Special Issue "Advances in Artificial Intelligence: Machine Learning, Data Mining and Data Sciences"

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (20 July 2022) | Viewed by 79240

Special Issue Editors

Center for Applied Intelligent Systems Research, Halmstad University, SE-301 18 Halmstad, Sweden
Interests: machine learning; autonomous knowledge creation; representation learning; aware intelligent systems; predictive maintenance
Special Issues, Collections and Topics in MDPI journals
Center for Applied Intelligent Systems Research, Halmstad University, Halmstad, Sweden
Interests: machine learning; anomaly and novelty detection; interactive learning; data stream mining; big data
Special Issues, Collections and Topics in MDPI journals
Center for Applied Intelligent Systems Research, Halmstad University, Halmstad, Sweden
Interests: data mining; machine learning; tensor analysis; anomaly detection; time series analysis; spatiotemporal data mining

Special Issue Information

Dear Colleagues,

Machine learning (ML), data mining (DM), and data sciences in general are among the most exciting and rapidly growing research fields today. In recent years, ML and DM have been successfully used to solve practical problems in various domains, including engineering, healthcare, medicine, manufacturing, transportation, and finance.

In this era of big data, considerable research is being focused on designing efficient ML and DM methods. Nonetheless, practical applications of ML face several challenges, such as dealing with either too small or big data, missing and uncertain data, highly multidimensional data, and the need for interpretable ML models that can provide trustable evidence and explanations of the predictions they make. Moreover, in a time where the complexity of systems is continuously growing, it becomes not always feasible to collect clean and exhaustive datasets and produce high-quality labels. In addition, most systems generate data that are subject to change over time due to external conditions resulting in non-stationary data distributions. Therefore, there is a need to do more “knowledge creation”: to develop ML and DM methods that sift through large amounts of streaming data and extract useful high-level knowledge from there, without human supervision or with very little of it. In addition, learning and obtaining good generalization from fewer training examples, efficient data/knowledge representation schemes, knowledge transfer between tasks and domains, and learning to adapt to varying contexts are also examples of important research problems.

To address such problems, this Special Issue invites researchers to contribute new methods and to demonstrate the applicability of existing methods in various fields.

Topics of interest for this Special Issue include but are not limited to the following:

  • Novel methods and algorithms in machine learning, data mining, data science, including data cleaning, clustering, classification, feature selection and extraction, neural networks and deep learning, representation learning, knowledge discovery, anomaly detection, fault detection, transfer learning, and active learning;
  • Solutions improving the state-of-the-art regarding important challenges such as big data, streaming data, time series, interactive learning, concept drift and nonstationary data, change detection, and dimensionality reduction;
  • Applications in various domains, for example, activity and event recognition, computational biology and bioinformatics, computational social science, game playing, healthcare, information retrieval, natural language processing, predictive maintenance, recommender systems, signal processing, web applications, and internet data;
  • Societal challenges associated with AI, such as fairness, accountability, and transparency or privacy, anonymity, and security.

Prof. Sławomir Nowaczyk
Dr. Mohamed-Rafik Bouguelia‬
Dr. Hadi Fanaee
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2300 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Active Learning
  • Anomaly Detection
  • Big Data
  • Classification         
  • Clustering 
  • Causal Inference
  • Concept Drift
  • Data Mining
  • Data Science
  • Deep Learning
  • Fairness, Accountability, and Transparency of AI
  • Knowledge Discovery
  • Machine Learning
  • Medical Decision Support Systems
  • Multitask Learning  
  • Neural Networks
  • Predictive Models      
  • Representation Learning     
  • Semi-Supervised Learning 
  • Supervised Learning
  • Transfer Learning    
  • Unsupervised Learning       
  • Predictive Maintenance
  • Privacy, Anonymity, and Security of AI…

Published Papers (28 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Review

Article
Predictive Maintenance and Fault Monitoring Enabled by Machine Learning: Experimental Analysis of a TA-48 Multistage Centrifugal Plant Compressor
Appl. Sci. 2023, 13(3), 1790; https://doi.org/10.3390/app13031790 - 30 Jan 2023
Cited by 2 | Viewed by 3075
Abstract
In an increasingly competitive industrial world, the need to adapt to any change at any time has become a major necessity for every industry to remain competitive and survive in their environments. Industries are undergoing rapid and perpetual changes on several levels. Indeed, [...] Read more.
In an increasingly competitive industrial world, the need to adapt to any change at any time has become a major necessity for every industry to remain competitive and survive in their environments. Industries are undergoing rapid and perpetual changes on several levels. Indeed, the latter requires companies to be more reactive and involved in their policies of continuous improvement in order to satisfy their customers and maximize the quantity and quality of production, while keeping the cost of production as low as possible. Reducing downtime is one of the major objectives of these industries of the future. This paper aimed to apply machine learning algorithms on a TA-48 multistage centrifugal compressor for failure prediction and remaining useful life (RUL), i.e., to reduce system downtime using a predictive maintenance (PdM) approach through the adoption of Industry 4.0 approaches. To achieve our goal, we followed the methodology of the predictive maintenance workflow that allows us to explore and process the data for the model training. Thus, a comparative study of different prediction algorithms was carried out to arrive at the final choice, which is based on the implementation of LSTM neural networks. In addition, its performance was improved as the data sets were fed and incremented. Finally, the model was deployed to allow operators to know the failure times of compressors and subsequently ensure minimum downtime rates by making decisions before failures occur. Full article
Show Figures

