Showing papers in "Applied Intelligence in 2022"

A tree-based stacking ensemble technique with feature selection for network intrusion detection

Abstract: Several studies have used machine learning algorithms to develop intrusion systems (IDS), which differentiate anomalous behaviours from the normal activities of network systems. Due to the ease of automated data collection and subsequently an increased size of collected data on network traffic and activities, the complexity of intrusion analysis is increasing exponentially. A particular issue, due to statistical and computation limitations, a single classifier may not perform well for large scale data as existent in modern IDS contexts. Ensemble methods have been explored in literature in such big data contexts. Although more complicated and requiring additional computation, literature has a note that ensemble methods can result in better accuracy than single classifiers in different large scale data classification contexts, and it is interesting to explore how ensemble approaches can perform in IDS. In this research, we introduce a tree-based stacking ensemble technique (SET) and test the effectiveness of the proposed model on two intrusion datasets (NSL-KDD and UNSW-NB15). We further enhance incorporate feature selection techniques to select the best relevant features with the proposed SET. A comprehensive performance analysis shows that our proposed model can better identify the normal and anomaly traffic in network than other existing IDS models. This implies the potentials of our proposed system for cybersecurity in Internet of Things (IoT) and large scale networks.

Machine learning techniques applied to mechanical fault diagnosis and fault prognosis in the context of real industrial manufacturing use-cases: a systematic literature review

Abstract: When put into practice in the real world, predictive maintenance presents a set of challenges for fault detection and prognosis that are often overlooked in studies validated with data from controlled experiments, or numeric simulations. For this reason, this study aims to review the recent advancements in mechanical fault diagnosis and fault prognosis in the manufacturing industry using machine learning methods. For this systematic review, we searched Web of Science, ACM Digital Library, Science Direct, Wiley Online Library, and IEEE Xplore between January 2015 and October 2021. Full-length studies that employed machine learning algorithms to perform mechanical fault detection or fault prognosis in manufacturing equipment and presented empirical results obtained from industrial case-studies were included, except for studies not written in English or published in sources other than peer-reviewed journals with JCR Impact Factor, conference proceedings and book chapters/sections. Of 4549 records, 44 primary studies were selected. In 37 of those studies, fault diagnosis and prognosis were performed using artificial neural networks (n = 12), decision tree methods (n = 11), hybrid models (n = 8), or latent variable models (n = 6), with one of the studies employing two different types of techniques independently. The remaining studies employed a variety of machine learning techniques, ranging from rule-based models to partition-based algorithms, and only two studies approached the problem using online learning methods. The main advantages of these algorithms include high performance, the ability to uncover complex nonlinear relationships and computational efficiency, while the most important limitation is the reduction in model performance in the presence of concept drift. This review shows that, although the number of studies performed in the manufacturing industry has been increasing in recent years, additional research is necessary to address the challenges presented by real-world scenarios.

General Performance Score for classification problems

Abstract: Abstract Several performance metrics are currently available to evaluate the performance of Machine Learning (ML) models in classification problems. ML models are usually assessed using a single measure because it facilitates the comparison between several models. However, there is no silver bullet since each performance metric emphasizes a different aspect of the classification. Thus, the choice depends on the particular requirements and characteristics of the problem. An additional problem arises in multi-class classification problems, since most of the well-known metrics are only directly applicable to binary classification problems. In this paper, we propose the General Performance Score (GPS) , a methodological approach to build performance metrics for binary and multi-class classification problems. The basic idea behind GPS is to combine a set of individual metrics, penalising low values in any of them. Thus, users can combine several performance metrics that are relevant in the particular problem based on their preferences obtaining a conservative combination. Different GPS -based performance metrics are compared with alternatives in classification problems using real and simulated datasets. The metrics built using the proposed method improve the stability and explainability of the usual performance metrics. Finally, the GPS brings benefits in both new research lines and practical usage, where performance metrics tailored for each particular problem are considered.

