Abstract: This paper aims to provide an interpretation of artificial neural networks (ANNs) and explore some of its implications. The interpretation views ANNs as a memory which encodes instances of experience. An experiment explores the behavior of encoding and retrieval of instances from memory. A localised representation ANN is created that allows control over encoding and retrieved memory sample size and is experimented with using the MNIST digits dataset. The relationship between input familiarity, conflict within retrieved samples, and error rates is described and demonstrated to be an effective driver for memory encoding. Results indicate that selective encoding and retrieval samples that allow detection of memory conflicts produce optimal performance, and that error rates are normally distributed with input familiarity and conflict. By using input familiarity and sample consistency to guide memory encoding, the number of encoding trials on the dataset were reduced to 18.33% of the training data while maintaining good recognition performance on the test data.
Abstract: The existing two-dimensional micro-Doppler features extraction ignores the correlation information between the spatial and temporal dimension features. For the range-Doppler map, the time dimension is introduced, and a frequency modulation continuous wave (FMCW) radar human fall detection algorithm based on time-varying range-Doppler features is proposed. Firstly, the range-Doppler sequence maps are generated from the echo signals of the continuous motion of the human body collected by the radar. Then the three-dimensional data cube composed of multiple frames of range-Doppler maps is input into the three-dimensional Convolutional Neural Network (3D CNN). The spatial and temporal features of time-varying range-Doppler are extracted by the convolution layer and pool layer at the same time. Finally, the extracted spatial and temporal features are input into the fully connected layer for classification. The experimental results show that the proposed fall detection algorithm has a detection accuracy of 95.66%.
Abstract: The paper proposes a method for constructing a self-organizing control system for unstable and deterministic chaotic processes in the class of catastrophe “hyperbolic umbilic” for objects with m-inputs and n-outputs. The self-organizing control system is investigated by the universal gradient-velocity method of Lyapunov vector-functions. The conditions for self-organization of the control system in the class of catastrophes “hyperbolic umbilic” are shown in the form of a system of algebraic inequalities that characterize the aperiodic robust stability in the stationary states of the system.
Abstract: On November 15th, 2017, a catastrophic flash flood devastated the city of Mandra in Central Greece, resulting in 24 fatalities and extensive damages to the built environment and infrastructure. It was Greece’s deadliest and most destructive flood event for the past 40 years. In this paper, we examine the consequences of this event to small and medium-sized enterprises (SMEs) operating in Mandra during the flood event, which were affected by the floodwaters to varying extents. In this context, we conducted semi-structured interviews with business owners-managers of 45 SMEs located in flood inundated areas and are still active nowadays, based on an interview guide that spanned 27 topics. The topics pertained to the disaster experience of the business and business owners-managers, knowledge and attitudes towards climate change and extreme weather, aspects of disaster preparedness and related assistance needs. Our findings reveal that the vast majority of the affected businesses experienced heavy damages in equipment and infrastructure or total destruction, which resulted in business interruption from several weeks up to several months. Assistance from relatives or friends helped for the damage repairs and business recovery, while state compensations were deemed insufficient compared to the extent of the damages. Most interviewees pinpoint flooding as one of the most critical risks, and many connect it with the climate crisis. However, they are either not willing or unable to apply property-level prevention measures in their businesses due to cost considerations or complex and cumbersome bureaucratic processes. In all cases, the business owners are fully aware of the flood hazard implications, and since the recovery from the event, they have engaged in basic mitigation measures and contingency plans in case of future flood events. Such plans include insurance contracts whenever possible (as the vast majority of the affected SMEs were uninsured at the time of the 2017 event) as well as simple relocations of critical equipment within their property. The study offers fruitful insights on latent drivers and barriers of SMEs’ resilience capacity to flash flooding. In this respect, findings such as ours, highlighting tensions that underpin behavioural responses and experiences, can feed into: a) bottom-up approaches for devising actionable and practical guidelines, manuals and/or standards on business preparedness to flooding, and, ultimately, b) policy-making for an enabling environment towards a flood-resilient SME sector.
Abstract: In recent years, SQL injection attacks have been identified as being prevalent against web applications. They affect network security and user data, which leads to a considerable loss of money and data every year. This paper presents the use of classification algorithms in machine learning using a method to classify the login data filtering inputs into "SQLi" or "Non-SQLi,” thus increasing the reliability and accuracy of results in terms of deciding whether an operation is an attack or a valid operation. A method as a Web-App is developed for auto-generated data replication to provide a twin of the targeted data structure. Shielding against SQLi attacks (WebAppShield) that verifies all users and prevents attackers (SQLi attacks) from entering and or accessing the database, which the machine learning module predicts as "Non-SQLi", has been developed. A special login form has been developed with a special instance of the data validation; this verification process secures the web application from its early stages. The system has been tested and validated, and up to 99% of SQLi attacks have been prevented.
