Article
Computer Science, Artificial Intelligence
Sourav Kumar Purohit, Sibarama Panigrahi, Prabira Kumar Sethy, Santi Kumari Behera
Summary: Accurate prediction of crop prices is crucial for farmers and government. This study proposes hybrid methods to predict the prices of three commonly used vegetable crops in India, showing superiority over statistical models and machine learning models through extensive statistical analyses.
APPLIED ARTIFICIAL INTELLIGENCE
(2021)
Article
Transportation Science & Technology
V. Rajalakshmi, S. Ganesh Vaidyanathan
Summary: Traffic flow forecast is critical for constructing traffic plans and reducing congestion on roadways. This research uses time-series forecasting models to estimate future traffic but minimizing prediction error is challenging. Hybrid ARIMA-MLP and ARIMA-RNN models are proposed to anticipate future traffic flow using real-time data from vehicles and roadways, using UK Highways dataset for validation.
PROMET-TRAFFIC & TRANSPORTATION
(2022)
Article
Computer Science, Artificial Intelligence
Igor Ilic, Berk Gorgulu, Mucahit Cevik, Mustafa Gokce Baydogan
Summary: Time series forecasting involves developing a model based on past observations to predict future events. The explainable boosted linear regression (EBLR) algorithm enhances predictions by explaining errors and incorporating nonlinear features. It provides interpretable results and high predictive accuracy, making it a promising method for time series forecasting.
PATTERN RECOGNITION
(2021)
Article
Green & Sustainable Science & Technology
Cyril Voyant, Gilles Notton, Jean-Laurent Duchaud, Luis Antonio Garcia Gutierrez, Jamie M. Bright, Dazhi Yang
Summary: With the increasing share of intermittent renewable energy, advanced solar power forecasting models are needed to optimize the operation of solar power plants. This study compares the performance of advanced models with naive reference methods and considers the benefits of ensemble forecasting. The combination method and ARTU method statistically offer the best results for the proposed study conditions.
Article
Engineering, Chemical
Mahmod Othman, Rachmah Indawati, Ahmad Abubakar Suleiman, Mochammad Bagus Qomaruddin, Rajalingam Sokkalingam
Summary: This study used time-series methods to forecast the incidence of Dengue Hemorrhagic Fever (DHF) in Surabaya City. By comparing different models, the seasonal ARIMA (SARIMA) model was found to be the most accurate forecasting method, and future outbreaks were predicted. The results showed significant seasonal outbreaks of DHF, with a significant correlation between DHF and air temperature.
Article
Mathematics
Oksana Mandrikova, Nadezhda Fetisova, Yuriy Polozov
Summary: The hybrid model for time series of complex structure (HMTS) combines wavelet series with ARIMA models to provide a more accurate modeling for time series with complicated structures. The identification of HMTS anomalous components using threshold functions and the detection of ionospheric anomalies have shown the efficiency of HMTS. Comparing HMTS with the NARX neural network confirms the effectiveness of HMTS.
Article
Computer Science, Information Systems
Ting Guo, Feng Hou, Yan Pang, Xiaoyun Jia, Zhongwei Wang, Ruili Wang
Summary: This paper categorizes recent methods for multivariate time series forecasting into graph-based approach and global-local approach. It points out the limitations of both approaches and proposes a new approach called Adaptive Global-Local Graph Structure Learning with Gated Recurrent Units (AGLG-GRU) to combine the advantages of both methods. Experimental results on real-world datasets demonstrate the superiority of the proposed approach.
INFORMATION SCIENCES
(2023)
Article
Green & Sustainable Science & Technology
Ashutosh Kumar Dubey, Abhishek Kumar, Vicente Garcia-Diaz, Arpit Kumar Sharma, Kishan Kanhaiya
Summary: Energy consumption forecasting using ARIMA, SARIMA, and LSTM models based on smart meter measurements showed that humidity has a high positive correlation with energy consumption, while temperature has a high negative correlation. LSTM outperformed ARIMA and SARIMA with a mean absolute error (MAE) of 0.23.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2021)
Article
Computer Science, Artificial Intelligence
Andreas Kanavos, Fotios Kounelis, Lazaros Iliadis, Christos Makris
Summary: This paper focuses on the analysis and modeling of passenger demand dynamics in the aviation industry, proposing a method using time series and deep learning techniques to forecast aviation demand, and developing related models. The results of the study show that the proposed methods exhibit satisfactory accuracy and robustness in predicting air travel demand.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Physics, Multidisciplinary
Zahra Hajirahimi, Mehdi Khashei
Summary: This study proposes a hybrid model named SHOP, which integrates a parallel hybrid model using a series hybridization approach, to improve forecasting accuracy and overcome the drawbacks of parallel models.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Borui Cai, Shuiqiao Yang, Longxiang Gao, Yong Xiang
Summary: This paper introduces a novel hybrid variational autoencoder (HyVAE) for forecasting time series by jointly learning the local patterns and temporal dynamics. Experimental results demonstrate that the proposed HyVAE achieves better results compared to other methods.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Biomedical
Nasrin Talkhi, Narges Akhavan Fatemi, Zahra Ataei, Mehdi Jabbari Nooghabi
Summary: The study identified the MLP network model as suitable for predicting confirmed cases in Iran, while the Holt-Winter model is suitable for forecasting future death cases. According to the data trend and forecast results, the number of confirmed cases is almost constant, while death cases are decreasing.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Zhangjing Yang, Wei-Wu Yan, Xiaolin Huang, Lin Mei
Summary: This paper proposes a novel adaptive temporal-frequency network (ATFN) for mid- and long-term time series forecasting. The model combines deep learning networks and frequency patterns to learn the trend feature and capture dynamic periodic patterns of time series data. The experimental results demonstrate that the ATFN has promising performance and strong adaptability for long-term time-series forecasting.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Statistics & Probability
Dileep Kumar Shetty, B. Ismail
Summary: In this study, a hybrid non-stationary model based on Elman's Recurrent Neural Networks was developed to forecast stock market price indices. The model is capable of capturing both linear and non-linear structures, and it exhibited the best forecasting accuracy when applied to real datasets.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2023)
Article
Computer Science, Information Systems
Ilias Kalouptsoglou, Dimitrios Tsoukalas, Miltiadis Siavvas, Dionysios Kehagias, Alexander Chatzigeorgiou, Apostolos Ampatzoglou
Summary: Software security is crucial for modern software products. This paper predicts the evolution of vulnerabilities in the future using statistical and deep learning models, based on security data from popular software projects. The results demonstrate that the performance of these models depends on the specific software project, with no significant difference in their accuracy.