Abstract: Following a review of various approaches that are utilized for classifying the tail behavior of a distribution, an easily implementable methodology that relies on an arctangent transformation is presented. The classification criterion is actually based on the difference between two specific quantiles of the transformed distribution. The resulting categories enable one to classify distributional tails as distinctly short, short, nearly medium, medium, extended medium and somewhat long, providing that at least two moments exist. Distributions possessing a single moment are said to be long tailed while those failing to have any finite moments are classified as having an extremely long tail. Several illustrative examples will be presented.
Abstract: Engagement is one of the most important factors in determining successful outcomes and deep learning in students. Existing approaches to detect student engagement involve periodic human observations that are subject to inter-rater reliability. Our solution uses real-time multimodal multisensor data labeled by objective performance outcomes to infer the engagement of students. The study involves four students with a combined diagnosis of cerebral palsy and a learning disability who took part in a 3-month trial over 59 sessions. Multimodal multisensor data were collected while they participated in a continuous performance test. Eye gaze, electroencephalogram, body pose, and interaction data were used to create a model of student engagement through objective labeling from the continuous performance test outcomes. In order to achieve this, a type of continuous performance test is introduced, the Seek-X type. Nine features were extracted including high-level handpicked compound features. Using leave-one-out cross-validation, a series of different machine learning approaches were evaluated. Overall, the random forest classification approach achieved the best classification results. Using random forest, 93.3% classification for engagement and 42.9% accuracy for disengagement were achieved. We compared these results to outcomes from different models: AdaBoost, decision tree, k-Nearest Neighbor, naïve Bayes, neural network, and support vector machine. We showed that using a multisensor approach achieved higher accuracy than using features from any reduced set of sensors. We found that using high-level handpicked features can improve the classification accuracy in every sensor mode. Our approach is robust to both sensor fallout and occlusions. The single most important sensor feature to the classification of engagement and distraction was shown to be eye gaze. It has been shown that we can accurately predict the level of engagement of students with learning disabilities in a real-time approach that is not subject to inter-rater reliability, human observation or reliant on a single mode of sensor input. This will help teachers design interventions for a heterogeneous group of students, where teachers cannot possibly attend to each of their individual needs. Our approach can be used to identify those with the greatest learning challenges so that all students are supported to reach their full potential.
Abstract: The aim of this study is to develop a system which can identify and sort peaberries automatically at low cost for coffee producers in developing countries. In this paper, the focus is on the classification of peaberries and normal coffee beans using image processing and machine learning techniques. The peaberry is not bad and not a normal bean. The peaberry is born in an only single seed, relatively round seed from a coffee cherry instead of the usual flat-sided pair of beans. It has another value and flavor. To make the taste of the coffee better, it is necessary to separate the peaberry and normal bean before green coffee beans roasting. Otherwise, the taste of total beans will be mixed, and it will be bad. In roaster procedure time, all the beans shape, size, and weight must be unique; otherwise, the larger bean will take more time for roasting inside. The peaberry has a different size and different shape even though they have the same weight as normal beans. The peaberry roasts slower than other normal beans. Therefore, neither technique provides a good option to select the peaberries. Defect beans, e.g., sour, broken, black, and fade bean, are easy to check and pick up manually by hand. On the other hand, the peaberry pick up is very difficult even for trained specialists because the shape and color of the peaberry are similar to normal beans. In this study, we use image processing and machine learning techniques to discriminate the normal and peaberry bean as a part of the sorting system. As the first step, we applied Deep Convolutional Neural Networks (CNN) and Support Vector Machine (SVM) as machine learning techniques to discriminate the peaberry and normal bean. As a result, better performance was obtained with CNN than with SVM for the discrimination of the peaberry. The trained artificial neural network with high performance CPU and GPU in this work will be simply installed into the inexpensive and low in calculation Raspberry Pi system. We assume that this system will be used in under developed countries. The study evaluates and compares the feasibility of the methods in terms of accuracy of classification and processing speed.
