Objective: Flood events, occurring as natural and unexpected phenomena, have become increasingly prevalent in recent decades. Assessing the probability of flood risk creating zoning maps vulnerable areas, particularly urban regions, are crucial for mitigating damages managing such events. This research aims to determine flood-prone zones along Darakeh River, located Tehran, capital city Iran, by integrating hydraulic model HEC-RAS with Geographic Information System (GIS). Methods: After...

The literature provides strong evidence that stock price values can be predicted from past data. Principal component analysis (PCA) identifies a small number of principle components explain most the variation in data set. This method is often used for dimensionality reduction and In this paper, we develop general prediction using time-varying covariance information. To address nature financial time series, assign exponential weights to so recent points are weighted more heavily. Our proposed...

Graph embedding is an important approach for graph analysis tasks such as node classification and link prediction. The goal of to find a low dimensional representation nodes that preserves the information. Recent methods like Convolutional Network (GCN) try consider attributes (if available) besides relations learn embeddings unsupervised semi-supervised on graphs. On other hand, multi-layer has been received attention recently. However, existing cannot incorporate all available information...

<p><strong><em> </em></strong>Increased concern for the environment has led to search more environment-friendly energy sources so that wind can be used as an endless option human consumption. Wind turbines offer a promising solution off-grid areas. However, they have certain drawbacks associated with different configurations. Darrieus turbine is one type efficient than other types. The poor start-up performance of critical problems restricting its development....

Adversarial approach has been widely used for data generation in the last few years. However, this not extensively utilized classifier training. In paper, we propose an adversarial framework training that can also handle imbalanced data. Indeed, a network is trained via to give weights samples of majority class such obtained classification problem becomes more challenging discriminator and thus boosts its capability. addition general problems, proposed method be problems as graph...

Detecting community structures in social networks has gained considerable attention recent years. However, lack of prior knowledge about the number communities, and their overlapping nature have made detection a challenging problem. Moreover, many existing methods only consider static networks, while most real world are dynamic evolve over time. Hence, finding consistent communities without any is still an interesting open research In this paper, we present method for called Dynamic Bayesian...

Abstract Instruction in ethical considerations is an important part of every engineering discipline. In many cases, a student’s exposure to issues delayed until the capstone senior design experience. For example, we have included lectures devoted ethics our Electrical and Computer Engineering program that start with introduction National Society Professional Engineers (NSPE) Institute Electronics (IEEE) codes ethics, then followed by discussion various case studies. While this common...

The literature provides strong evidence that stock prices can be predicted from past price data. Principal component analysis (PCA) is a widely used mathematical technique for dimensionality reduction and of data by identifying small number principal components to explain the variation found in set. In this paper, we describe general method prediction using covariance information, terms dimension operation based on principle analysis. Projecting noisy observation onto subspace leads...

We present five methods for probabilistic short-term load forecasting, namely, Bayesian estimation, a rank-reduction method based on principal component analysis, least absolute shrinkage and selection operator (Lasso) ridge regression, supervised learning approach called scaled conjugate gradient neural network. aim to incorporate the temperature effects directly in these reflect hourly patterns of electrical demand. provide empirical results data sets used Global Energy Forecasting...

The need for a large amount of labeled data in the supervised setting has led recent studies to utilize self-supervised learning pre-train deep neural networks using unlabeled data. Many training strategies have been investigated especially medical datasets leverage information available much fewer One fundamental image-based self-supervision is context prediction. In this approach, model trained reconstruct contents an arbitrary missing region image based on its surroundings. However,...