in shipping goods on courier services, especially in pre-delivery and delivery activities
Confirmatory Factor Analysis: User Behavior M-Commerce Gamification Service in Indonesia
Get PDFGamification of marketing is a fast-growing phenomenon and an innovation for mobile marketing. Gamification is a strategy for increasing the attractiveness of mobile consumers to encourage increased shopping behavior, loyalty, engagement, and product advocacy. Understanding the factors behind the use of gamification services in m-commerce is very interesting, and no one has done any research. This study investigates the theory of self-determination (competence, autonomy, and relatedness) and extrinsic motivation as a predictor of the use of m-commerce gamification services. Data was collected from 400 respondents who had experienced using gamification services on m-commerce. The data was then included in the analysis. Analysis of the data using confirmatory factor analysis to determine the dominant factors of gamification service users. The benefits of factor analysis confirm the dominant factors that motivate users of gamification services in m-commerce. Researchers use AMOS 18 for Windows software to assist in the data processing. The results show that extrinsic motivation is the dominant factor that motivates users of gamification services. This finding provides insight for m-commerce companies and game designers to improve gamification mechanisms based on virtual points to motivate users to be more active and continue using gamification services. For the next research, it is possible to validate the construct by using other theoretical approaches, such as adding flow theory to measure the motivational factors of gamification service users. The research object can use other applications, such as gamification services in health, education, and banking applications
An Intelligent Missing Data Imputation Techniques: A Review
Get PDFThe incomplete dataset is an unescapable problem in data preprocessing that primarily machine learning algorithms could not employ to train the model. Various data imputation approaches were proposed and challenged each other to resolve this problem. These imputations were established to predict the most appropriate value using different machine learning algorithms with various concepts. Furthermore, accurate estimation of the imputation method is exceptionally critical for some datasets to complete the missing value, especially imputing datasets in medical data. The purpose of this paper is to express the power of the distinguished state-of-the-art benchmarks, which have included the K-nearest Neighbors Imputation (KNNImputer) method, Bayesian Principal Component Analysis (BPCA) Imputation method, Multiple Imputation by Center Equation (MICE) Imputation method, Multiple Imputation with denoising autoencoder neural network (MIDAS) method. These methods have contributed to the achievable resolution to optimize and evaluate the appropriate data points for imputing the missing value. We demonstrate the experiment with all these imputation techniques based on the same four datasets which are collected from the hospital. Both Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) are utilized to measure the outcome of implementation and compare with each other to prove an extremely robust and appropriate method that overcomes missing data problems. As a result of the experiment, the KNNImputer and MICE have performed better than BPCA and MIDAS imputation, and BPCA has performed better than the MIDAS algorithm
Cataract Classification Based on Fundus Images Using Convolutional Neural Network
Get PDFA cataract is a disease that attacks the eye's lens and makes it difficult to see. Cataracts can occur due to hydration of the lens (addition of fluid) or denaturation of proteins in the lens. Cataracts that are not treated properly can lead to blindness. Therefore, early detection needs to be done to provide appropriate treatment according to the level of cataracts experienced. In this study, a comparison of cataract classification based on fundus images using GoogleNet, MobileNet, ResNet, and the proposed Convolutional Neural Network was carried out. We compared four CNN architectures when implementing the Adam optimizer with a learning rate of 0.001. The data used are 399 datasets and augmented to 3200 data. This test's best and most stable results were obtained from the proposed CNN model with 92% accuracy, followed by MobileNet at 92%, ResNet at 93%, and GoogLeNet at 86%. We also make comparisons with previous research. Most of the previous studies only used two to three class categories. In this study, the system was improved by increasing system classifies into four categories: Normal, Immature, Mature, and Hypermature. In addition, the accuracy obtained is also quite good compared to previous studies using manual feature extraction. This study is expected to help medical staff to carry out early detection of cataracts to prevent the dangerous effect of cataracts and appropriate medical treatment. In the future, we want to expand the number of datasets to improve the classification accuracy of the cataract detection system
Determining Forest Fire Position from UAV Photogrammetry using Color Filtration Algorithm
