Extended Feature Set Research Articles

Channels and Features Identification: A Review and a Machine-Learning Based Model With Large Scale Feature Extraction for Emotions and ASD Classification

Autism Spectrum Disorder (ASD) is characterized by impairments in social and cognitive skills, emotional disorders, anxiety, and depression. The prolonged conventional ASD diagnosis raises the sheer need for early meaningful intervention. Recently different works have proposed potential for ASD diagnosis and intervention through emotions prediction using deep neural networks (DNN) and machine learning algorithms. However, these systems lack an extensive large-scale feature extraction (LSFE) analysis through multiple benchmark data sets. LSFE analysis is required to identify and utilize the most relevant features and channels for emotion recognition and ASD prediction. Considering these challenges, for the first time, we have analyzed and evaluated an extensive feature set to select the optimal features using LSFE and feature selection algorithms (FSA). A set of up to eight most suitable channels was identified using different best-case FSA. The subject-wise importance of channels and features is also identified. The proposed method provides the best-case accuracies, precision, and recall of 95, 92, and 90%, respectively, for emotions prediction using a linear support vector machine (LSVM) classifier. It also provides the best-case accuracy, precision, and recall of 100% for ASD classification. This work utilized the largest number of benchmark data sets (5) and subjects (99) for validation reported till now in the literature. The LSVM classification algorithm proposed and utilized in this work has significantly lower complexity than the DNN, convolutional neural network (CNN), Naïve Bayes, and dynamic graph CNN used in recent ASD and emotion prediction systems.

Artificial neural network optimization to predict saturated hydraulic conductivity in arid and semi-arid regions

Saturated hydraulic conductivity (Ksat), one of the critical soil hydraulic properties, is used to model many soil hydrological processes. Measurement of Ksat on a routine basis is a labor-intensive, time-consuming, and expensive process. Alternatively, prediction of Ksat values from easy to obtain soil features is more economical and saves time. Artificial neural networks (ANNs) can be used to model and describe the most influential features affecting Ksat. This study aimed to develop and evaluate the potential use of generalized regression neural network (GRNN) to identify the optimal set of soil features to predict Ksat under arid and semi-arid environments. A total of 165 soil samples were collected from three depths (0–15, 15–30, and 30–60 cm) and analyzed for Ksat, texture, organic matter (OM), pH, bulk density (BD), and electrical conductivity (EC). Fourteen GRNN models were built with different feature combinations to identify the optimal set to predict Ksat. The results showed that soil texture explained 78% of the variability in soil Ksat while introducing EC improved model’s ability to estimate soil Ksat (R = 0.93, MSE = 2.89 × 10-12 m2 S-2). The optimum set of soil properties that should be included in the model were sand and clay percentages and EC values as evidenced from the cross-validation results. The GRNN model (using small dataset and set of features) provided reliable predictions of Ksat on bar with more complex models that included extensive set of features and used more extensive dataset. This work has implications for soil scients as provides an economical method to estimate Ksat values.

Privacy-preserving aware data aggregation for IoT-based healthcare with green computing technologies

Despite the rapid development of the Internet of Things technologies where more and more medical sensors and gadgets are connected to the Internet, limited energy resources due to transmission, and security are still main challenges. Most cases, patients wear multiple medical devices that transmit sensed medical data wirelessly to servers, which might cause big traffic on the communication networks, which in turn cause high energy consumption. Therefore, using the data aggregation, we can considerably reduce the energy consumption by eliminating redundant data; yet collected data must be fully protected. Secure data collection and transfer to centralized servers in healthcare applications employing IoT is quite challenging to protect against several attacks for illegal data access. For this reason, massive security measures should be taken to ensure that patients’ data can only be accessed by legitimate users. This paper proposes EPPADA: Efficient Privacy‑Preserving Authentication and Data Aggregation scheme in conjunction with Homomorphic Encryption concepts to meet requirements of healthcare using IoT with green computing technologies. The main objective of this proposed scheme is to decrease the communication overhead and energy consumption while maintaining safe and secure aggregation of the healthcare data between medical sensors and cloud servers. The proposed system is experimentally developed using E-health sensor shield V2.0 platform. Based on security analysis that has the most extensive set of security features, our multi-objective approach enhances the End-to-End Delay, Computational Cost, Communication Overhead, besides maintaining security features.

