The complexity and variety of language included in policy and academic documents make the automatic classification of research papers based on the United Nations Sustainable Development Goals (SDGs) somewhat difficult. Using both pre-trained and contextual word embeddings to increase semantic understanding, this study presents a complete deep learning pipeline combining Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Network (CNN) architectures which aims primarily to improve the comprehensibility and accuracy of SDG text classification, thereby enabling more effective policy monitoring and research evaluation. Successful document representation via Global Vector (GloVe), Bidirectional Encoder Representations from Tra

This paper focuses on developing a self-starting numerical approach that can be used for direct integration of higher-order initial value problems of Ordinary Differential Equations. The method is derived from power series approximation with the resulting equations discretized at the selected grid and off-grid points. The method is applied in a block-by-block approach as a numerical integrator of higher-order initial value problems. The basic properties of the block method are investigated to authenticate its performance and then implemented with some tested experiments to validate the accuracy and convergence of the method.

This study employs evolutionary optimization and Artificial Intelligence algorithms to determine an individual’s age using a single-faced image as the basis for the identification process. Additionally, we used the WIKI dataset, widely considered the most comprehensive collection of facial images to date, including descriptions of age and gender attributes. However, estimating age from facial images is a recent topic of study, even though much research has been undertaken on establishing chronological age from facial photographs. Retrained artificial neural networks are used for classification after applying reprocessing and optimization techniques to achieve this goal. It is possible that the difficulty of determining age could be reduce

Wind turbine (WT) is now a major renewable energy resource used in the modern world. One of the most significant technologies that use the wind speed (WS) to generate electric power is the horizontal-axis wind turbine. In order to enhance the output power over the rated WS, the blade pitch angle (BPA) is controlled and adjusted in WT. This paper proposes and compares three different controllers of BPA for a 500-kw WT. A PID controller (PIDC), a fuzzy logic controller (FLC) based on Mamdani and Sugeno fuzzy inference systems (FIS), and a hybrid fuzzy-PID controller (HFPIDC) have been applied and compared. Furthermore, Genetic Algorithm (GA) and Particle swarm optimization (PSO) have been applied and compared in order to identify the optimal

The blade pitch angle (BPA) controller is key factor to improve the power generation of wind turbine (WT). Due to the aerodynamic structural behavior of the rotor blades, wind turbine system performance is influenced by pitch angle and environmental conditions such as wind speed, which fluctuate throughout the day. Therefore, to overcome the pitch angle control (PAC) problem, high wind speed conditions, and due to type-1 and type-2 fuzzy logic limitations for handling high levels of uncertainty, the newly proposed optimal hybrid type-3 fuzzy logic controller has been applied and compared since type-3 fuzzy controllers utilize three-dimensional membership functions, unlike type-2 and type-1 fuzzy logic controllers. In this paper six differen

BP algorithm is the most widely used supervised training algorithms for multi-layered feedforward neural net works. However, BP takes long time to converge and quite sensitive to the initial weights of a network. In this paper, a modified cuckoo search algorithm is used to get the optimal set of initial weights that will be used by BP algorithm. And changing the value of BP learning rate to improve the error convergence. The performance of the proposed hybrid algorithm is compared with the stan dard BP using simple data sets. The simulation result show that the proposed algorithm has improved the BP training in terms of quick convergence of the solution depending on the slope of the error graph.