Abstract: Conventional reservoir prediction methods ar not sufficient to explore the implicit relation between seismic attributes, and thus data utilization is low. In order to improve the predictive classification accuracy of reservoir lithology, this paper proposes a deep learning lithology prediction method based on ResNet (Residual Neural Network) and SENet (Squeeze-and-Excitation Neural Network). The neural network model is built and trained by using seismic attribute data and lithology data of Shengli oilfield, and the nonlinear mapping relationship between seismic attribute and lithology marker is established. The experimental results show that this method can significantly improve the classification effect of reservoir lithology, and the classification accuracy is close to 70%. This study can effectively predict the lithology of undrilled area and provide support for exploration and development.
Abstract: Chronic venous diseases (CVD) are one of the common, though controversial problems in medicine. It is generally accepted that this pathology predominantly occurs in women. The issue of excessive weight as a risk factor for CVD is still considered debatable. To the author's best knowledge, today in Ukraine, there are barely any studies that describe the relationship between CVD and obesity. Our study aims to determine the association between age, body composition, obesity and CVD in postmenopausal women. The study was conducted in D. F. Chebotarev Institute of Gerontology, National Academy of Medical Sciences of Ukraine. We have examined 96 postmenopausal women aged 46-85 years (mean age – 66.19 ± 0.96 years), who were divided into two groups depending on the presence of CVD. The women were examined by vascular surgeons. For the diagnosis of CVD, we used clinical, anatomic and pathophysiologic classifications. We also performed clinical, ultrasound and densitometry examinations. We found that the CVD frequency in postmenopausal women increased with age (from 72% in those aged 45-59 years to 84% in those aged 75-89 years). A significant correlation between the total fat mass and age was determined in postmenopausal women with CVD. We also observed a significant correlation between the lower extremities’ fat mass and age in both examined groups. A significant correlation between body mass index and age was determined only in postmenopausal women without CVD.
Abstract: In the field of machine learning, the ensemble has been employed as a common methodology to improve the performance upon multiple base classifiers. However, the true predictions are often canceled out by the false ones during consensus due to a phenomenon called “curse of correlation” which is represented as the strong interferences among the predictions produced by the base classifiers. In addition, the existing practices are still not able to effectively mitigate the problem of imbalanced classification. Based on the analysis on our experiment results, we conclude that the two problems are caused by some inherent deficiencies in the approach of consensus. Therefore, we create an enhanced ensemble algorithm which adopts a designed rank-based chain-mode consensus to overcome the two problems. In order to evaluate the proposed ensemble algorithm, we employ a well-known benchmark data set NSL-KDD (the improved version of dataset KDDCup99 produced by University of New Brunswick) to make comparisons between the proposed and 8 common ensemble algorithms. Particularly, each compared ensemble classifier uses the same 22 base classifiers, so that the differences in terms of the improvements toward the accuracy and reliability upon the base classifiers can be truly revealed. As a result, the proposed rank-based chain-mode consensus is proved to be a more effective ensemble solution than the traditional consensus approach, which outperforms the 8 ensemble algorithms by 20% on almost all compared metrices which include accuracy, precision, recall, F1-score and area under receiver operating characteristic curve.
Abstract: Organizations, including governments, generate (big) data that are high in volume, velocity, veracity, and come from a variety of sources. Public Administrations are using (big) data, implementing base registries, and enforcing data sharing within the entire government to deliver (big) data related integrated services, provision of insights to users, and for good governance. Government (Big) data ecosystem actors represent distinct entities that provide data, consume data, manipulate data to offer paid services, and extend data services like data storage, hosting services to other actors. In this research work, we perform a systematic literature review. The key objectives of this paper are to propose a robust definition of government (big) data ecosystem and a classification of government (big) data ecosystem actors and their roles. We showcase a graphical view of actors, roles, and their relationship in the government (big) data ecosystem. We also discuss our research findings. We did not find too much published research articles about the government (big) data ecosystem, including its definition and classification of actors and their roles. Therefore, we lent ideas for the government (big) data ecosystem from numerous areas that include scientific research data, humanitarian data, open government data, industry data, in the literature.