Forecasting TV ratings of Turkish television series using a two-level machinelearning framework

TV rating is a numeric estimate of the popularity of television programs. Forecasting TV ratings is considered an important asset for investment planning of media due to its potential of reducing the risks of future ventures. The aim of this study is to develop a machine learning model capable of efficiently forecasting the TV ratings of Turkish TV series in a practical manner. To this end, two prediction models were proposed for forecasting the TV ratings of television series, facilitating an extensive set of features. A contribution of this study is the inclusion of social media-based features using search trends around television series and exploration of the viability of using these features in place of temporal features. The study presents an extensive evaluation of the forecast performance of the proposed models. The performance of the proposed models were evaluated using a data collection composed of ratings and various attributes of series and their episodes aired at prime-time Turkish broadcast during 2014 and 2018. In the experiments, a theoretical forecast performance was first established with the inclusion of temporal features. Next, a set of practical models were proposed, replacing temporal attributes with social media-based attributes, relating to internet popularity and visibility of the series. The experiments show that, the proposed models achieve up to 1.65% error rate for the theoretical setting and 7.06% error rate for the practical setting.

Data-Driven Market Segmentation in Hospitality Using Unsupervised Machine Learning

Within hospitality, marketing departments use segmentation to create tailored strategies to ensure personalized marketing. This study provides a data-driven approach by segmenting guest profiles via hierarchical clustering, based on an extensive set of features. The industry requires understandable outcomes that contribute to adaptability for marketing departments to make data-driven decisions and ultimately driving profit. A marketing department specified a business question that guides the unsupervised machine learning algorithm. Features of guests change over time; therefore, there is a probability that guests transition from one segment to another. The purpose of the study is to provide steps in the process from raw data to actionable insights, which serve as a guideline for how hospitality companies can adopt an algorithmic approach.

An Approach for Optimization of Features using Gorilla Troop Optimizer for Classification of Melanoma

The diagnosis and categorization of skin cancer, as well as the difference in skin textures and injuries, is a tough undertaking. Manually detecting skin lesions from dermoscopy images seems to be a difficult and cumbersome challenge. Recent advancements in the internet of things (IoT) and artificial intelligence for clinical applications have shown significant increase in precision and processing time. A lot of attention is given to deep learning models because they are effective at identifying cancer cells. The diagnosis and accuracy levels can be greatly increased by categorizing benign and malignant dermoscopy images. This work suggests an automated classification system based on a deep convolutional neural network (DCNN) in order to precisely perform multi-classification. The DCNN's structure was thoughtfully created by arranging a number of layers that are in charge of uniquely extracting different features from skin lesions. In this paper, we proposed a deep learning approach to tackle the three main tasks-deep extraction of features (task1) using transfer learning, selection of features (task2)-using metaheuristic algorithms such as Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), and Gorilla Troop Optimization (GTO) as a feature selector, the extensive feature set is optimized, and the amount of features is reduced to within the range, and a two-level classification (task3) was proposed that are emerging in the field of skin lesion image processing. On the HAM10000 dataset, the proposed deep learning frameworks were assessed. The accuracy achieved on the dataset is 93.58 percent. The proposed method outperforms state-of-the-art (SOTA) techniques in terms of accuracy. The suggested technique is however highly scalable.