Figure 1

Article
Metaheuristic Algorithms to Optimal Parameters Estimation of a Model of Two-Stage Anaerobic Digestion of Corn Steep Liquor
Appl. Sci. 2023, 13(1), 199; https://doi.org/10.3390/app13010199 - 23 Dec 2022
Viewed by 927
Abstract
Anaerobic Digestion (AD) of wastewater for hydrogen production is a promising technology resulting in the generation of value-added products and the reduction of the organic load of wastewater. The Two-Stage Anaerobic Digestion (TSAD) has several advantages over the conventional single-stage process due to [...] Read more.
Anaerobic Digestion (AD) of wastewater for hydrogen production is a promising technology resulting in the generation of value-added products and the reduction of the organic load of wastewater. The Two-Stage Anaerobic Digestion (TSAD) has several advantages over the conventional single-stage process due to the ability to control the acidification phase in the first bioreactor, preventing the overloading and/or the inhibition of the methanogenic population in the second bioreactor. To carry out any process research and process optimization, adequate mathematical models are required. To the best of our knowledge, no mathematical models of TSAD have been published in the literature so far. Therefore, the authors’ motivation is to present a high-quality model of the TSAD corn steeping process for the sequential production of H2 and CH4 considered in this paper. Four metaheuristics, namely Genetic Algorithm (GA), Firefly Algorithm (FA), Cuckoo Search Algorithm (CS), and Coyote Optimization Algorithm (COA), have been adapted and implemented for the first time for parameter identification of a new nonlinear mathematical model of TSAD of corn steep liquor proposed here. The superiority of some of the algorithms has been confirmed by a comparison of the observed numerical results, graphical results, and statistical analysis. The simulation results show that the four metaheuristics have achieved similar results in modelling the process dynamics in the first bioreactor. In the case of modelling the second bioreactor, a better description of the process dynamics trend has been obtained by FA, although GA has acquired the lowest value of the objective function. Full article
Show Figures

Figure 1

Article
Shear Wave Velocity Estimation Based on Deep-Q Network
Appl. Sci. 2022, 12(17), 8919; https://doi.org/10.3390/app12178919 - 05 Sep 2022
Cited by 1 | Viewed by 846
Abstract
Geoacoustic inversion is important for seabed geotechnical applications. It can be formulated as a problem that seeks an optimal solution in a high-dimensional parameter space. The conventional inversion approach exploits optimization methods with a pre-defined search strategy whose hyperparameters need to be fine-tuned [...] Read more.
Geoacoustic inversion is important for seabed geotechnical applications. It can be formulated as a problem that seeks an optimal solution in a high-dimensional parameter space. The conventional inversion approach exploits optimization methods with a pre-defined search strategy whose hyperparameters need to be fine-tuned for a specific scenario. A framework based on the deep-Q network is proposed in this paper and the environment and agent configurations of the framework are specially defined for geoacoustic inversion. Unlike a conventional optimization method with a pre-defined search strategy, the proposed framework determines a flexible strategy by trial and error. The proposed framework is evaluated by two case studies for estimating the shear wave velocity profile. Its performance is compared with three global optimization methods commonly used in underwater geoacoustic inversion. The results demonstrate that the proposed framework performs the inversion more efficiently and accurately. Full article
Show Figures

Figure 1

Article
Unsupervised and Supervised Feature Selection for Incomplete Data via L2,1-Norm and Reconstruction Error Minimization
Appl. Sci. 2022, 12(17), 8752; https://doi.org/10.3390/app12178752 - 31 Aug 2022
Viewed by 864
Abstract
Feature selection has been widely used in machine learning and data mining since it can alleviate the burden of the so-called curse of dimensionality of high-dimensional data. However, in previous works, researchers have designed feature selection methods with the assumption that all the [...] Read more.
Feature selection has been widely used in machine learning and data mining since it can alleviate the burden of the so-called curse of dimensionality of high-dimensional data. However, in previous works, researchers have designed feature selection methods with the assumption that all the information from a data set can be observed. In this paper, we propose unsupervised and supervised feature selection methods for use with incomplete data, further introducing an L2,1 norm and a reconstruction error minimization method. Specifically, the proposed feature selection objective functions take advantage of an indicator matrix reflecting unobserved information in incomplete data sets, and we present pairwise constraints, minimizing the L2,1-norm-robust loss functionand performing error reconstruction simultaneously. Furthermore, we derive two alternative iterative algorithms to effectively optimize the proposed objective functions and the convergence of the proposed algorithms is proven theoretically. Extensive experimental studies were performed on both real and synthetic incomplete data sets to demonstrate the performance of the proposed methods. Full article
Show Figures

Figure 1

Article
Generation of Controlled Synthetic Samples and Impact of Hyper-Tuning Parameters to Effectively Classify the Complex Structure of Overlapping Region
Appl. Sci. 2022, 12(16), 8371; https://doi.org/10.3390/app12168371 - 22 Aug 2022
Viewed by 1007
Abstract
The classification of imbalanced and overlapping data has provided customary insight over the last decade, as most real-world applications comprise multiple classes with an imbalanced distribution of samples. Samples from different classes overlap near class boundaries, creating a complex structure for the underlying [...] Read more.
The classification of imbalanced and overlapping data has provided customary insight over the last decade, as most real-world applications comprise multiple classes with an imbalanced distribution of samples. Samples from different classes overlap near class boundaries, creating a complex structure for the underlying classifier. Due to the imbalanced distribution of samples, the underlying classifier favors samples from the majority class and ignores samples representing the least minority class. The imbalanced nature of the data—resulting in overlapping regions—greatly affects the learning of various machine learning classifiers, as most machine learning classifiers are designed to handle balanced datasets and perform poorly when applied to imbalanced data. To improve learning on multi-class problems, more expertise is required in both traditional classifiers and problem domain datasets. Some experimentation and knowledge of hyper-tuning the parameters and parameters of the classifier under consideration are required. Several techniques for learning from multi-class problems have been reported in the literature, such as sampling techniques, algorithm adaptation methods, transformation methods, hybrid methods, and ensemble techniques. In the current research work, we first analyzed the learning behavior of state-of-the-art ensemble and non-ensemble classifiers on imbalanced and overlapping multi-class data. After analysis, we used grid search techniques to optimize key parameters (by hyper-tuning) of ensemble and non-ensemble classifiers to determine the optimal set of parameters to enhance the learning from a multi-class imbalanced classification problem, performed on 15 public datasets. After hyper-tuning, 20% of the dataset samples are synthetically generated to add to the majority class of each respective dataset to make it more overlapped (complex structure). After the synthetic sample’s addition, the hyper-tuned ensemble and non-ensemble classifiers are tested over that complex structure. This paper also includes a brief description of tuned parameters and their effects on imbalanced data, followed by a detailed comparison of ensemble and non-ensemble classifiers with the default and tuned parameters for both original and synthetically overlapped datasets. We believe that the underlying paper is the first kind of effort in this domain, which will furnish various research aspects to with a greater focus on the parameters of the classifier in the field of learning from imbalanced data problems using machine-learning algorithms. Full article
Show Figures