Deepfakes generation and detection: state-of-the-art, open challenges, countermeasures, and way forward

Abstract: Easy access to audio-visual content on social media, combined with the availability of modern tools such as Tensorflow or Keras, and open-source trained models, along with economical computing infrastructure, and the rapid evolution of deep-learning (DL) methods have heralded a new and frightening trend. Particularly, the advent of easily available and ready to use Generative Adversarial Networks (GANs), have made it possible to generate deepfakes media partially or completely fabricated with the intent to deceive to disseminate disinformation and revenge porn, to perpetrate financial frauds and other hoaxes, and to disrupt government functioning. Existing surveys have mainly focused on the detection of deepfake images and videos; this paper provides a comprehensive review and detailed analysis of existing tools and machine learning (ML) based approaches for deepfake generation, and the methodologies used to detect such manipulations in both audio and video. For each category of deepfake, we discuss information related to manipulation approaches, current public datasets, and key standards for the evaluation of the performance of deepfake detection techniques, along with their results. Additionally, we also discuss open challenges and enumerate future directions to guide researchers on issues which need to be considered in order to improve the domains of both deepfake generation and detection. This work is expected to assist readers in understanding how deepfakes are created and detected, along with their current limitations and where future research may lead.

Three-dimensional DenseNet self-attention neural network for automatic detection of student’s engagement

Abstract: Today, due to the widespread outbreak of the deadly coronavirus, popularly known as COVID-19, the traditional classroom education has been shifted to computer-based learning. Students of various cognitive and psychological abilities participate in the learning process. However, most students are hesitant to provide regular and honest feedback on the comprehensiveness of the course, making it difficult for the instructor to ensure that all students are grasping the information at the same rate. The students' understanding of the course and their emotional engagement, as indicated via facial expressions, are intertwined. This paper attempts to present a three-dimensional DenseNet self-attention neural network (DenseAttNet) used to identify and evaluate student participation in modern and traditional educational programs. With the Dataset for Affective States in E-Environments (DAiSEE), the proposed DenseAttNet model outperformed all other existing methods, achieving baseline accuracy of 63.59% for engagement classification and 54.27% for boredom classification, respectively. Besides, DenseAttNet trained on all four multi-labels, namely boredom, engagement, confusion, and frustration has registered an accuracy of 81.17%, 94.85%, 90.96%, and 95.85%, respectively. In addition, we performed a regression experiment on DAiSEE and obtained the lowest Mean Square Error (MSE) value of 0.0347. Finally, the proposed approach achieves a competitive MSE of 0.0877 when validated on the Emotion Recognition in the Wild Engagement Prediction (EmotiW-EP) dataset.

Attention-based residual autoencoder for video anomaly detection

Abstract: Abstract Automatic anomaly detection is a crucial task in video surveillance system intensively used for public safety and others. The present system adopts a spatial branch and a temporal branch in a unified network that exploits both spatial and temporal information effectively. The network has a residual autoencoder architecture, consisting of a deep convolutional neural network-based encoder and a multi-stage channel attention-based decoder, trained in an unsupervised manner. The temporal shift method is used for exploiting the temporal feature, whereas the contextual dependency is extracted by channel attention modules. System performance is evaluated using three standard benchmark datasets. Result suggests that our network outperforms the state-of-the-art methods, achieving 97.4% for UCSD Ped2, 86.7% for CUHK Avenue, and 73.6% for ShanghaiTech dataset in term of Area Under Curve, respectively.

A Cluster-based Stratified Hybrid Decision Support Model under Uncertainty: Sustainable Healthcare Landfill Location Selection

Abstract: Nowadays, healthcare waste management has become one of the significant environmental, health, and social problems. Due to population and urbanization growth and an increase in healthcare waste disposals according to the growing number of diseases and pandemics like COVID-19, disposal of healthcare waste has become a critical issue. Authorities in big cities require reliable decision support systems to empower them to make strategic decisions to provide safe disposal methods with a prospective vision. Since inappropriate healthcare waste management systems would definitely bring up dangerous environmental, social, health, and economic issues for every city. Therefore, this paper attempts to address the landfill location selection problem for healthcare waste using a novel decision support system. Novel decision support model integrates K-means algorithms with Stratified Best-Worst Method (SBWM) and a novel hybrid MARCOS-CoCoSo under grey interval numbers. The proposed decision support system considers waste generate rate in medical centers, future unforeseen but potential events, and uncertainty in experts' opinion to optimally locate required landfills for safe and economical disposal of dangerous healthcare waste. To investigate the feasibility and applicability of the proposed methodology, a real case study is performed for Mazandaran province in Iran. Our proposed methodology could efficiently deal with 79 medical centers within 4 clusters addressing 9 criteria to prioritize candidate locations. Moreover, the sensitivity analysis of weight coefficients is carried out to evaluate the results. Finally, the efficiency of the methodology is compared with several well-known methods and its high efficiency is demonstrated. Results recommend adherence to local rules and regulations, and future expansion potential as the top two criteria with importance values of 0.173 and 0.164, respectively. Later, best location alternatives are determined for each cluster of medical centers.

