Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
56
result(s) for
"Computational intelligence Congresses."
Sort by:
Performance analysis of stirling engine using computational intelligence techniques (ANN & Fuzzy Mamdani Model) and hybrid algorithms (ANN-PSO & ANFIS)
Stirling engine is considered as one of the most promising alternatives to conventional combustion units due to its versatility and potential to achieve relatively high efficiency. The output power and torque are the main performance indicators that depend on many variables. Many studies have pointed out that the relationship between the performance indicators of the Stirling engine and its input variables was nonlinear. This study analyses the prediction performance of power and torque in a Stirling engine system using soft computing techniques—artificial neural network (ANN) and Fuzzy Mamdani Model (FMM) and hybrid algorithms—adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network trained with particle swarm optimization (ANN-PSO). The performance of these approaches has been discussed using a dataset from a test conducted on an existing Stirling engine. The performance indicators of the different models considering the power and the torque were predicted and analysed. A parametric analysis has been performed for the ANN-PSO model to identify the best model configuration considering the number of neurons in hidden layers, the number of swarm size and acceleration factors. A detailed description of the process leading to the identification of the best networks architecture for the power and torque model has been provided. The comparison of the four approaches indicates that FMM exhibits the highest performance prediction considering the power while the ANN-PSO and ANFIS model exhibit the highest performance considering the torque. This study demonstrates the suitability of soft computing techniques and hybrid algorithms for the prediction of Stirling engine characteristics and its potential to optimize time and experimental cost.
Journal Article
Evolutionary algorithm applied to time-series landing flight path and control optimization of supersonic transport
An evolutionary algorithm (EA) was applied in this study to optimize the landing flight path of a delta-winged supersonic transport (SST). However, it is difficult for a delta wing with a large sweepback angle to reduce the aerodynamic drag during supersonic cruising to gain sufficient lift force at low speeds, particularly during takeoff and landing. Besides, high-fidelity computational fluid dynamics is required to evaluate the flight path with a complex flowfield. This study performed an efficient flight simulation based on the Kriging model-assisted aerodynamic estimation to carry out global optimization. Then, the designs of the flight and control sequence were realized for time-series optimization of effective SST landing. To develop the EA, two design scenarios were considered; one involved only the elevator, which is an aerodynamic control surface that controls the aircraft, and the other involved introducing thrust control in addition to elevator control. In the scenario involving only elevator control, feasible solutions could not be obtained owing to the poor low-speed aerodynamic performance of the SST. This paper presents several feasible solutions enabling reasonable SST landing performance in the scenario involving the elevator and thrust controls along with descriptions regarding the optimum flight and control sequences. In addition, we analyzed the solutions by analyzing the variance to obtain qualitative information. Consequently, we determined that elevator control was considerably effective in cases with the microburst effect than in cases without the microburst effect.
Journal Article
Verbal aggression detection on Twitter comments: convolutional neural network for short-text sentiment analysis
Cyberbullying and hate speeches are common issues in online etiquette. To tackle this highly concerned problem, we propose a text classification model based on convolutional neural networks for the de facto verbal aggression dataset built in our previous work and observe significant improvement, thanks to the proposed 2D TF-IDF features instead of pre-trained methods. Experiments are conducted to demonstrate that the proposed system outperforms our previous methods and other existing methods. A case study of word vectors is carried out to address the difficulty in using pre-trained word vectors for our short-text classification task, demonstrating the necessities of introducing 2D TF-IDF features. Furthermore, we also conduct visual analysis on the convolutional and pooling layers of the convolutional neural networks trained.
Journal Article
Selection of optimal wavelet features for epileptic EEG signal classification with LSTM
Epilepsy remains one of the most common chronic neurological disorders; hence, there is a need to further investigate various models for automatic detection of seizure activity. An effective detection model can be achieved by minimizing the complexity of the model in terms of trainable parameters while still maintaining high accuracy. One way to achieve this is to select the minimum possible number of features. In this paper, we propose a long short-term memory (LSTM) network for the classification of epileptic EEG signals. Discrete wavelet transform (DWT) is employed to remove noise and extract 20 eigenvalue features. The optimal features were then identified using correlation and
P
value analysis. The proposed method significantly reduces the number of trainable LSTM parameters required to attain high accuracy. Finally, our model outperforms other proposed frameworks, including popular classifiers such as logistic regression (LR), support vector machine (SVM), K-nearest neighbor (K-NN) and decision tree (DT).
Journal Article
Pattern-based image retrieval using GLCM
Gray-level co-occurrence matrix (GLCM) is one of the oldest techniques used for texture analysis. It has two important parameters, i.e., distance and direction. In this paper, various combinations of distance and directional angles used for GLCM calculation are analyzed in order to recognize certain patterned images based on their textural features. In the proposed approach, the work is divided into two modules: determining the pattern of the image and pattern retrieval from the dataset. Patterns considered in this paper are horizontal striped, vertical striped, right and left diagonally striped, checkered and other images. For recognizing the pattern, the proposed approach has achieved a percentage accuracy of 96, 98, 96, 90, 96 and 94 for horizontal striped, vertical striped, right and left diagonally striped, checkered and other irregular patterns (not fully stripped), respectively. The proposed approach has a practical implementation in the fashion industry so to filter the search according to the pattern of the cloth.
Journal Article
Fake review and reviewer detection through behavioral graph partitioning integrating deep neural network
With a profound effect of online reviews on customers’ decisions about purchasing products or services, untruthful (fake) reviews written to deceive product quality and receive unfair commercial benefits have become a crucial problem. In this work, we propose a graph partitioning approach (BeGP) and its extension (BeGPX) to distinguish fake reviewers from benign ones. The main idea of BeGP is to first construct a behavioral graph in which reviewers are connected if they share common characteristic features that capture their similar behavior. Then, the algorithm starts with a small subgraph of known fake reviewers and afterwards repeatedly expands the subgraph by inducing other connected suspicious reviewers. Subsequently, all reviews of those suspects are hypothesized to be untruthful. Moreover, to enhance the performance of fake review(er) detection, BeGPX employs additional analysis of semantic content and emotions expressed in reviews. In particular, we use the deep neural network to learn word embeddings representation and lexicon-based emotion indicators in order to integrate into the graph construction process. We demonstrate the effectiveness of BeGP and BeGPX on two real-world review datasets from Yelp.com. The results show that both approaches outperform state-of-the-art methods with accurately identifying fake review(er)s within the
k
-first order of rankings. In addition, BeGPX shows significant enhancement although being provided with only a few amount of learning labeled data.
Journal Article
A flexible framework for anomaly Detection via dimensionality reduction
Anomaly detection is challenging, especially for large datasets in high dimensions. Here, we explore a general anomaly detection framework based on dimensionality reduction and unsupervised clustering. DRAMA is released as a general python package that implements the general framework with a wide range of built-in options. This approach identifies the primary prototypes in the data with anomalies detected by their large distances from the prototypes, either in the latent space or in the original, high-dimensional space. DRAMA is tested on a wide variety of simulated and real datasets, in up to 3000 dimensions, and is found to be robust and highly competitive with commonly used anomaly detection algorithms, especially in high dimensions. The flexibility of the DRAMA framework allows for significant optimization once some examples of anomalies are available, making it ideal for online anomaly detection, active learning, and highly unbalanced datasets. Besides, DRAMA naturally provides clustering of outliers for subsequent analysis.
Journal Article