Figure 1

Article
MFHE: Multi-View Fusion-Based Heterogeneous Information Network Embedding
Appl. Sci. 2022, 12(16), 8218; https://doi.org/10.3390/app12168218 - 17 Aug 2022
Viewed by 930
Abstract
Depending on the type of information network, information network embedding is classified into homogeneous information network embedding and heterogeneous information network (HIN) embedding. Compared with the homogeneous network, HIN composition is more complex and contains richer semantics. At present, the research on homogeneous [...] Read more.
Depending on the type of information network, information network embedding is classified into homogeneous information network embedding and heterogeneous information network (HIN) embedding. Compared with the homogeneous network, HIN composition is more complex and contains richer semantics. At present, the research on homogeneous information network embedding is relatively mature. However, if the homogeneous information network model is directly applied to HIN, it will cause incomplete information extraction. It is necessary to build a specialized embedding model for HIN. Learning information network embedding based on the meta-path is an effective approach to extracting semantic information. Nevertheless, extracting HIN embedding only from a single view will cause information loss. To solve these problems, we propose a multi-view fusion-based HIN embedding model, called MFHE. MFHE includes four parts: node feature space transformation, subview information extraction, multi-view information fusion, and training. MFHE divides HIN into different subviews based on meta-paths, models the local information accurately in the subviews based on the multi-head attention mechanism, and then fuses subview information through a spatial matrix. In this paper, we consider the relationship between subviews; thus, the MFHE is applicable to complex HIN embedding. Experiments are conducted on ACM and DBLP datasets. Compared with baselines, the experimental results demonstrate that the effectiveness of MFHE and HIN embedding has been improved. Full article
Show Figures

Figure 1

Article
An Earthquake Forecast Model Based on Multi-Station PCA Algorithm
Appl. Sci. 2022, 12(7), 3311; https://doi.org/10.3390/app12073311 - 24 Mar 2022
Cited by 2 | Viewed by 1500
Abstract
With the continuous development of human society, earthquakes are becoming more and more dangerous to the production and life of human society. Researchers continue to try to predict earthquakes, but the results are still not significant. With the development of data science, sensing [...] Read more.
With the continuous development of human society, earthquakes are becoming more and more dangerous to the production and life of human society. Researchers continue to try to predict earthquakes, but the results are still not significant. With the development of data science, sensing and communication technologies, there are increasing efforts to use machine learning methods to predict earthquakes. Our work raises a method that applies big data analysis and machine learning algorithms to earthquakes prediction. All data are accumulated by the Acoustic and Electromagnetic Testing All in One System (AETA). We propose the multi-station Principal Component Analysis (PCA) algorithm and extract features based on this method. At last, we propose a weekly-scale earthquake prediction model, which has a 60% accuracy using LightGBM (LGB). Full article
Show Figures

Figure 1

Article
Research on the Feasibility of Applying GRU and Attention Mechanism Combined with Technical Indicators in Stock Trading Strategies
Appl. Sci. 2022, 12(3), 1007; https://doi.org/10.3390/app12031007 - 19 Jan 2022
Cited by 9 | Viewed by 1949
Abstract
The vigorous development of Time Series Neural Network in recent years has brought many potential possibilities to the application of financial technology. This research proposes a stock trend prediction model that combines Gate Recurrent Unit and Attention mechanism. In the proposed framework, the [...] Read more.
The vigorous development of Time Series Neural Network in recent years has brought many potential possibilities to the application of financial technology. This research proposes a stock trend prediction model that combines Gate Recurrent Unit and Attention mechanism. In the proposed framework, the model takes the daily opening price, closing price, highest price, lowest price and trading volume of stocks as input, and uses technical indicator transition prediction as a label to predict the possible rise and fall probability of future trading days. The research results show that the proposed model and labels designed by this research can effectively predict important stock price fluctuations and can be effectively applied to financial commodity trading strategies. Full article
Show Figures

Figure 1

Article
Enhanced DQN Framework for Selecting Actions and Updating Replay Memory Considering Massive Non-Executable Actions
Appl. Sci. 2021, 11(23), 11162; https://doi.org/10.3390/app112311162 - 24 Nov 2021
Cited by 3 | Viewed by 1688
Abstract
A Deep-Q-Network (DQN) controls a virtual agent as the level of a player using only screenshots as inputs. Replay memory selects a limited number of experience replays according to an arbitrary batch size and updates them using the associated Q-function. Hence, relatively fewer [...] Read more.
A Deep-Q-Network (DQN) controls a virtual agent as the level of a player using only screenshots as inputs. Replay memory selects a limited number of experience replays according to an arbitrary batch size and updates them using the associated Q-function. Hence, relatively fewer experience replays of different states are utilized when the number of states is fixed and the state of the randomly selected transitions becomes identical or similar. The DQN may not be applicable in some environments where it is necessary to perform the learning process using more experience replays than is required by the limited batch size. In addition, because it is unknown whether each action can be executed, a problem of an increasing amount of repetitive learning occurs as more non-executable actions are selected. In this study, an enhanced DQN framework is proposed to resolve the batch size problem and reduce the learning time of a DQN in an environment with numerous non-executable actions. In the proposed framework, non-executable actions are filtered to reduce the number of selectable actions to identify the optimal action for the current state. The proposed method was validated in Gomoku, a strategy board game, in which the application of a traditional DQN would be difficult. Full article
Show Figures

