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: 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: 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: 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: Human pose estimation and tracking are to accurately identify and locate the positions of human joints in the video. It is a computer vision task which is of great significance for human motion recognition, behavior understanding and scene analysis. There has been remarkable progress on human pose estimation in recent years. However, more researches are needed for human pose tracking especially for online tracking. In this paper, a framework, called PoseSRPN, is proposed for online single-person pose estimation and tracking. We use Siamese network attaching a pose estimation branch to incorporate Single-person Pose Tracking (SPT) and Visual Object Tracking (VOT) into one framework. The pose estimation branch has a simple network structure that replaces the complex upsampling and convolution network structure with deconvolution. By augmenting the loss of fully convolutional Siamese network with the pose estimation task, pose estimation and tracking can be trained in one stage. Once trained, PoseSRPN only relies on a single bounding box initialization and producing human joints location. The experimental results show that while maintaining the good accuracy of pose estimation on COCO and PoseTrack datasets, the proposed method achieves a speed of 59 frame/s, which is superior to other pose tracking frameworks.
Abstract: Visual search and identification of immunohistochemically stained tissue of meningioma was performed manually in pathologic laboratories to detect and diagnose the cancers type of meningioma. This task is very tedious and time-consuming. Moreover, because of cell's complex nature, it still remains a challenging task to segment cells from its background and analyze them automatically. In this paper, we develop and test a computerized scheme that can automatically identify cells in microscopic images of meningioma and classify them into positive (proliferative) and negative (normal) cells. Dataset including 150 images are used to test the scheme. The scheme uses Fuzzy C-means algorithm as a color clustering method based on perceptually uniform hue, saturation, value (HSV) color space. Since the cells are distinguishable by the human eye, the accuracy and stability of the algorithm are quantitatively compared through application to a wide variety of real images.
Abstract: Data assets protection is a crucial issue in the
cybersecurity field. Companies use logical access control tools to
vault their information assets and protect them against external
threats, but they lack solutions to counter insider threats. Nowadays,
insider threats are the most significant concern of security analysts.
They are mainly individuals with legitimate access to companies
information systems, which use their rights with malicious intents.
In several fields, behavior anomaly detection is the method used by
cyber specialists to counter the threats of user malicious activities
effectively. In this paper, we present the step toward the construction
of a user and entity behavior analysis framework by proposing a
behavior anomaly detection model. This model combines machine
learning classification techniques and graph-based methods, relying
on linear algebra and parallel computing techniques. We show the
utility of an ensemble learning approach in this context. We present
some detection methods tests results on an representative access
control dataset. The use of some explored classifiers gives results
up to 99% of accuracy.
Abstract: This paper compares different types of profitability measures of cooperative banks from two developed regions: the European Union and the United States of America together with Canada. We created balanced dataset of more than 200 cooperative banks covering 2011-2016 period. We made series of tests and run Random Effects estimation on panel data. We found that American and Canadian cooperatives are more profitable in terms of return on assets (ROA) and return on equity (ROE). There is no significant difference in net interest margin (NIM). Our results show that the North American cooperative banks accommodated better to the current market environment.
Abstract: Prediction of wall shear stress in a rectangular channel, with non-homogeneous roughness distribution, was studied. Estimation of shear stress is an important subject in hydraulic engineering, since it affects the flow structure directly. In this study, the Genetic Algorithm Artificial (GAA) neural network is introduced as a hybrid methodology of the Artificial Neural Network (ANN) and modified Genetic Algorithm (GA) combination. This GAA method was employed to predict the wall shear stress. Various input combinations and transfer functions were considered to find the most appropriate GAA model. The results show that the proposed GAA method could predict the wall shear stress of open channels with high accuracy, by Root Mean Square Error (RMSE) of 0.064 in the test dataset. Thus, using GAA provides an accurate and practical simple-to-use equation.
Abstract: The neural network quantization is highly desired
procedure to perform before running neural networks on mobile
devices. Quantization without fine-tuning leads to accuracy drop of
the model, whereas commonly used training with quantization is done
on the full set of the labeled data and therefore is both time- and
resource-consuming. Real life applications require simplification and
acceleration of quantization procedure that will maintain accuracy of
full-precision neural network, especially for modern mobile neural
network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with
quantization procedure by introducing the trained scale factors for
discretization thresholds that are separate for each filter. Using the
proposed technique, we quantize the modern mobile architectures of
neural networks with the set of train data of only ∼ 10% of the
total ImageNet 2012 sample. Such reduction of train dataset size and
small number of trainable parameters allow to fine-tune the network
for several hours while maintaining the high accuracy of quantized
model (accuracy drop was less than 0.5%). Ready-for-use models and
code are available in the GitHub repository.
