Social ski driver conditional autoregressive-based deep learning classifier for flight delay prediction

Abstract

The importance of robust flight delay prediction has recently increased in the air transportation industry. This industry seeks alternative methods and technologies for more robust flight delay prediction because of its significance for all stakeholders. The most affected are airlines that suffer from monetary and passenger loyalty losses. Several studies have attempted to analysed and solve flight delay prediction problems using machine learning methods. This research proposes a novel alternative method, namely social ski driver conditional autoregressive-based (SSDCA-based) deep learning. Our proposed method combines the Social Ski Driver algorithm with Conditional Autoregressive Value at Risk by Regression Quantiles. We consider the most relevant instances from the training dataset, which are the delayed flights. We applied data transformation to stabilise the data variance using Yeo-Johnson. We then perform the training and testing of our data using deep recurrent neural network (DRNN) and SSDCA-based algorithms. The SSDCA-based optimisation algorithm helped us choose the right network architecture with better accuracy and less error than the existing literature. The results of our proposed SSDCA-based method and existing benchmark methods were compared. The efficiency and computational time of our proposed method are compared against the existing benchmark methods. The SSDCA-based DRNN provides a more accurate flight delay prediction with 0.9361 and 0.9252 accuracy rates on both dataset-1 and dataset-2, respectively. To show the reliability of our method, we compared it with other meta-heuristic approaches. The result is that the SSDCA-based DRNN outperformed all existing benchmark methods tested in our experiment.

Enhancement of available transfer capability usingTCSCwith hybridized model: Combining lion and moth flame algorithms

“…The equivalent circuit of TCSC is shown in Figure 2. The TCSC functioning can be understood by means of analyzing the behavior of a variable inductor L that is connected in parallel with the capacitor C. Further, the mathematical expression for computing the equivalent impedance Z eq of this LC combinational circuit is depicted in Equation (1).…”

“…The power system is a compound electrical network, which is extremely bulky that includes generation networks, transmission networks and distribution networks along with loads and these networks are being stretched all over the huge geographical zone. 1,2 Recently, the demand for electric power escalates in an exponential mode and hence ceaseless efforts are being implied by the power system industries in distribution, transmission, and generation of electric power. [3][4][5] These industries keep promoting their techniques to sustain the structure of the competitive electric power markets.…”

Self‐improved moth flame for optimal container resource allocation in cloud

“…In contrast to VMs, container‐oriented virtualization allows an existing operating system to offer further isolation 14–17 . Thus, a container simplifies the exploitation of cloud appliances and satisfies the most important needs of cloud, that is, density and elasticity 7,18–21 …”

“…[14][15][16][17] Thus, a container simplifies the exploitation of cloud appliances and satisfies the most important needs of cloud, that is, density and elasticity. 7,[18][19][20][21] In spite of the speedy growth in technologies, there are certain problems in managing and developing MS in the cloud. 22,23 The allocation of container resources in the cloud is an NP-hard issue and it should be resolved using genetic algorithm, reinforcement learning approach, chemical reaction optimization algorithm, and reinforcement learning-based microservice allocation (RL-MA), improved particle swarm optimization based quantum evolutionary algorithm (IPOQEA).…”

Optimization assisted convolutional neural network for detection of thyroid

“…Recently, the deep learning (DL) field has been grown rapidly in different visual recognition tasks as the CNNs 20‐23 have its additive performance 24 . Further, the merits of CNN include the hierarchical abstractions through end‐to‐end training of image data, and the ability to learn high‐level automatically.…”