Figure 1

Article
A Novel Embedded Feature Selection and Dimensionality Reduction Method for an SVM Type Classifier to Predict Periventricular Leukomalacia (PVL) in Neonates
Appl. Sci. 2021, 11(23), 11156; https://doi.org/10.3390/app112311156 - 24 Nov 2021
Cited by 3 | Viewed by 1343
Abstract
This paper is concerned with the prediction of the occurrence of periventricular leukomalacia (PVL) in neonates after heart surgery. Our prior work shows that the Support Vector Machine (SVM) classifier can be a powerful tool in predicting clinical outcomes of such complicated and [...] Read more.
This paper is concerned with the prediction of the occurrence of periventricular leukomalacia (PVL) in neonates after heart surgery. Our prior work shows that the Support Vector Machine (SVM) classifier can be a powerful tool in predicting clinical outcomes of such complicated and uncommon diseases, even when the number of data samples is low. In the presented work, we first illustrate and discuss the shortcomings of the traditional automatic machine learning (aML) approach. Consequently, we describe our methodology for addressing these shortcomings, while utilizing the designed interactive ML (iML) algorithm. Finally, we conclude with a discussion of the developed method and the results obtained. In sum, by adding an additional (Genetic Algorithm) optimization step in the SVM learning framework, we were able to (a) reduce the dimensionality of an SVM model from 248 to 53 features, (b) increase generalization that was confirmed by a 100% accuracy assessed on an unseen testing set, and (c) improve the overall SVM model’s performance from 65% to 100% testing accuracy, utilizing the proposed iML method. Full article
Show Figures

Figure 1

Article
A Study on the Anomaly Detection of Engine Clutch Engagement/Disengagement Using Machine Learning for Transmission Mounted Electric Drive Type Hybrid Electric Vehicles
Appl. Sci. 2021, 11(21), 10187; https://doi.org/10.3390/app112110187 - 30 Oct 2021
Cited by 3 | Viewed by 1659
Abstract
Transmission mounted electric drive type hybrid electric vehicles (HEVs) engage/disengage an engine clutch when EV↔HEV mode transitions occur. If this engine clutch is not adequately engaged or disengaged, driving power is not transmitted correctly. Therefore, it is required to verify whether engine clutch [...] Read more.
Transmission mounted electric drive type hybrid electric vehicles (HEVs) engage/disengage an engine clutch when EV↔HEV mode transitions occur. If this engine clutch is not adequately engaged or disengaged, driving power is not transmitted correctly. Therefore, it is required to verify whether engine clutch engagement/disengagement operates normally in the vehicle development process. This paper studied machine learning-based methods for detecting anomalies in the engine clutch engagement/disengagement process. We trained the various models based on multi-layer perceptron (MLP), long short-term memory (LSTM), convolutional neural network (CNN), and one-class support vector machine (one-class SVM) with the actual vehicle test data and compared their results. The test results showed the one-class SVM-based models have the highest anomaly detection performance. Additionally, we found that configuring the training architecture to determine normal/anomaly by data instance and conducting one-class classification is proper for detecting anomalies in the target data. Full article
Show Figures

Figure 1

Article
Semantic Fusion with Deep Learning and Formal Ontologies for Evaluation of Policies and Initiatives in the Smart City Domain
Appl. Sci. 2021, 11(21), 10037; https://doi.org/10.3390/app112110037 - 26 Oct 2021
Cited by 1 | Viewed by 1755
Abstract
Decision makers and policy analysts at different administrative levels often lack a holistic view of the problem as there are semantic variations in policy documents due to domain-specific content. For example, smart city initiatives are derived from national and international initiatives which may [...] Read more.
Decision makers and policy analysts at different administrative levels often lack a holistic view of the problem as there are semantic variations in policy documents due to domain-specific content. For example, smart city initiatives are derived from national and international initiatives which may influence the incentives for local participants, but local initiatives reflect the local contextual elements of the city. Balanced assessment of smart city initiatives should include a systemic evaluation of the initiatives at multiple levels including the city, the country in which the city resides as well as at international level. In this paper, a knowledge elicitation methodology is presented for multi-granularity evaluation of policies and initiatives. The methodology is demonstrated on the evaluation of smart city initiatives generated at different administrative levels. Semantic networks are constructed using formal ontologies and deep learning methods for automatic semantic evaluation of initiatives to abstract knowledge found in text. Three smart city initiatives published by different administrative levels including international, national, and city level are evaluated in terms of relevance, coherence, and alignment of multi-level smart city initiatives. Experiments and analysis ultimately provide a holistic view of the problem which is necessary for decision makers and policy analysts of smart cities. Full article
Show Figures

Figure 1

Article
Phishing Webpage Classification via Deep Learning-Based Algorithms: An Empirical Study
Appl. Sci. 2021, 11(19), 9210; https://doi.org/10.3390/app11199210 - 03 Oct 2021
Cited by 14 | Viewed by 2391
Abstract
Phishing detection with high-performance accuracy and low computational complexity has always been a topic of great interest. New technologies have been developed to improve the phishing detection rate and reduce computational constraints in recent years. However, one solution is insufficient to address all [...] Read more.
Phishing detection with high-performance accuracy and low computational complexity has always been a topic of great interest. New technologies have been developed to improve the phishing detection rate and reduce computational constraints in recent years. However, one solution is insufficient to address all problems caused by attackers in cyberspace. Therefore, the primary objective of this paper is to analyze the performance of various deep learning algorithms in detecting phishing activities. This analysis will help organizations or individuals select and adopt the proper solution according to their technological needs and specific applications’ requirements to fight against phishing attacks. In this regard, an empirical study was conducted using four different deep learning algorithms, including deep neural network (DNN), convolutional neural network (CNN), Long Short-Term Memory (LSTM), and gated recurrent unit (GRU). To analyze the behaviors of these deep learning architectures, extensive experiments were carried out to examine the impact of parameter tuning on the performance accuracy of the deep learning models. In addition, various performance metrics were measured to evaluate the effectiveness and feasibility of DL models in detecting phishing activities. The results obtained from the experiments showed that no single DL algorithm achieved the best measures across all performance metrics. The empirical findings from this paper also manifest several issues and suggest future research directions related to deep learning in the phishing detection domain. Full article
Show Figures