Get PDFForest fires frequently happen worldwide, especially in the dry season. A forest fire early warning system (EWS) is needed to prevent this disaster. The main part of EWS is the hotspot detection system. On the other side, Unmanned Aerial Vehicle (UAV) technology offers an alternative solution to detect the hotspot for poor satellite image processing accuracy. Remote sensing techniques with UAV working drones are progressively challenging. Drones can provide results in 2D and 3D images with high resolution and real-time. Therefore, in this research, we have used a photogrammetry application from the number of images collected by a UAV with an optimum flight plan for the mission to determine the location of the forest fire. This paper describes remote sensing experiments using drones to detect land fires. The experiment was carried out using a quadcopter drone of the DJI Phantom 4 Pro. The photos are processed using Agisoft Metashape Professional image processing software and become a 2D image. These images captured a fire simulation in a known location. After a high resolution (GSD – Ground Sampling Distance – 0.87cm/px) orthophoto had been generated, a color filtration algorithm detected a hotspot to detect a fire at the exact location. The results are almost zero deviation of longitude and latitude from the real location with 1.44 m2 and 1.06 m2 fire area from 2 experiments. This algorithm program has TPR and FPR are 0,78 and 0, respectively. Further research can develop an EWS with a combination of UAV and Wireless Sensor Networks
Performance Evaluation of Successive Interference Cancellation on Gain Ratio Power Allocation using Underwater Visible Light Communication Channel
Get PDFUnderwater Visible light communication (UVLC) is a network communication wirelessly where information is transmitted employing light through visible waves; in this case, the light source comes from a light-emitting diode (LED) as a transmitter underwater. VLC has several advantages over radio frequency technology, such as safer communication because light propagation cannot penetrate the wall, so it is difficult to do hacking, easy to get a license, relatively build cheap cost, and has no side effects on health. However, VLC has several limitations, one of which is the narrow bandwidth modulation. VLC undergoes a distribution of modulated bandwidth to allocate against each user. This bandwidth sharing has an impact on reduced system capacity. This study applied non-orthogonal multiple access (NOMA) to increase system capacity. This research analyzes the performance of the two best power allocation methods in a water medium, including gain ratio power allocation (GRPA) and static power allocation (SPA). In the results obtained in the NOMA-UVLC power allocation value, GRPA is more stable than SPA power allocation. Then applying residue in the successive interference cancellation (SIC) process will result in a decrease in system capacity compared to no residue in the SIC process. This study found that the GRPA power allocation is more stable in capacity performance compared to the application of SPA power allocation. Average capacity increase of 48.5% in GRPA power allocatio
Batik Classification Using Convolutional Neural Network with Data Improvements
Get PDFBatik is one of the Indonesian cultures that UNESCO has recognized. Batik has a variety of unique and distinctive patterns that reflect the area of origin of the batik motif. Batik motifs usually have a 'core motif' printed repeatedly on the fabric. The entry of digitization makes batik motif designs more diverse and unique. However, with so many batik motifs spread on the internet, it is difficult for ordinary people to recognize the types of batik motifs. This makes an automatic classification of batik motifs must continue to be developed. Automation of batik motif classification can be assisted with artificial intelligence. Machine learning and deep learning have produced much good performance in image recognition. In this study, we use deep learning based on a Convolutional Neural Network (CNN) to automate the classification of batik motifs. There are two datasets used in this study. The old dataset comes from a public repository with 598 data with five types of motifs. Meanwhile, the new dataset updates the old dataset by replacing the anomalous data in the old dataset with 621 data with five types of motifs. The lereng motif is changed to pisanbali due to the difficulty of obtaining the lereng motif. Each dataset was divided into three ways: original, balance patch, and patch. We used ResNet-18 architecture, which used a pre-trained model to shorten the training time. The best test results were obtained in the new dataset with the patch way of 88.88 % ±0.88, and in the old dataset, the best accuracy was found in the patch way on the test data of 66.14 % ±3.7. The data augmentation in this study did not significantly affect the accuracy because the most significant increase in accuracy is only up to 1.22%
MLP-NARX Bitcoin Price Prediction Model Integrating System Identification Modelling Principles
Get PDFBitcoin is a decentralized digital currency that enables people to exchange value without requiring a third-party intermediary. Due to its many advantages, it has received much interest from institutional and individual investors. Despite its meteoric increase, the price of Bitcoin extremely volatile asset class as it purely relies on supply and demand. This presents an interesting opportunity to create a forecasting model. However, many research papers in this area does not analyse the residuals as part of the forecasting resulting in potentially biased models. In this paper, we demonstrate System Identification (SI) residual analysis techniques to the analysis of our forecasting model. The Multi-Layer Perceptron (MLP) Nonlinear Autoregressive with Exogeneous Inputs (NARX) uses historical price data and several technical indicators to predict the future price movements of Bitcoin. The Particle Swarm Optimization (PSO) algorithm was used to find optimal parameters for the model. The model was able to predict one day ahead price in the prediction test. The model has successfully captured the dynamics of the data through the tests performed on residuals. It is also proving the randomness of residuals, albeit some minor violations