Abstract: Fourier Transform Infrared (FT-IR) spectroscopy coupled with chemometrics was used to distinguish between butter samples and non-butter samples. Further, quantification of the content of margarine in adulterated butter samples was investigated. Fingerprinting region (1400-800 cm–1) was used to develop unsupervised pattern recognition (Principal Component Analysis, PCA), supervised modeling (Soft Independent Modelling by Class Analogy, SIMCA), classification (Partial Least Squares Discriminant Analysis, PLS-DA) and regression (Partial Least Squares Regression, PLS-R) models. PCA of the fingerprinting region shows a clustering of the two sample types. All samples were classified in their rightful class by SIMCA approach; however, nine adulterated samples (between 1% and 30% w/w of margarine) were classified as belonging both at the butter class and at the non-butter one. In the two-class PLS-DA model’s (R2 = 0.73, RMSEP, Root Mean Square Error of Prediction = 0.26% w/w) sensitivity was 71.4% and Positive Predictive Value (PPV) 100%. Its threshold was calculated at 7% w/w of margarine in adulterated butter samples. Finally, PLS-R model (R2 = 0.84, RMSEP = 16.54%) was developed. PLS-DA was a suitable classification tool and PLS-R a proper quantification approach. Results demonstrate that FT-IR spectroscopy combined with PLS-R can be used as a rapid, simple and safe method to identify pure butter samples from adulterated ones and to determine the grade of adulteration of margarine in butter samples.
Abstract: Learning from very big datasets is a significant problem for most present data mining and machine learning algorithms. MicroRNA (miRNA) is one of the important big genomic and non-coding datasets presenting the genome sequences. In this paper, a hybrid method for the classification of the miRNA data is proposed. Due to the variety of cancers and high number of genes, analyzing the miRNA dataset has been a challenging problem for researchers. The number of features corresponding to the number of samples is high and the data suffer from being imbalanced. The feature selection method has been used to select features having more ability to distinguish classes and eliminating obscures features. Afterward, a Convolutional Neural Network (CNN) classifier for classification of cancer types is utilized, which employs a Genetic Algorithm to highlight optimized hyper-parameters of CNN. In order to make the process of classification by CNN faster, Graphics Processing Unit (GPU) is recommended for calculating the mathematic equation in a parallel way. The proposed method is tested on a real-world dataset with 8,129 patients, 29 different types of tumors, and 1,046 miRNA biomarkers, taken from The Cancer Genome Atlas (TCGA) database.
