Yaya Heryadi

Bio: Yaya Heryadi is an academic researcher from Binus University. The author has contributed to research in topics: Deep learning & Weather forecasting. The author has an hindex of 10, co-authored 74 publications receiving 451 citations. Previous affiliations of Yaya Heryadi include International University, Cambodia.

Single Layer & Multi-layer Long Short-Term Memory (LSTM) Model with Intermediate Variables for Weather Forecasting

Abstract: Weather forecasting has gained attention many researchers from various research communities due to its effect to the global human life. The emerging deep learning techniques in the last decade coupled and the wide availability of massive weather observation data have motivated many researches to explore hidden hierarchical pattern in the large volume of weather dataset for weather forecasting. The purposes of this research are to build a robust and adaptive statistical model for forecasting univariate weather variable in Indonesian airport area and to explore the effect of intermediate weather variable related to accuracy prediction using single layer Long Short Memory Model (LSTM) model and multi layers LSTM model. The proposed forecasting model is an extension of LSTM model by adding intermediate variable signal into LSTM memory block. The premise is that two highly related patterns in input dataset will rectify the input patterns so make it easier for the model to learn and recognize the pattern from the training dataset. In an effort to achieve a robust model for learning and recognizing weather pattern, this research will also explore various architectures such as single layer LSTM and Multiple Layer LSTM (4 layers LSTM). The dataset is weather variable data collected by Weather Underground at Hang Nadim Indonesia airport. This research used visibility as predicted data and temperature, pressure, humidity, dew point as intermediates data. The best model of LSTM in this experiment is multiple layers LSTM and the best intermediate data is pressure variable. Using the pressure variable this model has gained the validation accuracy 0.8060 and RMSE 0.0775.

Weather forecasting using deep learning techniques

Abstract: Weather forecasting has gained attention many researchers from various research communities due to its effect to the global human life. The emerging deep learning techniques in the last decade coupled with the wide availability of massive weather observation data and the advent of information and computer technology have motivated many researches to explore hidden hierarchical pattern in the large volume of weather dataset for weather forecasting. This study investigates deep learning techniques for weather forecasting. In particular, this study will compare prediction performance of Recurrence Neural Network (RNN), Conditional Restricted Boltzmann Machine (CRBM), and Convolutional Network (CN) models. Those models are tested using weather dataset provided by BMKG (Indonesian Agency for Meteorology, Climatology, and Geophysics) which are collected from a number of weather stations in Aceh area from 1973 to 2009 and El-Nino Southern Oscilation (ENSO) data set provided by International Institution such as National Weather Service Center for Environmental Prediction Climate (NOAA). Forecasting accuracy of each model is evaluated using Frobenius norm. The result of this study expected to contribute to weather forecasting for wide application domains including flight navigation to agriculture and tourism.

Learning temporal representation of transaction amount for fraudulent transaction recognition using CNN, Stacked LSTM, and CNN-LSTM

Abstract: This paper aims to explore deep learning model to learn short-term and long-term patterns from imbalanced input dataset. Data for this study are imbalanced card transactions from an Indonesia bank in period 2016–2017 with binary labels (nonfraud or fraud). From 50 features of the dataset, 30 principal components of data contribute to 87 % of the cumulative Eigenvalues. This study explores the effect of nonfraud to fraud sample ratio from 1 to 4 and three models: Convolutional Neural Network (CNN), Stacked Long Short-term Memory (SLSTM), and Hybrid of CNN-LSTM. Using Area Under the ROC Curve (AUC) as model performance, CNN achieved the highest AUC for R=1,2,3,4 followed by SLSTM and CNN-LSTM.

American sign language-based finger-spelling recognition using k-Nearest Neighbors classifier

Abstract: This paper presents finger-spelling recognition method for American Sign Language (ASL) Alphabet using k-Nearest Neighbours (k-NN) Classifier. This research also examines the effect of PCA for dimensional reduction to k-NN performance. The empiric results show that k-NN classifier achieves the highest accuracy (99.8 percent) for k=3 when the pattern is represented by full dimensional feature. However, k-NN classifier only achieves 28.6 percent accuracy (for k=5) when the pattern is represented by PCAreduced dimensional feature. This low accuracy is due to several factors, among others, is the presence of high numbers of redundant or highly correlated features among ASL alphabet that makes PCA unable to separate data. Although kNN classifier accuracy is higher than the proposed classifier in [7], recognition time of k-NN classifier is longer than that of the method proposed in [7]. Therefore, k-NN classifier is suitable for early child education-based application such as self-assessment system for special need student who learns ASL alphabet finger-spelling.