Abstract: This paper presents the analysis of six different classes of Petri nets: fuzzy Petri nets (FPN), generalized fuzzy Petri nets (GFPN), parameterized fuzzy Petri nets (PFPN), T2GFPN, flexible generalized fuzzy Petri nets (FGFPN), binary Petri nets (BPN). These classes were simulated in the special software PNeS® for the analysis of its pros and cons on the example of models which are dedicated to the decision-making process of passenger transport logistics. The paper includes the analysis of two approaches: when input values are filled with the experts’ knowledge; when fuzzy expectations represented by output values are added to the point. These approaches fulfill the possibilities of triples of functions which are replaced with different combinations of t-/s-norms.
Abstract: Face recognition is the problem of identifying or recognizing individuals in an image. This paper investigates a possible method to bring a solution to this problem. The method proposes an amalgamation of Principal Component Analysis (PCA), K-Means clustering, and Convolutional Neural Network (CNN) for a face recognition system. It is trained and evaluated using the ORL dataset. This dataset consists of 400 different faces with 40 classes of 10 face images per class. Firstly, PCA enabled the usage of a smaller network. This reduces the training time of the CNN. Thus, we get rid of the redundancy and preserve the variance with a smaller number of coefficients. Secondly, the K-Means clustering model is trained using the compressed PCA obtained data which select the K-Means clustering centers with better characteristics. Lastly, the K-Means characteristics or features are an initial value of the CNN and act as input data. The accuracy and the performance of the proposed method were tested in comparison to other Face Recognition (FR) techniques namely PCA, Support Vector Machine (SVM), as well as K-Nearest Neighbour (kNN). During experimentation, the accuracy and the performance of our suggested method after 90 epochs achieved the highest performance: 99% accuracy F1-Score, 99% precision, and 99% recall in 463.934 seconds. It outperformed the PCA that obtained 97% and KNN with 84% during the conducted experiments. Therefore, this method proved to be efficient in identifying faces in the images.
Abstract: Time Series Forecasting (TSF) is used to predict the target variables at a future time point based on the learning from previous time points. To keep the problem tractable, learning methods use data from a fixed length window in the past as an explicit input. In this paper, we study how the performance of predictive models change as a function of different look-back window sizes and different amounts of time to predict into the future. We also consider the performance of the recent attention-based transformer models, which had good success in the image processing and natural language processing domains. In all, we compare four different deep learning methods (Recurrent Neural Network (RNN), Long Short-term Memory (LSTM), Gated Recurrent Units (GRU), and Transformer) along with a baseline method. The dataset (hourly) we used is the Beijing Air Quality Dataset from the website of University of California, Irvine (UCI), which includes a multivariate time series of many factors measured on an hourly basis for a period of 5 years (2010-14). For each model, we also report on the relationship between the performance and the look-back window sizes and the number of predicted time points into the future. Our experiments suggest that Transformer models have the best performance with the lowest Mean Absolute Errors (MAE = 14.599, 23.273) and Root Mean Square Errors (RSME = 23.573, 38.131) for most of our single-step and multi-steps predictions. The best size for the look-back window to predict 1 hour into the future appears to be one day, while 2 or 4 days perform the best to predict 3 hours into the future.
Abstract: Digital magnitude comparators based on Gate Diffusion Input (GDI) implementation technique are high speed and area-efficient, and they consume less power as compared to other implementation techniques. However, they are less efficient for some logic gates and have no full voltage swing. In this paper, we made a performance comparison between the GDI implementation technique and other implementation methods, such as Static CMOS, Pass Transistor Logic (PTL), and Transmission Gate (TG) in 90 nm, 120 nm, and 180 nm CMOS technologies using BSIM4 MOS model. We proposed a methodology (hybrid implementation) of implementing digital magnitude comparators which significantly improved the power, speed, area, and voltage swing requirements. Simulation results revealed that the hybrid implementation of digital magnitude comparators show a 10.84% (power dissipation), 41.6% (propagation delay), 47.95% (power-delay product (PDP)) improvement compared to the usual GDI implementation method. We used Microwind & Dsch Version 3.5 as well as the Tanner EDA 16.0 tools for simulation purposes.