A Deep Learning Approach for IoT Traffic Multi-Classification in a Smart-City Scenario

As the number of Internet of Things (IoT) devices and applications increases, the capacity of the IoT access networks is considerably stressed. This can create significant performance bottlenecks in various layers of an end-to-end communication path, including the scheduling of the spectrum, the resource requirements for processing the IoT data at the Edge and/or Cloud, and the attainable delay for critical emergency scenarios. Thus, a proper classification or prediction of the time varying traffic characteristics of the IoT devices is required. However, this classification remains at large an open challenge. Most of the existing solutions are based on machine learning techniques, which nonetheless present high computational cost, whereas they are not considering the fine-grained flow characteristics of the traffic. To this end, this paper introduces the following four contributions. Firstly, we provide an extended feature set including, flow, packet and device level features to characterize the IoT devices in the context of a smart environment. Secondly, we propose a custom weighting based preprocessing algorithm to determine the importance of the data values. Thirdly, we present insights into traffic characteristics using feature selection and correlation mechanisms. Finally, we develop a two-stage learning algorithm and we demonstrate its ability to accurately categorize the IoT devices in two different datasets. The evaluation results show that the proposed learning framework achieves 99.9% accuracy for the first dataset and 99.8% accuracy for the second. Additionally, for the first dataset we achieve a precision and recall performance of 99.6% and 99.5%, while for the second dataset the precission and recall attained is of 99.6% and 99.7% respectively. These results show that our approach clearly outperforms other well-known machine learning methods. Hence, this work provides a useful model deployed in a realistic IoT scenario, where IoT traffic and devices’ profiles are predicted and classified, while facilitating the data processing in the upper layers of an end-to-end communication model.

An efficient malicious webpage static detection framework based on optimized Bayesian and hybrid machine learning

SummaryMalicious webpage detection is a crucial work in both theory and practical environment. In practical applications, static detection methods are usually regarded as a priority choice, which can quickly detect unknown malicious web pages and avoid a costly in‐depth analysis. However, existing solution of static detection typically has the following problems. For example, a single static detection may lead to a higher false positive rate, and the integrated detection usually has a lower detection efficiency. In this article, we propose an efficient webpage static detection framework, especially considering both the detection efficiency and the detection accuracy. Then, on the basis of the extended feature sets from URL, HTML, and JavaScript, we introduce an optimized naive Bayesian algorithm, in which a novel amplification factor strategy is proposed. Finally, a webpage threat assessment model oriented to general machine learning is presented to achieve the refined detection. Three main properties are provided: high detection efficiency, high detection accuracy, and better applicability. Furthermore, the comprehensive experimental results and comparative analysis is given to show the advantages of the proposed framework.

Vstplugin~] – A Pd external for hosting VST plugins

[vstplugin~] is a cross-platform Pure Data external for hosting VST 2 and VST 3 plugins that has been developed at the IEM in Graz. The external provides an extensive set of features which span several topics, such as channel layout, plugin search, real-time safety, GUI editor, parameter automation, preset management, MIDI events, transport and timing, bridging and sandboxing, multithreading and offline processing. We show the development process and discuss previous work in this area. We also cover the history of the VST technology, explain the differences between the VST 2 and VST 3 SDK and offer insight into their inner workings. Finally, we discuss possible improvements and extensions for the future.

QuickFigures: A toolkit and ImageJ PlugIn to quickly transform microscope images into scientific figures.