Figure 1

Article
Gaze Tracking Using an Unmodified Web Camera and Convolutional Neural Network
Appl. Sci. 2021, 11(19), 9068; https://doi.org/10.3390/app11199068 - 29 Sep 2021
Cited by 5 | Viewed by 2308
Abstract
Gaze estimation plays a significant role in understating human behavior and in human–computer interaction. Currently, there are many methods accessible for gaze estimation. However, most approaches need additional hardware for data acquisition which adds an extra cost to gaze tracking. The classic gaze [...] Read more.
Gaze estimation plays a significant role in understating human behavior and in human–computer interaction. Currently, there are many methods accessible for gaze estimation. However, most approaches need additional hardware for data acquisition which adds an extra cost to gaze tracking. The classic gaze tracking approaches usually require systematic prior knowledge or expertise for practical operations. Moreover, they are fundamentally based on the characteristics of the eye region, utilizing infrared light and iris glint to track the gaze point. It requires high-quality images with particular environmental conditions and another light source. Recent studies on appearance-based gaze estimation have demonstrated the capability of neural networks, especially convolutional neural networks (CNN), to decode gaze information present in eye images and achieved significantly simplified gaze estimation. In this paper, a gaze estimation method that utilizes a CNN for gaze estimation that can be applied to various platforms without additional hardware is presented. An easy and fast data collection method is used for collecting face and eyes images from an unmodified desktop camera. The proposed method registered good results; it proves that it is possible to predict the gaze with reasonable accuracy without any additional tools. Full article
Show Figures

Figure 1

Article
Framework for the Ensemble of Feature Selection Methods
Appl. Sci. 2021, 11(17), 8122; https://doi.org/10.3390/app11178122 - 01 Sep 2021
Cited by 14 | Viewed by 2387
Abstract
Feature selection (FS) has attracted the attention of many researchers in the last few years due to the increasing sizes of datasets, which contain hundreds or thousands of columns (features). Typically, not all columns represent relevant values. Consequently, the noise or irrelevant columns [...] Read more.
Feature selection (FS) has attracted the attention of many researchers in the last few years due to the increasing sizes of datasets, which contain hundreds or thousands of columns (features). Typically, not all columns represent relevant values. Consequently, the noise or irrelevant columns could confuse the algorithms, leading to a weak performance of machine learning models. Different FS algorithms have been proposed to analyze highly dimensional datasets and determine their subsets of relevant features to overcome this problem. However, very often, FS algorithms are biased by the data. Thus, methods for ensemble feature selection (EFS) algorithms have become an alternative to integrate the advantages of single FS algorithms and compensate for their disadvantages. The objective of this research is to propose a conceptual and implementation framework to understand the main concepts and relationships in the process of aggregating FS algorithms and to demonstrate how to address FS on datasets with high dimensionality. The proposed conceptual framework is validated by deriving an implementation framework, which incorporates a set of Phyton packages with functionalities to support the assembly of feature selection algorithms. The performance of the implementation framework was demonstrated in several experiments discovering relevant features in the Sonar, SPECTF, and WDBC datasets. The experiments contrasted the accuracy of two machine learning classifiers (decision tree and logistic regression), trained with subsets of features generated either by single FS algorithms or the set of features selected by the ensemble feature selection framework. We observed that for the three datasets used (Sonar, SPECTF, and WD), the highest precision percentages (86.95%, 74.73%, and 93.85%, respectively) were obtained when the classifiers were trained with the subset of features generated by our framework. Additionally, the stability of the feature sets generated using our ensemble method was evaluated. The results showed that the method achieved perfect stability for the three datasets used in the evaluation. Full article
Show Figures

Figure 1

Article
Detection Model of Occluded Object Based on YOLO Using Hard-Example Mining and Augmentation Policy Optimization
Appl. Sci. 2021, 11(15), 7093; https://doi.org/10.3390/app11157093 - 31 Jul 2021
Cited by 6 | Viewed by 4214
Abstract
A study on object detection utilizing deep learning is in continuous progress to promptly and accurately determine the surrounding situation in the driving environment. Existing studies have tried to improve object detection performance considering occlusion through various processes. However, recent studies use R-CNN-based [...] Read more.
A study on object detection utilizing deep learning is in continuous progress to promptly and accurately determine the surrounding situation in the driving environment. Existing studies have tried to improve object detection performance considering occlusion through various processes. However, recent studies use R-CNN-based deep learning to provide high accuracy at slow speeds, so there are limitations to real-time. In addition, since such previous studies never took into consideration the data imbalance problem of the objects of interest in the model training process, it is necessary to make additional improvements. Accordingly, we proposed a detection model of occluded object based on YOLO using hard-example mining and augmentation policy optimization. The proposed procedures were as follows: diverse augmentation policies were applied to the base model in sequence and the optimized policy suitable for training data were strategically selected through the gradient-based performance improvement rate. Then, in the model learning process, the occluded objects and the objects likely to induce a false-positive detection were extracted, and fine-tuning using transfer learning was conducted. As a result of the performance evaluation, the model proposed in this study showed an [email protected] value of 90.49% and an F1-score value of 90%. It showed that this model detected occluded objects more stably and significantly enhanced the self-driving object detection accuracy compared with existing model. Full article
Show Figures