The Impact of the COVID-19 Pandemic in Indonesia (Face to face versus Online Learning)

Abstract: Covid-19 pandemic is an international disaster that is experienced by almost all countries in the world. This has an impact on all lines of the life of each country. Among them is the education sector. Aside from efforts to solve this co-19 problem, the state must continue to maintain the stability and sustainability of the learning process that is the right of all citizens. Indonesia experienced the same thing. face to face learning “shock” and immediately take the fastest action by utilizing existing technology, but not all of them are ready. Using survey methods and adopting the theories of Delon and McLean, this study aims to determine the readiness of organizers, lectures, and students for current conditions, their readiness in undergoing the learning process while maintaining the quality of education and user satisfaction (instructors and students) towards learning. The results of this study prove that we all tend to be unprepared but strangely, on the other hand, the fact is that the positive things from this pandemic prove that education practitioners in Indonesia are better prepared by online learning because they are more comfortable and satisfied with online learning while supported by the government and a good system (96% of respondents) compared to face to face (4% of respondents).

Flow The Psychology Of Optimal Experience

Abstract: Thank you very much for downloading flow the psychology of optimal experience. As you may know, people have search numerous times for their chosen readings like this flow the psychology of optimal experience, but end up in infectious downloads. Rather than reading a good book with a cup of tea in the afternoon, instead they juggled with some harmful bugs inside their desktop computer. flow the psychology of optimal experience is available in our digital library an online access to it is set as public so you can get it instantly. Our digital library saves in multiple countries, allowing you to get the most less latency time to download any of our books like this one. Merely said, the flow the psychology of optimal experience is universally compatible with any devices to read.

Deep Learning for Anomaly Detection: A Survey.

Abstract: Anomaly detection is an important problem that has been well-studied within diverse research areas and application domains. The aim of this survey is two-fold, firstly we present a structured and comprehensive overview of research methods in deep learning-based anomaly detection. Furthermore, we review the adoption of these methods for anomaly across various application domains and assess their effectiveness. We have grouped state-of-the-art research techniques into different categories based on the underlying assumptions and approach adopted. Within each category we outline the basic anomaly detection technique, along with its variants and present key assumptions, to differentiate between normal and anomalous behavior. For each category, we present we also present the advantages and limitations and discuss the computational complexity of the techniques in real application domains. Finally, we outline open issues in research and challenges faced while adopting these techniques.

Sentiment analysis using deep learning architectures: a review

Abstract: Social media is a powerful source of communication among people to share their sentiments in the form of opinions and views about any topic or article, which results in an enormous amount of unstructured information. Business organizations need to process and study these sentiments to investigate data and to gain business insights. Hence, to analyze these sentiments, various machine learning, and natural language processing-based approaches have been used in the past. However, deep learning-based methods are becoming very popular due to their high performance in recent times. This paper provides a detailed survey of popular deep learning models that are increasingly applied in sentiment analysis. We present a taxonomy of sentiment analysis and discuss the implications of popular deep learning architectures. The key contributions of various researchers are highlighted with the prime focus on deep learning approaches. The crucial sentiment analysis tasks are presented, and multiple languages are identified on which sentiment analysis is done. The survey also summarizes the popular datasets, key features of the datasets, deep learning model applied on them, accuracy obtained from them, and the comparison of various deep learning models. The primary purpose of this survey is to highlight the power of deep learning architectures for solving sentiment analysis problems.

Change-Point Detection in Time-Series Data by Relative Density-Ratio Estimation (情報論的学習理論と機械学習)

Abstract: The objective of change-point detection is to discover abrupt property changes lying behind time-series data. In this paper, we present a novel statistical change-point detection algorithm based on non-parametric divergence estimation between time-series samples from two retrospective segments. Our method uses the relative Pearson divergence as a divergence measure, and it is accurately and efficiently estimated by a method of direct density-ratio estimation. Through experiments on artificial and real-world datasets including human-activity sensing, speech, and Twitter messages, we demonstrate the usefulness of the proposed method.