Related references
Note: Only part of the references are listed.
Article
Mathematics
Shahzad Zaheer et al.
Summary: Financial data, particularly time series data, offer valuable information for data analysis, especially in forecasting stock prices, which remains a challenging task due to noise, non-linearity, and volatility. Previous studies primarily focused on a single stock parameter, such as the close price. This study presents a hybrid deep-learning forecasting model that predicts both the close and high prices of stocks for the next day based on input stock data. Experimental results on the Shanghai Composite Index (000001) demonstrate the superiority of a suggested single Layer RNN model compared to existing methods, with an improvement of 2.2%, 0.4%, 0.3%, 0.2%, and 0.1%. These findings validate the effectiveness of the proposed model in assisting investors to make profitable decisions.
Review
Business, Finance
Htet Htet Htun et al.
Summary: The identification of critical features is crucial in stock market forecasting for accurate predictions. This survey analyzes 32 research works that combine feature study and ML approaches in various stock market applications. The most widely used feature selection and extraction techniques for accurate predictions include correlation criteria, random forest, principal component analysis, and autoencoder.
FINANCIAL INNOVATION
(2023)
Article
Engineering, Multidisciplinary
Metin Ifraz et al.
Summary: This study conducted demand forecasting for spare parts of vehicle fleets using regression-based, rule-based, tree-based methods, and artificial neural networks. It was observed that the artificial neural network provided the most accurate forecasts with the highest forecast accuracy rate.
JOURNAL OF ENGINEERING RESEARCH
(2023)
Article
Environmental Sciences
Eatedal Alabdulkreem et al.
Summary: Groundwater is a crucial water resource and its management is essential for various purposes such as human life, production, irrigation, and economic development. This research aims to predict groundwater using the spatiotemporal attention mechanism. Due to urbanization, population growth, and industrialization, the vulnerability of groundwater depletion has increased, necessitating effective management in terms of quality and quantity. This paper proposes a stacked LSTM with a deep neural network (SLSTM-DNN) approach for managing groundwater and predicting demand, achieving an accuracy rate of 91.45%, higher than existing methods such as CNN, DNN, and LSTM (81.32%, 82.34%, and 88.12% respectively).
Article
Business, Finance
Eka Nurhalimatus Sifa et al.
Summary: This study aims to simulate and compare the effect of two financing schemes, Salam and conventional financing, on farmers' cash flows. The findings suggest that the Salam scheme provides a higher income that can contribute to improving farmer welfare, and it also requires less adjustment to meet the farmers' needs.
JOURNAL OF ISLAMIC ACCOUNTING AND BUSINESS RESEARCH
(2023)
Article
Green & Sustainable Science & Technology
Chen-Yu Tai et al.
Summary: Adequate data is essential for accurate trend prediction. This study utilizes TimeGAN to generate agricultural sensing data and train neural network models to predict future pest populations. The results show that the generated data effectively compensates for the lack of actual data and produces similar predictions to those trained on actual data. Accurate prediction of pest populations would be a breakthrough in pest control.
Article
Computer Science, Artificial Intelligence
Madini O. Alassafi et al.
Summary: The study developed a prediction model for the spread of COVID-19 in Malaysia, Morocco, and Saudi Arabia using public datasets from the European Centre for Disease Prevention and Control. Deep learning models were utilized with a focus on LSTM networks. The study also compared the number of cases and deaths in the three countries.
Article
Computer Science, Information Systems
Lkhagvadorj Munkhdalai et al.
Summary: This study proposes a novel locally adaptive interpretable deep learning architecture augmented by recurrent neural networks, which can provide model explainability and high predictive accuracy for time-series data. It shows better predictive performance than the state-of-the-art baselines and discovers the dynamic relationship between input and output variables.
Article
Mathematics, Applied
Jong-Min Kim et al.
Summary: This paper introduces methodologies for forecasting oil prices (Brent and WTI) using multivariate time series of major S&P 500 stock prices with Gaussian process modeling, deep learning, and vine copula regression. Bayesian variable selection and nonlinear principal component analysis (NLPCA) are also applied for data dimension reduction. The results suggest that vine copula regression with NLPCA outperforms other proposed methods in terms of prediction error measures.
Article
Computer Science, Information Systems
Srivinay et al.
Summary: Stock prices are volatile due to various factors and accurately predicting them can help reduce investment risk. The authors proposed a hybrid stock prediction model using the prediction rule ensembles (PRE) technique and deep neural network (DNN). This model showed better performance in estimating stock volatility compared to single prediction models such as DNN and ANN.
Article
Multidisciplinary Sciences
Akib Mohi Ud Din Khanday et al.
Summary: Social networks have become a dominant platform for information dissemination, but are also being used to manipulate emotions and spread misinformation. This research utilized machine learning algorithms to classify propagandist text from non-propagandist text, with SVM outperforming MNB in accuracy.
BAGHDAD SCIENCE JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Hui Liu et al.
DIGITAL SIGNAL PROCESSING
(2020)
Article
Statistics & Probability
Igor G. Zurbenko et al.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL STATISTICS
(2018)