Figure 1

Article
Efficient High-Dimensional Kernel k-Means++ with Random Projection
Appl. Sci. 2021, 11(15), 6963; https://doi.org/10.3390/app11156963 - 28 Jul 2021
Cited by 2 | Viewed by 1415
Abstract
Using random projection, a method to speed up both kernel k-means and centroid initialization with k-means++ is proposed. We approximate the kernel matrix and distances in a lower-dimensional space Rd before the kernel k-means clustering motivated by upper error bounds. With random [...] Read more.
Using random projection, a method to speed up both kernel k-means and centroid initialization with k-means++ is proposed. We approximate the kernel matrix and distances in a lower-dimensional space Rd before the kernel k-means clustering motivated by upper error bounds. With random projections, previous work on bounds for dot products and an improved bound for kernel methods are considered for kernel k-means. The complexities for both kernel k-means with Lloyd’s algorithm and centroid initialization with k-means++ are known to be O(nkD) and Θ(nkD), respectively, with n being the number of data points, the dimensionality of input feature vectors D and the number of clusters k. The proposed method reduces the computational complexity for the kernel computation of kernel k-means from O(n2D) to O(n2d) and the subsequent computation for k-means with Lloyd’s algorithm and centroid initialization from O(nkD) to O(nkd). Our experiments demonstrate that the speed-up of the clustering method with reduced dimensionality d=200 is 2 to 26 times with very little performance degradation (less than one percent) in general. Full article
Show Figures

Figure 1

Article
Identifying the Author Group of Malwares through Graph Embedding and Human-in-the-Loop Classification
Appl. Sci. 2021, 11(14), 6640; https://doi.org/10.3390/app11146640 - 20 Jul 2021
Viewed by 1596
Abstract
Malware are developed for various types of malicious attacks, e.g., to gain access to a user’s private information or control of the computer system. The identification and classification of malware has been extensively studied in academic societies and many companies. Beyond the traditional [...] Read more.
Malware are developed for various types of malicious attacks, e.g., to gain access to a user’s private information or control of the computer system. The identification and classification of malware has been extensively studied in academic societies and many companies. Beyond the traditional research areas in this field, including malware detection, malware propagation analysis, and malware family clustering, this paper focuses on identifying the “author group” of a given malware as a means of effective detection and prevention of further malware threats, along with providing evidence for proper legal action. Our framework consists of a malware-feature bipartite graph construction, malware embedding based on DeepWalk, and classification of the target malware based on the k-nearest neighbors (KNN) classification. However, our KNN classifier often faced ambiguous cases, where it should say “I don’t know” rather than attempting to predict something with a high risk of misclassification. Therefore, our framework allows human experts to intervene in the process of classification for the final decision. We also developed a graphical user interface that provides the points of ambiguity for helping human experts to effectively determine the author group of the target malware. We demonstrated the effectiveness of our human-in-the-loop classification framework via extensive experiments using real-world malware data. Full article
Show Figures

Figure 1

Article
Linked Data Triples Enhance Document Relevance Classification
Appl. Sci. 2021, 11(14), 6636; https://doi.org/10.3390/app11146636 - 20 Jul 2021
Cited by 2 | Viewed by 1729
Abstract
Standardized approaches to relevance classification in information retrieval use generative statistical models to identify the presence or absence of certain topics that might make a document relevant to the searcher. These approaches have been used to better predict relevance on the basis of [...] Read more.
Standardized approaches to relevance classification in information retrieval use generative statistical models to identify the presence or absence of certain topics that might make a document relevant to the searcher. These approaches have been used to better predict relevance on the basis of what the document is “about”, rather than a simple-minded analysis of the bag of words contained within the document. In more recent times, this idea has been extended by using pre-trained deep learning models and text representations, such as GloVe or BERT. These use an external corpus as a knowledge-base that conditions the model to help predict what a document is about. This paper adopts a hybrid approach that leverages the structure of knowledge embedded in a corpus. In particular, the paper reports on experiments where linked data triples (subject-predicate-object), constructed from natural language elements are derived from deep learning. These are evaluated as additional latent semantic features for a relevant document classifier in a customized news-feed website. The research is a synthesis of current thinking in deep learning models in NLP and information retrieval and the predicate structure used in semantic web research. Our experiments indicate that linked data triples increased the F-score of the baseline GloVe representations by 6% and show significant improvement over state-of-the art models, like BERT. The findings are tested and empirically validated on an experimental dataset and on two standardized pre-classified news sources, namely the Reuters and 20 News groups datasets. Full article
Show Figures

Figure 1

Article
BHHO-TVS: A Binary Harris Hawks Optimizer with Time-Varying Scheme for Solving Data Classification Problems
Appl. Sci. 2021, 11(14), 6516; https://doi.org/10.3390/app11146516 - 15 Jul 2021
Cited by 14 | Viewed by 1503
Abstract
Data classification is a challenging problem. Data classification is very sensitive to the noise and high dimensionality of the data. Being able to reduce the model complexity can help to improve the accuracy of the classification model performance. Therefore, in this research, we [...] Read more.
Data classification is a challenging problem. Data classification is very sensitive to the noise and high dimensionality of the data. Being able to reduce the model complexity can help to improve the accuracy of the classification model performance. Therefore, in this research, we propose a novel feature selection technique based on Binary Harris Hawks Optimizer with Time-Varying Scheme (BHHO-TVS). The proposed BHHO-TVS adopts a time-varying transfer function that is applied to leverage the influence of the location vector to balance the exploration and exploitation power of the HHO. Eighteen well-known datasets provided by the UCI repository were utilized to show the significance of the proposed approach. The reported results show that BHHO-TVS outperforms BHHO with traditional binarization schemes as well as other binary feature selection methods such as binary gravitational search algorithm (BGSA), binary particle swarm optimization (BPSO), binary bat algorithm (BBA), binary whale optimization algorithm (BWOA), and binary salp swarm algorithm (BSSA). Compared with other similar feature selection approaches introduced in previous studies, the proposed method achieves the best accuracy rates on 67% of datasets. Full article
Show Figures

