As climates change globally, water-related disasters increase, causing substantial economic losses and safety risks. During floods, river water levels show unpredictable fluctuations, introducing substantial noise that complicates accurate prediction. A hybrid model that uses eight-dimensional input data from hydrological and meteorological stations is proposed to address these challenges. Initially, the Variational Mode Decomposition preprocesses and denoises water level data, resulting in decomposed Intrinsic Mode Functions (IMFs). Then, the Pearson correlation coefficient between each IMF and input characteristics is computed, and the fluctuation factor for each IMF is defined. IMFs are categorized based on a threshold, leading to a hybrid prediction model. This model integrates convolutional neural networks (CNNs) for spatial information and bidirectional long short-term memory (BiLSTM) networks with an attention mechanism for learning from past and future data points. Comparative evaluations of mean absolute percentage error, root mean square error, mean absolute error, and goodness of fit (R2) show that the proposed model outperforms existing LSTM and CNN–BiLSTM frameworks, reducing RMSE by at least 20% and increasing R2 by approximately 10% on average. The model's practical significance lies in improving the accuracy and efficiency of meteorological forecasting and flood warning systems, contributing substantially to global disaster preparedness and response strategies.

Real life logical rule is not always satisfiable in nature due to the redundant variable that represents the logical formulation. Thus, the intelligence system must be optimally governed to ensure the system can behave according to non-satisfiable structure that finds practical applications particularly in knowledge discovery tasks. In this paper, we a propose non-satisfiability logical rule that combines two sub-logical rules, namely Maximum 2 Satisfiability and Random 2 Satisfiability, that play a vital role in creating explainable artificial intelligence. Interestingly, the combination will result in the negative logical outcome where the cost function of the proposed logic is always more than zero. The proposed logical rule is implemented into Discrete Hopfield Neural Network by computing the cost function associated with each variable in Random 2 Satisfiability. Since the proposed logical rule is difficult to be optimized during training phase of DHNN, Election Algorithm is implemented to find consistent interpretation that minimizes the cost function of the proposed logical rule. Election Algorithm has become the most popular optimization metaheuristic technique for resolving constraint optimization problems. The fundamental concepts of Election Algorithm are taken from socio-political phenomena which use new and efficient processes to produce the best outcome. The behavior of Random Maximum 2 Satisfiability in Discrete Hopfield Neural Network is investigated based on several performance metrics. The performance is compared between existing conventional methods with Genetic Algorithm and Election Algorithm. The results demonstrate that the proposed Random Maximum 2 Satisfiability can become the symbolic instruction in Discrete Hopfield Neural Network where Election Algorithm has performed as an effective training process of Discrete Hopfield Neural Network compared to Genetic Algorithm and Exhaustive Search.

Flood disasters occur worldwide, and flood risk prediction is conducive to protecting human life and property safety. Influenced by topographic changes and rainfall, the water level fluctuates randomly and violently during the flood, introducing many noises and directly increasing the difficulty of flood prediction. A data-driven flood forecasting method is proposed based on data preprocessing and a two-layer BiLSTM-Attention network to improve forecast accuracy. First, the Variational Mode Decomposition (VMD) is used to decompose the data for reducing noise and produce suitable Intrinsic Mode Functions (IMFs); Then, an optimized two-layer attention-based Bidirectional Long Sshort-Term memory (BiLSTM-Attention) network is constructed to predict each IMF. Finally, two optimization algorithms are used to obtain the optimized parameters of VMD and BiLSTM intelligently, increasing the self-adaptability. The inertia factor of particle swarm optimization is improved and then used to optimize the five hyperparameters of BiLSTM. The proposed model reduces storage errors for smaller training sets and can achieve good performance. Three water level data sets from the Yangtze River in China are used for comparative experiments. Numerical results show that the peak height absolute error is within 2 cm, and the relative error of peak time arrival is within 30%. Compared with LSTM, BiLSTM, CNN-BiLSTM-attention, etc., the proposed model reduces the root mean square error by at least 50% and has advantages for high-risk forecasting when the water level exceeds the defense line and fluctuates prominently.

As global climates undergo changes, the frequency of water-related disasters rises, leading to significant economic losses and safety hazards. During flood events, river water levels exhibit unpredictable fluctuations, introducing considerable noise that poses challenges for accurate prediction. A prediction of water level by using existing water level data makes a major contribution to forecasting flood. Enhanced least-squares support vector machine (ELSSVM) is utilized by integrating an additional extra bias error control term. In this study, least-squares support vector machine (LSSVM) and ELSSVM optimized by the genetic algorithm (GA) were chosen to be compared with the help of data decomposition methods to improve daily water level prediction accuracy. Double empirical mode decomposition (DEMD) will be integrated with LSSVM and ELSSVM. Thus, the models are named LSSVM-GA, ELSSVM-GA, empirical mode decomposition (EMD)-LSSVM-GA, EMD-ELSSVM-GA, DEMD-LSSVM-GA, and DEMD-ELSSVM-GA. The proposed models are used in forecasting the water level of Klang River in Sri Muda, Malaysia. The behavior proposed models are investigated and compared based on several performance metrics. The results demonstrated that the DEMD-ELSSVM-GA model outperformed the other models based on the performance analysis in forecasting the water level with RMSE = 0.2536 m and R2 = 0.8596 for testing data that indicate the forecasting accuracy.

This paper presents investigation on the efficient market hypothesis of extreme stock return based on peaks over threshold method, by application of 10% Value at risk (VaR) quantile threshold level. The efficient market hypothesis (EMH) in the stock market index is validated by utilising autocorrelation, Kwiatkowski Phillips Schmidt Shin (KPSS), and variance ratio tests. The tests constituted of daily, extreme maximum and minimum, and three sub-periods data reflecting different economic condition in the market. Results indicated the strong influences of the financial crisis in the series movement. Mixed evidences were acquired; nonetheless, the overall results show that the Malaysia extreme stock return does not follow a random walk, only the series during the crisis and recovery period are in weak-form market efficiency.