COVID-CT-Mask-Net: prediction of COVID-19 from CT scans using regional features

Abstract: We present COVID-CT-Mask-Net model that predicts COVID-19 in chest CT scans. The model works in two stages: in the first stage, Mask R-CNN is trained to localize and detect two types of lesions in images. In the second stage, these detections are fused to classify the whole input image. To develop the solution for the three-class problem (COVID-19, Common Pneumonia and Control), we used the COVIDx-CT data split derived from the dataset of chest CT scans collected by China National Center for Bioinformation. We use 3000 images (about 5% of the train split of COVIDx-CT) to train the model. Without any complicated data normalization, balancing and regularization, and training only a small fraction of the model’s parameters, we achieve a 90.80% COVID-19 sensitivity, 91.62% Common Pneumonia sensitivity and 92.10% true negative rate (Control sensitivity), an overall accuracy of 91.66% and F1-score of 91.50% on the test data split with 21192 images, bringing the ratio of test to train data to 7.06. We also establish an important result that regional predictions (bounding boxes with confidence scores) detected by Mask R-CNN can be used to classify whole images. The full source code, models and pretrained weights are available on https://github.com/AlexTS1980/COVID-CT-Mask-Net .

Semantic segmentation in medical images through transfused convolution and transformer networks

Abstract: Recent decades have witnessed rapid development in the field of medical image segmentation. Deep learning-based fully convolution neural networks have played a significant role in the development of automated medical image segmentation models. Though immensely effective, such networks only take into account localized features and are unable to capitalize on the global context of medical image. In this paper, two deep learning based models have been proposed namely USegTransformer-P and USegTransformer-S. The proposed models capitalize upon local features and global features by amalgamating the transformer-based encoders and convolution-based encoders to segment medical images with high precision. Both the proposed models deliver promising results, performing better than the previous state of the art models in various segmentation tasks such as Brain tumor, Lung nodules, Skin lesion and Nuclei segmentation. The authors believe that the ability of USegTransformer-P and USegTransformer-S to perform segmentation with high precision could remarkably benefit medical practitioners and radiologists around the world.

Rescuing emergency cases of COVID-19 patients: An intelligent real-time MSC transfusion framework based on multicriteria decision-making methods

Abstract: Mesenchymal stem cells (MSCs) have shown promising ability to treat critical cases of coronavirus disease 2019 (COVID-19) by regenerating lung cells and reducing immune system overreaction. However, two main challenges need to be addressed first before MSCs can be efficiently transfused to the most critical cases of COVID-19. First is the selection of suitable MSC sources that can meet the standards of stem cell criteria. Second is differentiating COVID-19 patients into different emergency levels automatically and prioritising them in each emergency level. This study presents an efficient real-time MSC transfusion framework based on multicriteria decision-making(MCDM) methods. In the methodology, the testing phase represents the ability to adhere to plastic surfaces, the upregulation and downregulation of specific surface protein markers and finally the ability to differentiate into different kinds of cells. In the development phase, firstly, two scenarios of an augmented dataset based on the medical perspective are generated to produce 80 patients with different emergency levels. Secondly, an automated triage algorithm based on a formal medical guideline is proposed for real-time monitoring of COVID-19 patients with different emergency levels (i.e. mild, moderate, severe and critical) considering the improvement and deterioration procedures from one level to another. Thirdly, a unique decision matrix for each triage level (except mild) is constructed on the basis of the intersection between the evaluation criteria of each emergency level and list of COVID-19 patients. Thereafter, MCDM methods (i.e. analytic hierarchy process [AHP] and vlsekriterijumska optimizcija i kaompromisno resenje [VIKOR]) are integrated to assign subjective weights for the evaluation criteria within each triage level and then prioritise the COVID-19 patients on the basis of individual and group decision-making(GDM) contexts. Results show that: (1) in both scenarios, the proposed algorithm effectively classified the patients into four emergency levels, including mild, moderate, severe and critical, taking into consideration the improvement and deterioration cases. (2) On the basis of experts' perspectives, clear differences in most individual prioritisations for patients with different emergency levels in both scenarios were found. (3) In both scenarios, COVID-19 patients were prioritised identically between the internal and external group VIKOR. During the evaluation, the statistical objective method indicated that the patient prioritisations underwent systematic ranking. Moreover, comparison analysis with previous work proved the efficiency of the proposed framework. Thus, the real-time MSC transfusion for COVID-19 patients can follow the order achieved in the group VIKOR results.