Figure 1

Article
Opportunities for Machine Learning in District Heating
Appl. Sci. 2021, 11(13), 6112; https://doi.org/10.3390/app11136112 - 30 Jun 2021
Cited by 10 | Viewed by 3639
Abstract
The district heating (DH) industry is facing an important transformation towards more efficient networks that utilise significantly lower water temperatures to distribute the heat. This change requires taking advantage of new technologies, and Machine Learning (ML) is a popular direction. In the last [...] Read more.
The district heating (DH) industry is facing an important transformation towards more efficient networks that utilise significantly lower water temperatures to distribute the heat. This change requires taking advantage of new technologies, and Machine Learning (ML) is a popular direction. In the last decade, we have witnessed an extreme growth in the number of published research papers that focus on applying ML techniques to the DH domain. However, based on our experience in the field, and an extensive review of the state-of-the-art, we perceive a mismatch between the most popular research directions, such as forecasting, and the challenges faced by the DH industry. In this work, we present our findings, explain and demonstrate the key gaps between the two communities and suggest a road-map ahead towards increasing the impact of ML research in the DH industry. Full article
Show Figures

Figure 1

Article
Seismic Reflection Analysis of AETA Electromagnetic Signals
Appl. Sci. 2021, 11(13), 5869; https://doi.org/10.3390/app11135869 - 24 Jun 2021
Cited by 4 | Viewed by 1486
Abstract
Acoustic and electromagnetics to artificial intelligence (AETA) is a system used to predict seismic events through monitoring of electromagnetic and geoacoustic signals. It is widely deployed in the Sichuan–Yunnan region (22° N–34° N, 98° E–107° E) of China. Generally, the electromagnetic signals of [...] Read more.
Acoustic and electromagnetics to artificial intelligence (AETA) is a system used to predict seismic events through monitoring of electromagnetic and geoacoustic signals. It is widely deployed in the Sichuan–Yunnan region (22° N–34° N, 98° E–107° E) of China. Generally, the electromagnetic signals of AETA stations near the epicenter have abnormal disturbances before an earthquake. When a significant decrease or increase in the signal is observed, it is difficult to quantify this change using only visual observation and confirm that it is related to an upcoming large earthquake. Considering that the AETA data comprise a typical time series, current work has analyzed the anomalism of AETA electromagnetic signals using the long short-term memory (LSTM) autoencoder method to prove that the electromagnetic anomaly of the AETA station can be regarded as an earthquake precursor. The results show that there are 2–4% anomalous points and some outliers exceeding 0.7 (after normalization) in the AETA stations within 200 km of the epicenter of the Jiuzaigou earthquake (M. 7.0) and the Yibin earthquake (M. 6.0) half a month before the earthquakes. Therefore, the AETA electromagnetic disturbance signal can be used as an earthquake precursor and for further earthquake prediction. Full article
Show Figures

Figure 1

Article
NPU RGBD Dataset and a Feature-Enhanced LSTM-DGCN Method for Action Recognition of Basketball Players+
Appl. Sci. 2021, 11(10), 4426; https://doi.org/10.3390/app11104426 - 13 May 2021
Cited by 7 | Viewed by 2435
Abstract
Computer vision-based action recognition of basketball players in basketball training and competition has gradually become a research hotspot. However, owing to the complex technical action, diverse background, and limb occlusion, it remains a challenging task without effective solutions or public dataset benchmarks. In [...] Read more.
Computer vision-based action recognition of basketball players in basketball training and competition has gradually become a research hotspot. However, owing to the complex technical action, diverse background, and limb occlusion, it remains a challenging task without effective solutions or public dataset benchmarks. In this study, we defined 32 kinds of atomic actions covering most of the complex actions for basketball players and built the dataset NPU RGB+D (a large scale dataset of basketball action recognition with RGB image data and Depth data captured in Northwestern Polytechnical University) for 12 kinds of actions of 10 professional basketball players with 2169 RGB+D videos and 75 thousand frames, including RGB frame sequences, depth maps, and skeleton coordinates. Through extracting the spatial features of the distances and angles between the joint points of basketball players, we created a new feature-enhanced skeleton-based method called LSTM-DGCN for basketball player action recognition based on the deep graph convolutional network (DGCN) and long short-term memory (LSTM) methods. Many advanced action recognition methods were evaluated on our dataset and compared with our proposed method. The experimental results show that the NPU RGB+D dataset is very competitive with the current action recognition algorithms and that our LSTM-DGCN outperforms the state-of-the-art action recognition methods in various evaluation criteria on our dataset. Our action classifications and this NPU RGB+D dataset are valuable for basketball player action recognition techniques. The feature-enhanced LSTM-DGCN has a more accurate action recognition effect, which improves the motion expression ability of the skeleton data. Full article
Show Figures

Figure 1

Article
Learning-Based Dissimilarity for Clustering Categorical Data
Appl. Sci. 2021, 11(8), 3509; https://doi.org/10.3390/app11083509 - 14 Apr 2021
Cited by 5 | Viewed by 1883
Abstract
Comparing data objects is at the heart of machine learning. For continuous data, object dissimilarity is usually taken to be object distance; however, for categorical data, there is no universal agreement, for categories can be ordered in several different ways. Most existing category [...] Read more.
Comparing data objects is at the heart of machine learning. For continuous data, object dissimilarity is usually taken to be object distance; however, for categorical data, there is no universal agreement, for categories can be ordered in several different ways. Most existing category dissimilarity measures characterize the distance among the values an attribute may take using precisely the number of different values the attribute takes (the attribute space) and the frequency at which they occur. These kinds of measures overlook attribute interdependence, which may provide valuable information when capturing per-attribute object dissimilarity. In this paper, we introduce a novel object dissimilarity measure that we call Learning-Based Dissimilarity, for comparing categorical data. Our measure characterizes the distance between two categorical values of a given attribute in terms of how likely it is that such values are confused or not when all the dataset objects with the remaining attributes are used to predict them. To that end, we provide an algorithm that, given a target attribute, first learns a classification model in order to compute a confusion matrix for the attribute. Then, our method transforms the confusion matrix into a per-attribute dissimilarity measure. We have successfully tested our measure against 55 datasets gathered from the University of California, Irvine (UCI) Machine Learning Repository. Our results show that it surpasses, in terms of various performance indicators for data clustering, the most prominent distance relations put forward in the literature. Full article
Show Figures