Abstract: Iraq is a middle income country where the road safety issue is considered one of the leading causes of deaths. To control the road risk issue, the Iraqi Ministry of Planning, General Statistical Organization started to organise a collection system of traffic accidents data with details related to their causes and severity. These data are published as an annual report. In this paper, a review of the available crash data in Iraq will be presented. The available data represent the rate of accidents in aggregated level and classified according to their types, road users’ details, and crash severity, type of vehicles, causes and number of causalities. The review is according to the types of models used in road safety studies and research, and according to the required road safety data in the road constructions tasks. The available data are also compared with the road safety dataset published in the United Kingdom as an example of developed country. It is concluded that the data in Iraq are suitable for descriptive and exploratory models, aggregated level comparison analysis, and evaluation and monitoring the progress of the overall traffic safety performance. However, important traffic safety studies require disaggregated level of data and details related to the factors of the likelihood of traffic crashes. Some studies require spatial geographic details such as the location of the accidents which is essential in ranking the roads according to their level of safety, and name the most dangerous roads in Iraq which requires tactic plan to control this issue. Global Road safety agencies interested in solve this problem in low and middle-income countries have designed road safety assessment methodologies which are basing on the road attributes data only. Therefore, in this research it is recommended to use one of these methodologies.
Abstract: Seizure is the main factor that affects the quality of life of epileptic patients. The diagnosis of epilepsy, and hence the identification of epileptogenic zone, is commonly made by using continuous Electroencephalogram (EEG) signal monitoring. Seizure identification on EEG signals is made manually by epileptologists and this process is usually very long and error prone. The aim of this paper is to describe an automated method able to detect seizures in EEG signals, using knowledge discovery in database process and data mining methods and algorithms, which can support physicians during the seizure detection process. Our detection method is based on Artificial Neural Network classifier, trained by applying the multilayer perceptron algorithm, and by using a software application, called Training Builder that has been developed for the massive extraction of features from EEG signals. This tool is able to cover all the data preparation steps ranging from signal processing to data analysis techniques, including the sliding window paradigm, the dimensionality reduction algorithms, information theory, and feature selection measures. The final model shows excellent performances, reaching an accuracy of over 99% during tests on data of a single patient retrieved from a publicly available EEG dataset.
Abstract: Artificial Intelligence (AI) has the potential to transform
manufacturing by improving shop floor processes such as production,
maintenance and quality. However, industrial datasets are notoriously
difficult to extract in a real-time, streaming fashion thus, negating
potential AI benefits. The main example is some specialized industrial
controllers that are operated by custom software which complicates
the process of connecting them to an Information Technology (IT)
based data acquisition network. Security concerns may also limit
direct physical access to these controllers for data acquisition.
To connect the Operational Technology (OT) data stored in these
controllers to an AI application in a secure, reliable and available
way, we propose a novel Industrial IoT (IIoT) solution in this paper.
In this solution, we demonstrate how video cameras can be installed
in a factory shop floor to continuously obtain images of the controller
HMIs. We propose image pre-processing to segment the HMI into
regions of streaming data and regions of fixed meta-data. We then
evaluate the performance of multiple Optical Character Recognition
(OCR) technologies such as Tesseract and Google vision to recognize
the streaming data and test it for typical factory HMIs and realistic
lighting conditions. Finally, we use the meta-data to match the OCR
output with the temporal, domain-dependent context of the data to
improve the accuracy of the output. Our IIoT solution enables reliable
and efficient data extraction which will improve the performance of
subsequent AI applications.
Abstract: Tracking sports players is a widely challenging
scenario, specially in single-feed videos recorded in tight courts,
where cluttering and occlusions cannot be avoided. This paper
presents an analysis of several geometric and semantic visual features
to detect and track basketball players. An ablation study is carried
out and then used to remark that a robust tracker can be built with
Deep Learning features, without the need of extracting contextual
ones, such as proximity or color similarity, nor applying camera
stabilization techniques. The presented tracker consists of: (1) a
detection step, which uses a pretrained deep learning model to
estimate the players pose, followed by (2) a tracking step, which
leverages pose and semantic information from the output of a
convolutional layer in a VGG network. Its performance is analyzed
in terms of MOTA over a basketball dataset with more than 10k
instances.
Abstract: This research presents an analytical model for the development of an energy harvester with piezoelectric rings stacked at the boundary of the structure based on the Adomian decomposition method. The model is applied to geometrically non-uniform beams to derive the steady-state dynamic response of the structure subjected to base motion excitation and efficiently harvest the subsequent vibrational energy. The in-plane polarization of the piezoelectric rings is employed to enhance the electrical power output. A parametric study for the proposed energy harvester with various design parameters is done to prepare the dataset required for optimization. Finally, simulation-based optimization technique helps to find the optimum structural design with maximum efficiency. To solve the optimization problem, an artificial neural network is first trained to replace the simulation model, and then, a genetic algorithm is employed to find the optimized design variables. Higher geometrical non-uniformity and length of the beam lowers the structure natural frequency and generates a larger power output.