Abstract: A well-defined insulin resistance (IR) is one of the requirements for the good understanding and evaluation of metabolic syndrome (MetS). However, underlying causes for the development of IR are not clear. Endothelial dysfunction also participates in the pathogenesis of this disease. IR indices are being determined in various obesity groups and also in diagnosing MetS. Components of MetS have been well established and used in adult studies. However, there are some ambiguities particularly in the field of pediatrics. The aims of this study were to compare the performance of fasting blood glucose (FBG), one of MetS components, with some other IR indices and check whether FBG may be replaced by some other parameter or ratio for a better evaluation of pediatric MetS. Five-hundred and forty-nine children were involved in the study. Five groups were constituted. Groups 109, 40, 100, 166, 110, 24 children were included in normal-body mass index (N-BMI), overweight (OW), obese (OB), morbid obese (MO), MetS with two components (MetS2) and MetS with three components (MetS3) groups, respectively. Age and sex-adjusted BMI percentiles tabulated by World Health Organization were used for the classification of obesity groups. MetS components were determined. Aside from one of the MetS components-FBG, eight measures of IR [homeostatic model assessment of IR (HOMA-IR), homeostatic model assessment of beta cell function (HOMA-%β), alanine transaminase-to-aspartate transaminase ratio (ALT/AST), alanine transaminase (ALT), insulin (INS), insulin-to-FBG ratio (INS/FBG), the product of fasting triglyceride and glucose (TyG) index, McAuley index] were evaluated. Statistical analyses were performed. A p value less than 0.05 was accepted as the statistically significance degree. Mean values for BMI of the groups were 15.7 kg/m2, 21.0 kg/m2, 24.7 kg/m2, 27.1 kg/m2, 28.7 kg/m2, 30.4 kg/m2 for N-BMI, OW, OB, MO, MetS2, MetS3, respectively. Differences between the groups were significant (p < 0.001). The only exception was MetS2-MetS3 couple, in spite of an increase detected in MetS3 group. Waist-to-hip circumference ratios significantly differed only for N-BMI vs, OB, MO, MetS2; OW vs MO; OB vs MO, MetS2 couples. ALT and ALT/AST did not differ significantly among MO-MetS2-MetS3. HOMA-%β differed only between MO and MetS2. INS/FBG, McAuley index and TyG were not significant between MetS2 and MetS3. HOMA-IR and FBG were not significant between MO and MetS2. INS was the only parameter, which showed statistically significant differences between MO-MetS2, MO-MetS3, and MetS2-MetS3. In conclusion, these findings have suggested that FBG presently considered as one of the five MetS components, may be replaced by INS during the evaluation of pediatric morbid obesity and MetS.
Abstract: The impact of metabolic syndrome (MetS) on cognition and functions of the brain is being investigated. Iron deficiency and deficiencies of B9 (folate) as well as B12 (cobalamin) vitamins are best-known nutritional anemias. They are associated with cognitive disorders and learning difficulties. The antidepressant effects of vitamin D are known and the deficiency state affects mental functions negatively. The aim of this study is to investigate possible correlations of MetS with serum brain-derived neurotrophic factor (BDNF), iron, folate, cobalamin and vitamin D in pediatric patients. 30 children, whose age- and sex-dependent body mass index (BMI) percentiles vary between 85 and 15, 60 morbid obese children with above 99th percentiles constituted the study population. Anthropometric measurements were taken. BMI values were calculated. Age- and sex-dependent BMI percentile values were obtained using the appropriate tables prepared by the World Health Organization (WHO). Obesity classification was performed according to WHO criteria. Those with MetS were evaluated according to MetS criteria. Serum BDNF was determined by enzyme-linked immunosorbent assay. Serum folate was analyzed by an immunoassay analyzer. Serum cobalamin concentrations were measured using electrochemiluminescence immunoassay. Vitamin D status was determined by the measurement of 25-hydroxycholecalciferol [25-hydroxy vitamin D3, 25(OH)D] using high performance liquid chromatography. Statistical evaluations were performed using SPSS for Windows, version 16. The p values less than 0.05 were accepted as statistically significant. Although statistically insignificant, lower folate and cobalamin values were found in MO children compared to those observed for children with normal BMI. For iron and BDNF values, no alterations were detected among the groups. Significantly decreased vitamin D concentrations were noted in MO children with MetS in comparison with those in children with normal BMI (p ≤ 0.05). The positive correlation observed between iron and BDNF in normal-BMI group was not found in two MO groups. In THE MetS group, the partial correlation among iron, BDNF, folate, cobalamin, vitamin D controlling for waist circumference and BMI was r = -0.501; p ≤ 0.05. None was calculated in MO and normal BMI groups. In conclusion, vitamin D should also be considered during the assessment of pediatric MetS. Waist circumference and BMI should collectively be evaluated during the evaluation of MetS in children. Within this context, BDNF appears to be a key biochemical parameter during the examination of obesity degree in terms of mental functions, cognition and learning capacity. The association observed between iron and BDNF in children with normal BMI was not detected in MO groups possibly due to development of inflammation and other obesity-related pathologies. It was suggested that this finding may contribute to mental function impairments commonly observed among obese children.