Publications involving fluorescent microscopy images generally contain many panels with split channels, merged images, scale bars and label text. Similar layouts of panels are used when displaying other microscopy images, electron micrographs, photographs, and other images. Assembling and editing these figures with even spacing, consistent font, text position, accurate scale bars, and other features can be tedious and time consuming. In order to save time, I have created a toolset and ImageJ Plugin called QuickFigures. QuickFigures includes many helpful features that streamline the process of creating, aligning, and editing scientific figures. Those features include tools that automatically create split channel figures from a region of interest (“Quick Figure” button and “Inset Tool”), layouts that make it easy to rearrange panels, multiple tools to align objects, and “Figure Format” menu options that help a user ensure that large numbers of figures have consistent appearance. QuickFigures was compared to previous tools by measuring the amount of time needed for a user to create a figure using each software (QuickFigures, OMERO.figure. EZFig, FigureJ and PowerPoint). QuickFigures significantly reduced the amount of time required to create a figure. The toolsets were also compared by checking each software against a list of features. QuickFigures had the most extensive set of features. Therefore, QuickFigures is an advantageous alternative to traditional methods of constructing scientific figures. After a user has saved time by creating their work in QuickFigures, the figures can be exported to a variety of formats including PowerPoint, PDF, SVG, PNG, TIFF and Adobe Illustrator. Export was successfully tested for each file format and object type. Exported objects and text are editable in their target software, making them suitable for sharing with collaborators. The software is free, open source and can be installed easily.

Predicting residential electricity consumption using aerial and street view images

Reducing the electricity consumption of buildings is an important lever in the global effort to reduce greenhouse gas emissions. However, for privacy and other reasons, there is a lack of data on building electricity consumption. As a consequence, data-driven tools that support decision-makers in this area are scarce. To address this problem, we present an innovative approach to modeling building electricity consumption that relies exclusively on publicly available aerial and street view images. We evaluate our approach in a case study based on real world data from Gainesville, Florida. The results show that our model can predict electricity consumption about as well as conventional models, which are trained on commonly used features that are typically not publicly available at a large scale. Furthermore, our model achieves 68% of the potential accuracy improvements of a model that relies on an extensive set of fine-grained tabular features. Spatially aggregating the predictions from the level of buildings to areas of up to 1km2 further improves the results.

Health Condition Estimation of Bearings with Multiple Faults by a Composite Learning-Based Approach.

Bearings are critical components found in most rotating machinery; their health condition is of immense importance to many industries. The varied conditions and environments in which bearings operate make them prone to single and multiple faults. Widespread interest in the improvements of single fault diagnosis meant limited attention was spent on multiple fault diagnosis. However, multiple fault diagnosis poses extra challenges due to the submergence of the weak fault by the strong fault, presence of non-Gaussian noise, coupling of the frequency components, etc. A number of existing convolutional neural network models operate on a distinct feature that is not enough to assure reliable results in the presence of these challenges. In this paper, extended feature sets in three homogenous deep learning models are used for multiple fault diagnosis. This ensures a measure of diversity is introduced to the health management dataset to obtain complementary solutions from the models. The outputs of the models are fused through blending ensemble learning. Experiments using vibration datasets based on bearing multiple faults show an accuracy of 98.54%, with an improvement of 2.74% in the overall effectiveness over the single models. Compared with other technologies, the results show that this approach provides an improved generalized diagnostic capability.

Identifying factors associated with roadside work zone collisions using machine learning techniques

Identifying factors that are associated with the probability of roadside work zone collisions enables decision makers to better assess and control the risk of scheduling a particular maintenance or construction activity by modifying the characteristics of the operation. This can be achieved by studying the effect of work zone properties on the risk of roadside work zone collisions. Much of the existing work in this area is based on data in the police traffic collision reports, which do not include data on the characteristics of the work zone itself. This paper develops a comprehensive data set of 42 features describing time, location, work zone characteristics, traffic volume, and road properties. Using recent machine learning techniques such as extreme gradient boosting classifiers on this extensive set of features allows for more accurate analysis to identify factors that affect the risk of work zone collisions or indicate higher than baseline chances of a roadside crash. Our statistical analysis reveals 10 important features and shows that four of these features are significantly associated with higher probabilities of roadside work zone collisions.