Figure 1

Article
EvoSplit: An Evolutionary Approach to Split a Multi-Label Data Set into Disjoint Subsets
Appl. Sci. 2021, 11(6), 2823; https://doi.org/10.3390/app11062823 - 22 Mar 2021
Cited by 3 | Viewed by 1515
Abstract
This paper presents a new evolutionary approach, EvoSplit, for the distribution of multi-label data sets into disjoint subsets for supervised machine learning. Currently, data set providers either divide a data set randomly or using iterative stratification, a method that aims to maintain the [...] Read more.
This paper presents a new evolutionary approach, EvoSplit, for the distribution of multi-label data sets into disjoint subsets for supervised machine learning. Currently, data set providers either divide a data set randomly or using iterative stratification, a method that aims to maintain the label (or label pair) distribution of the original data set into the different subsets. Following the same aim, this paper first introduces a single-objective evolutionary approach that tries to obtain a split that maximizes the similarity between those distributions independently. Second, a new multi-objective evolutionary algorithm is presented to maximize the similarity considering simultaneously both distributions (labels and label pairs). Both approaches are validated using well-known multi-label data sets as well as large image data sets currently used in computer vision and machine learning applications. EvoSplit improves the splitting of a data set in comparison to the iterative stratification following different measures: Label Distribution, Label Pair Distribution, Examples Distribution, folds and fold-label pairs with zero positive examples. Full article
Show Figures

Figure 1

Article
Recommendation Systems: Algorithms, Challenges, Metrics, and Business Opportunities
Appl. Sci. 2020, 10(21), 7748; https://doi.org/10.3390/app10217748 - 02 Nov 2020
Cited by 94 | Viewed by 14772
Abstract
Recommender systems are widely used to provide users with recommendations based on their preferences. With the ever-growing volume of information online, recommender systems have been a useful tool to overcome information overload. The utilization of recommender systems cannot be overstated, given its potential [...] Read more.
Recommender systems are widely used to provide users with recommendations based on their preferences. With the ever-growing volume of information online, recommender systems have been a useful tool to overcome information overload. The utilization of recommender systems cannot be overstated, given its potential influence to ameliorate many over-choice challenges. There are many types of recommendation systems with different methodologies and concepts. Various applications have adopted recommendation systems, including e-commerce, healthcare, transportation, agriculture, and media. This paper provides the current landscape of recommender systems research and identifies directions in the field in various applications. This article provides an overview of the current state of the art in recommendation systems, their types, challenges, limitations, and business adoptions. To assess the quality of a recommendation system, qualitative evaluation metrics are discussed in the paper. Full article
Show Figures

Figure 1

Article
Transitional SAX Representation for Knowledge Discovery for Time Series
Appl. Sci. 2020, 10(19), 6980; https://doi.org/10.3390/app10196980 - 06 Oct 2020
Cited by 6 | Viewed by 2262
Abstract
Numerous dimensionality-reducing representations of time series have been proposed in data mining and have proved to be useful, especially in handling a high volume of time series data. Among them, widely used symbolic representations such as symbolic aggregate approximation and piecewise aggregate approximation [...] Read more.
Numerous dimensionality-reducing representations of time series have been proposed in data mining and have proved to be useful, especially in handling a high volume of time series data. Among them, widely used symbolic representations such as symbolic aggregate approximation and piecewise aggregate approximation focus on information of local averages of time series. To compensate for such methods, several attempts were made to include trend information. However, the included trend information is quite simple, leading to great information loss. Such information is hardly extendable, so adjusting the level of simplicity to a higher complexity is difficult. In this paper, we propose a new symbolic representation method called transitional symbolic aggregate approximation that incorporates transitional information into symbolic aggregate approximations. We show that the proposed method, satisfying a lower bound of the Euclidean distance, is able to preserve meaningful information, including dynamic trend transitions in segmented time series, while still reducing dimensionality. We also show that this method is advantageous from theoretical aspects of interpretability, and practical and superior in terms of time-series classification tasks when compared with existing symbolic representation methods. Full article
Show Figures

Figure 1

Review

Jump to: Research

Review
On Predictive Maintenance in Industry 4.0: Overview, Models, and Challenges
Appl. Sci. 2022, 12(16), 8081; https://doi.org/10.3390/app12168081 - 12 Aug 2022
Cited by 30 | Viewed by 9037
Abstract
In the era of the fourth industrial revolution, several concepts have arisen in parallel with this new revolution, such as predictive maintenance, which today plays a key role in sustainable manufacturing and production systems by introducing a digital version of machine maintenance. The [...] Read more.
In the era of the fourth industrial revolution, several concepts have arisen in parallel with this new revolution, such as predictive maintenance, which today plays a key role in sustainable manufacturing and production systems by introducing a digital version of machine maintenance. The data extracted from production processes have increased exponentially due to the proliferation of sensing technologies. Even if Maintenance 4.0 faces organizational, financial, or even data source and machine repair challenges, it remains a strong point for the companies that use it. Indeed, it allows for minimizing machine downtime and associated costs, maximizing the life cycle of the machine, and improving the quality and cadence of production. This approach is generally characterized by a very precise workflow, starting with project understanding and data collection and ending with the decision-making phase. This paper presents an exhaustive literature review of methods and applied tools for intelligent predictive maintenance models in Industry 4.0 by identifying and categorizing the life cycle of maintenance projects and the challenges encountered, and presents the models associated with this type of maintenance: condition-based maintenance (CBM), prognostics and health management (PHM), and remaining useful life (RUL). Finally, a novel applied industrial workflow of predictive maintenance is presented including the decision support phase wherein a recommendation for a predictive maintenance platform is presented. This platform ensures the management and fluid data communication between equipment throughout their life cycle in the context of smart maintenance. Full article
Show Figures

Figure 1

Back to TopTop