Abstract: Fake news and false information are big challenges of all types of media, especially social media. There is a lot of false information, fake likes, views and duplicated accounts as big social networks such as Facebook and Twitter admitted. Most information appearing on social media is doubtful and in some cases misleading. They need to be detected as soon as possible to avoid a negative impact on society. The dimensions of the fake news datasets are growing rapidly, so to obtain a better result of detecting false information with less computation time and complexity, the dimensions need to be reduced. One of the best techniques of reducing data size is using feature selection method. The aim of this technique is to choose a feature subset from the original set to improve the classification performance. In this paper, a feature selection method is proposed with the integration of K-means clustering and Support Vector Machine (SVM) approaches which work in four steps. First, the similarities between all features are calculated. Then, features are divided into several clusters. Next, the final feature set is selected from all clusters, and finally, fake news is classified based on the final feature subset using the SVM method. The proposed method was evaluated by comparing its performance with other state-of-the-art methods on several specific benchmark datasets and the outcome showed a better classification of false information for our work. The detection performance was improved in two aspects. On the one hand, the detection runtime process decreased, and on the other hand, the classification accuracy increased because of the elimination of redundant features and the reduction of datasets dimensions.
Abstract: This paper presents wavelet based classification of various heart diseases. Electrocardiogram signals of different heart patients have been studied. Statistical natures of electrocardiogram signals for different heart diseases have been compared with the statistical nature of electrocardiograms for normal persons. Under this study four different heart diseases have been considered as follows: Myocardial Ischemia (MI), Congestive Heart Failure (CHF), Arrhythmia and Sleep Apnea. Statistical nature of electrocardiograms for each case has been considered in terms of kurtosis values of two types of wavelet coefficients: approximate and detail. Nine wavelet decomposition levels have been considered in each case. Kurtosis corresponding to both approximate and detail coefficients has been considered for decomposition level one to decomposition level nine. Based on significant difference, few decomposition levels have been chosen and then used for classification.
Abstract: The more an educational system knows about a learner, the more personalised interaction it can provide, which leads to better learning. However, asking a learner directly is potentially disruptive, and often ignored by learners. Especially in the booming realm of MOOC Massive Online Learning platforms, only a very low percentage of users disclose demographic information about themselves. Thus, in this paper, we aim to predict learners’ demographic characteristics, by proposing an approach using linguistically motivated Deep Learning Architectures for Learner Profiling, particularly targeting gender prediction on a FutureLearn MOOC platform. Additionally, we tackle here the difficult problem of predicting the gender of learners based on their comments only – which are often available across MOOCs. The most common current approaches to text classification use the Long Short-Term Memory (LSTM) model, considering sentences as sequences. However, human language also has structures. In this research, rather than considering sentences as plain sequences, we hypothesise that higher semantic - and syntactic level sentence processing based on linguistics will render a richer representation. We thus evaluate, the traditional LSTM versus other bleeding edge models, which take into account syntactic structure, such as tree-structured LSTM, Stack-augmented Parser-Interpreter Neural Network (SPINN) and the Structure-Aware Tag Augmented model (SATA). Additionally, we explore using different word-level encoding functions. We have implemented these methods on Our MOOC dataset, which is the most performant one comparing with a public dataset on sentiment analysis that is further used as a cross-examining for the models' results.