Power production strategies for two-stage PV systems during grid faults

The paper comprehensively addresses the control algorithm for the photovoltaic (PV) system. The system’s operation during both nominal and faulty grid conditions, most notably voltage unbalances, is addressed. The control of the boost converters and the inverter that secures reliable and maximized power production is proposed. Specifically, the paper offers a novel DC-link voltage stabilization algorithm that is not based on conventional approaches, such as droop schemes. Rather, the DC-link voltage rate of change is used to realize safe primary energy sources’ power management during transients caused by grid disturbances. This offers improved robustness, reliability and scalability. Moreover, the improved grid-connected converter control is proposed. It enables the production of arbitrary currents and powers profiles, during different working regimes, respecting the converter’s ratings. Consequently, higher-layer control functions can be realized by utilizing the proposed scheme. Also, the flexible power production prioritization feature is integrated into the overall solution. Such an extensive set of features was not demonstrated for PV systems in the literature previously. Different hardware topologies are regarded – centralized, decentralized and distributed. The proposed control schemes were tested using the hardware-in-the-loop (HIL) environment and were validated through the analysis of the obtained results.

Digital Ethnography: A Systematic Literature Review

Digital ethnography can be defined as a contemporary form of ethnography which considers online social spaces of discussion following the advances in data transmission technology. In the last few years, various attempts of considering online spaces in ethnographic research have been made producing different styles of online ethnography, each identified by a different label. This paper aims to provide a systematic review of the topic to map the practice of digital ethnography. The research process followed four key steps: search, selection, analysis and synthesis. In the search phase, we searched contributions iteratively in Web of Science and Scopus by using a variety of keywords corresponding to the different labels used to refer to digital ethnography. In the selection phase, we adopted a selective stance that aims to provide a critical review of the existing research and practices in the context of digital ethnography. In the analysis phase, we carried out a content analysis of the papers combining deductive and inductive coding. The synthesis phase involves a process of typology development to pragmatically reduce and systematize an extensive set of features and digital ethnography practices. Basing on the type of data collected (Small vs Big Data) and the type of fieldwork (meta or contextual field), we detected four types of ethnographic research: social media ethnography , contextual digital ethnography , meta digital ethnography , cross-media ethnography .

Classification and feature analysis of the Human Connectome Project dataset for differentiating between males and females

We analysed features relevant for differentiation between males and females based on the data available from the Human Connectome Project (HCP) S1200 dataset. We used 354 features containing cognitive and emotional measures as well as measures derived from task functional magnetic resonance imaging (MRI) and structural brain MRI. The paper presents a thorough analysis of this extensive set of features using a machine learning approach with a goal of identifying features that have the ability to differentiate between males and females. We used two state of the art classification algorithms with different properties: support vector machine (SVM) and random forest classifier (RFC). For each classifier the hyperparameters were obtained and classifiers were optimized using nested cross validation and grid search. This resulted in the classification performance of 91% and 89% accuracy using SVM and RFC, respectively. Using SHAP (SHapley Additive exPlanations) method we obtained relevance of features as indicators of sex differences and identified features with high discriminative power for sex classification. The majority of top features were brain morphological measures, and only a small proportion were features related to cognitive performance. Our results demonstrate the importance and advantages of using a machine learning approach when analysing sex differences.

Multi-Feature Sample Database for Enhancing Deep Learning Tasks in Operational Humanitarian Applications

Amongst the many benefits of remote sensing techniques in disaster- or conflict-related applications, timeliness and objectivity may be the most critical assets. Recently, increasing sensor quality and data availability have shifted the attention more towards the information extraction process itself. With promising results obtained by deep learning (DL), the notion arises that DL is not agnostic to input errors or biases introduced, in particular in sample-scarce situations. The present work seeks to understand the influence of different sample quality aspects propagating through network layers in automated image analysis. In this paper, we broadly discuss the conceptualisation of such a sample database in an early stage of realisation: (1) inherited properties (quality parameters of the underlying image such as cloud cover, seasonality, etc.); (2) individual (i.e., per-sample) properties, including a. lineage and provenance, b. geometric properties (size, orientation, shape), c. spectral features (standardized colour code); (3) context-related properties (arrangement Several hundred samples collected from different camp settings were hand-selected and annotated with computed features in an initial stage. The supervised annotation routine is automated so that thousands of existing samples can be labelled with this extended feature set. This should better condition the subsequent DL tasks in a hybrid AI approach.