Abstract: In data-driven prognostic methods, the prediction
accuracy of the estimation for remaining useful life of bearings
mainly depends on the performance of health indicators, which
are usually fused some statistical features extracted from vibrating
signals. However, the existing health indicators have the following
two drawbacks: (1) The differnet ranges of the statistical features
have the different contributions to construct the health indicators,
the expert knowledge is required to extract the features. (2) When
convolutional neural networks are utilized to tackle time-frequency
features of signals, the time-series of signals are not considered.
To overcome these drawbacks, in this study, the method combining
convolutional neural network with gated recurrent unit is proposed to
extract the time-frequency image features. The extracted features are
utilized to construct health indicator and predict remaining useful life
of bearings. First, original signals are converted into time-frequency
images by using continuous wavelet transform so as to form the
original feature sets. Second, with convolutional and pooling layers
of convolutional neural networks, the most sensitive features of
time-frequency images are selected from the original feature sets.
Finally, these selected features are fed into the gated recurrent unit
to construct the health indicator. The results state that the proposed
method shows the enhance performance than the related studies which
have used the same bearing dataset provided by PRONOSTIA.
Abstract: In order to reduce the number of deaths due to heart
problems, we propose the use of Hierarchical Temporal Memory
Algorithm (HTM) which is a real time anomaly detection algorithm.
HTM is a cortical learning algorithm based on neocortex used for
anomaly detection. In other words, it is based on a conceptual theory
of how the human brain can work. It is powerful in predicting unusual
patterns, anomaly detection and classification. In this paper, HTM
have been implemented and tested on ECG datasets in order to detect
cardiac anomalies. Experiments showed good performance in terms
of specificity, sensitivity and execution time.
Abstract: The objective of this paper is to investigate the relationship between the increase of accessibility brought by high speed rail (HSR) systems and the tourism market in Italy. The impacts of HSR projects on tourism can be quantified in different ways. In this manuscript, an empirical analysis has been carried out with the aid of a dataset containing information both on tourism and transport for 99 Italian provinces during the 2006-2016 period. Panel data regression models have been considered, since they allow modelling a wide variety of correlation patterns. Results show that HSR has an impact on the choice of a given destination for Italian tourists while the presence of a second level hub mainly affects foreign tourists. Attraction variables are also significant for both categories and the variables concerning security, such as number of crimes registered in a given destination, have a negative impact on the choice of a destination.
Abstract: It is commonly observed that aftershocks follow the mainshock. Aftershocks continue over a period of time with a decreasing frequency and typically there is not sufficient time for repair and retrofit between a mainshock–aftershock sequence. Usually, aftershocks are smaller in magnitude; however, aftershock ground motion characteristics such as the intensity and duration can be greater than the mainshock due to the changes in the earthquake mechanism and location with respect to the site. The seismic performance of slopes is typically evaluated based on the sliding displacement predicted to occur along a critical sliding surface. Various empirical models are available that predict sliding displacement as a function of seismic loading parameters, ground motion parameters, and site parameters but these models do not include the aftershocks. The seismic risks associated with the post-mainshock slopes ('damaged slopes') subjected to aftershocks is significant. This paper extends the empirical sliding displacement models for flexible slopes subjected to earthquake mainshock-aftershock sequences (a multi hazard approach). A dataset was developed using 144 pairs of as-recorded mainshock-aftershock sequences using the Pacific Earthquake Engineering Research Center (PEER) database. The results reveal that the combination of mainshock and aftershock increases the seismic demand on slopes relative to the mainshock alone; thus, seismic risks are underestimated if aftershocks are neglected.
Abstract: Digital Terrain Model (DTM) is a digital numerical representation of the Earth's surface. DTMs have been applied to a diverse field of tasks, such as urban planning, military, glacier mapping, disaster management. In the expression of the Earth' surface as a mathematical model, an infinite number of point measurements are needed. Because of the impossibility of this case, the points at regular intervals are measured to characterize the Earth's surface and DTM of the Earth is generated. Hitherto, the classical measurement techniques and photogrammetry method have widespread use in the construction of DTM. At present, RADAR, LiDAR, and stereo satellite images are also used for the construction of DTM. In recent years, especially because of its superiorities, Airborne Light Detection and Ranging (LiDAR) has an increased use in DTM applications. A 3D point cloud is created with LiDAR technology by obtaining numerous point data. However recently, by the development in image mapping methods, the use of unmanned aerial vehicles (UAV) for photogrammetric data acquisition has increased DTM generation from image-based point cloud. The accuracy of the DTM depends on various factors such as data collection method, the distribution of elevation points, the point density, properties of the surface and interpolation methods. In this study, the random data reduction method is compared for DTMs generated from image based point cloud data. The original image based point cloud data set (100%) is reduced to a series of subsets by using random algorithm, representing the 75, 50, 25 and 5% of the original image based point cloud data set. Over the ANS campus of Afyon Kocatepe University as the test area, DTM constructed from the original image based point cloud data set is compared with DTMs interpolated from reduced data sets by Kriging interpolation method. The results show that the random data reduction method can be used to reduce the image based point cloud datasets to 50% density level while still maintaining the quality of DTM.