Abstract: Loose soils normally are of weak bearing capacity due to their structural nature. Being exposed to heavy traffic loads, they would fail in most cases. To tackle the aforementioned issue, geotechnical engineers have come up with different approaches; one of which is making use of geosynthetic-reinforced soil-aggregate systems. As these polymeric reinforcements have highlighted economic and environmentally-friendly features, they have become widespread in practice during the last decades. The present research investigates the efficiency of four different types of these reinforcements in increasing the bearing capacity of two-layered soil sections using a series California Bearing Ratio (CBR) test. The studied sections are comprised of a 10 cm-thick layer of no. 161 Firouzkooh sand (weak subgrade) and a 10 cm-thick layer of compacted aggregate materials (base course) classified as SP and GW according to the United Soil Classification System (USCS), respectively. The aggregate layer was compacted to the relative density (Dr) of 95% at the optimum water content (Wopt) of 6.5%. The applied reinforcements were including two kinds of geocomposites (type A and B), a geotextile, and a geogrid that were embedded at the interface of the lower and the upper layers of the soil-aggregate system. As the standard CBR mold was not appropriate in height for this study, the mold used for soaked CBR tests were utilized. To make a comparison between the results of stress-settlement behavior in the studied specimens, CBR values pertinent to the penetrations of 2.5 mm and 5 mm were considered. The obtained results demonstrated 21% and 24.5% increments in the amount of CBR value in the presence of geocomposite type A and geogrid, respectively. On the other hand, the effect of both geotextile and geocomposite type B on CBR values was generally insignificant in this research.
Abstract: Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.
Abstract: In the context of the development process of two assistive navigation mobile apps for blind and visually impaired people (BVI) an extensive qualitative analysis of the requirements of potential users has been conducted. The analysis was based on interviews with BVIs and aimed to elicit not only their needs with respect to autonomous navigation but also their preferences on specific features of the apps under development. The elicited requirements were structured into four main categories, namely, requirements concerning the capabilities, functionality and usability of the apps, as well as compatibility requirements with respect to other apps and services. The main categories were then further divided into nine sub-categories. This classification, along with its content, aims to become a useful tool for the researcher or the developer who is involved in the development of digital services for BVI.
Abstract: Certain systems can function well only if they recognize the sound environment as humans do. In this research, we focus on sound classification by adopting a convolutional neural network and aim to develop a method that automatically classifies various environmental sounds. Although the neural network is a powerful technique, the performance depends on the type of input data. Therefore, we propose an approach via a slice bispectrogram, which is a third-order spectrogram and is a slice version of the amplitude for the short-time bispectrum. This paper explains the slice bispectrogram and discusses the effectiveness of the derived method by evaluating the experimental results using the ESC‑50 sound dataset. As a result, the proposed scheme gives high accuracy and stability. Furthermore, some relationship between the accuracy and non-Gaussianity of sound signals was confirmed.
Abstract: Over the last years, a new paradigm known as Physical Internet has been developed, and studied in logistics management. The purpose of this global and open system is to deal with logistics grand challenge by setting up an efficient and sustainable Logistics Web. The purpose of this paper is to review scientific articles dedicated to Physical Internet topic, and to provide a clustering strategy enabling to classify the literature on the Physical Internet, to follow its evolution, as well as to criticize it. The classification is based on three factors: Logistics Web, organization, and resources. Several papers about Physical Internet have been classified and analyzed along the Logistics Web, resources and organization views at a strategic, tactical and operational level, respectively. A developed cluster analysis shows which topics of the Physical Internet that are the less covered actually. Future researches are outlined for these topics.
Abstract: Speech to text in Malay language is a system that converts Malay speech into text. The Malay language recognition system is still limited, thus, this paper aims to investigate the performance of ten Malay words obtained from the online Malay news. The methodology consists of three stages, which are preprocessing, feature extraction, and speech classification. In preprocessing stage, the speech samples are filtered using pre emphasis. After that, feature extraction method is applied to the samples using Mel Frequency Cepstrum Coefficient (MFCC). Lastly, speech classification is performed using Feedforward Neural Network (FFNN). The accuracy of the classification is further investigated based on the hidden layer size. From experimentation, the classifier with 40 hidden neurons shows the highest classification rate which is 94%.