Digital Twin in the IoT Context: A Survey on Technical Features, Scenarios, and Architectural Models

Digital twin (DT) is an emerging concept that is gaining attention in various industries. It refers to the ability to clone a physical object (PO) into a software counterpart. The softwarized object, termed logical object, reflects all the important properties and characteristics of the original object within a specific application context. To fully determine the expected properties of the DT, this article surveys the state-of-the-art starting from the original definition within the manufacturing industry. It takes into account related proposals emerging in other fields, namely augmented and virtual reality (e.g., avatars), multiagent systems, and virtualization. This survey thereby allows for the identification of an extensive set of DT features that point to the “softwarization” of POs. To properly consolidate a shared DT definition, a set of foundational properties is identified and proposed as a common ground outlining the essential characteristics (must-haves) of a DT. Once the DT definition has been consolidated, its technical and business value is discussed in terms of applicability and opportunities. Four application scenarios illustrate how the DT concept can be used and how some industries are applying it. The scenarios also lead to a generic DT architectural model. This analysis is then complemented by the identification of software architecture models and guidelines in order to present a general functional framework for the DT. This article, eventually, analyses a set of possible evolution paths for the DT considering its possible usage as a major enabler for the softwarization process.

Emotion Assessment Using Feature Fusion and Decision Fusion Classification Based on Physiological Data: Are We There Yet?

Emotion recognition based on physiological data classification has been a topic of increasingly growing interest for more than a decade. However, there is a lack of systematic analysis in literature regarding the selection of classifiers to use, sensor modalities, features and range of expected accuracy, just to name a few limitations. In this work, we evaluate emotion in terms of low/high arousal and valence classification through Supervised Learning (SL), Decision Fusion (DF) and Feature Fusion (FF) techniques using multimodal physiological data, namely, Electrocardiography (ECG), Electrodermal Activity (EDA), Respiration (RESP), or Blood Volume Pulse (BVP). The main contribution of our work is a systematic study across five public datasets commonly used in the Emotion Recognition (ER) state-of-the-art, namely: (1) Classification performance analysis of ER benchmarking datasets in the arousal/valence space; (2) Summarising the ranges of the classification accuracy reported across the existing literature; (3) Characterising the results for diverse classifiers, sensor modalities and feature set combinations for ER using accuracy and F1-score; (4) Exploration of an extended feature set for each modality; (5) Systematic analysis of multimodal classification in DF and FF approaches. The experimental results showed that FF is the most competitive technique in terms of classification accuracy and computational complexity. We obtain superior or comparable results to those reported in the state-of-the-art for the selected datasets.

Death Risk Prediction of Intensive Care Unit Patients Combined with Treatment Process Mining

Precise death risk prediction of ICU patients plays an important role in developing follow-up treatment options and reducing rescue cost. Previous work mainly relied on the vital signs of patients to predict the risk of death. In fact, patients’ treatment process also implies valuable information about patient’s condition. To use the treatment process information in patient risk assessment, this paper proposes a new death risk prediction method combined with treatment process mining. Specifically, historical patients are firstly clustered into several classes according to their final treatment outcome and the drugs they took. Patients in one cluster are assumed to share the same treatment process. Then, these treatment processes are mined by means of the LDA topic model and the probability suffix tree training method. Next, for a patient to do death risk prediction, the similarity between his/her treatment procedure and those probability suffix trees is computed. The similarity value is taken as a new feature complementary to those traditional ones in disease severity scoring systems. Finally, the random forest algorithm is used to train the death risk prediction model based on the extended feature set. Taking the prescription data of sepsis patients in the MIMIC-III database as input, an experiment is conduced to evaluate the proposed method. As expected, the proposed method achieves better prediction accuracy, recall rate, and F1 score compared to methods.