Abstract: This work describes a system that uses electromyography (EMG) signals obtained from muscle sensors and an Artificial Neural Network (ANN) for signal classification and pattern recognition that is used to control a small unmanned aerial vehicle using specific arm movements. The main objective of this endeavor is the development of an intelligent interface that allows the user to control the flight of a drone beyond direct manual control. The sensor used were the MyoWare Muscle sensor which contains two EMG electrodes used to collect signals from the posterior (extensor) and anterior (flexor) forearm, and the bicep. The collection of the raw signals from each sensor was performed using an Arduino Uno. Data processing algorithms were developed with the purpose of classifying the signals generated by the arm’s muscles when performing specific movements, namely: flexing, resting, and motion of the arm. With these arm motions roll control of the drone was achieved. MATLAB software was utilized to condition the signals and prepare them for the classification. To generate the input vector for the ANN and perform the classification, the root mean square and the standard deviation were processed for the signals from each electrode. The neuromuscular information was trained using an ANN with a single 10 neurons hidden layer to categorize the four targets. The result of the classification shows that an accuracy of 97.5% was obtained. Afterwards, classification results are used to generate the appropriate control signals from the computer to the drone through a Wi-Fi network connection. These procedures were successfully tested, where the drone responded successfully in real time to the commanded inputs.
Abstract: With the advent of three-dimension (3D) technology, there are lots of research in converting 2D images to 3D images. The main difference between 2D and 3D is the visual illusion of depth in 3D images. In the recent era, there are more depth estimation techniques. The objective of this paper is to convert 2D images to 3D images with less computation time. For this, the input image is divided into blocks from which the depth information is obtained. Having the depth information, a depth map is generated. Then the 3D image is warped using the original image and the depth map. The proposed method is tested on Make3D dataset and NYU-V2 dataset. The experimental results are compared with other recent methods. The proposed method proved to work with less computation time and good accuracy.
Abstract: The Purdue University Research Reactor-1 (PUR-1) is a 10 kWth pool-type research reactor located at Purdue University’s West Lafayette campus. The reactor was recently upgraded to use entirely digital instrumentation and control systems. However, currently, there is no automated control system to regulate the power in the reactor. We propose a fuzzy logic controller as a form of digital twin to complement the existing digital instrumentation system to monitor and stabilize power control using existing experimental data. This work assesses the feasibility of a power controller based on a Fuzzy Rule-Based System (FRBS) by modelling and simulation with a MATLAB algorithm. The controller uses power error and reactor period as inputs and generates reactivity insertion as output. The reactivity insertion is then converted to control rod height using a logistic function based on information from the recorded experimental reactor control rod data. To test the capability of the proposed fuzzy controller, a point-kinetic reactor model is utilized based on the actual PUR-1 operation conditions and a Monte Carlo N-Particle simulation result of the core to numerically compute the neutronics parameters of reactor behavior. The Point Kinetic Equation (PKE) was employed to model dynamic characteristics of the research reactor since it explains the interactions between the spatial and time varying input and output variables efficiently. The controller is demonstrated computationally using various cases: startup, power maneuver, and shutdown. From the test results, it can be proved that the implemented fuzzy controller can satisfactorily regulate the reactor power to follow demand power without compromising nuclear safety measures.
Abstract: Against the background of President Xi Jinping and Vladimir Putin hailing economics at the heart of their bilateral relationship, Putin conveyed that economic cooperation and trade are important to the revitalization of China and Russia. Likewise, President Xi Jinping noted that Russia and China bask in unprecedented great confidence and collaboration. Such amicable relationships may translate into enhanced economic cooperation and Chinese foreign investment in Russia. This paper employed a retrospective review and attempted to analyze Chinese agricultural engagement in Russian Far East, in addition to examining what was dubbed as land grabbing in Russia, probing into the political, economics, and human dimensions on the land debate. Such an analysis is significant as it contributes to the literature on foreign investment in agriculture.
Abstract: The digital control system of mobile robots (MR) control is considered. It is shown that sequential interpretation of control algorithm operators, unfolding in physical time, suggests the occurrence of time delays between inputting data from sensors and outputting data to actuators. Another destabilizing control factor is presence of backlash in the joints of an actuator with an executive unit. Complex model of control system, which takes into account the dynamics of the MR, the dynamics of the digital controller and backlash in actuators, is worked out. The digital controller model is divided into two parts: the first part describes the control law embedded in the controller in the form of a control program that realizes a polling procedure when organizing transactions to sensors and actuators. The second part of the model describes the time delays that occur in the Von Neumann-type controller when processing data. To estimate time intervals, the algorithm is represented in the form of an ergodic semi-Markov process. For an ergodic semi-Markov process of common form, a method is proposed for estimation a wandering time from one arbitrary state to another arbitrary state. Example shows how the backlash and time delays affect the quality characteristics of the MR control system functioning.
Abstract: Composting is one of the conventional techniques adopted for organic waste management but the practice is very limited in emerging cities despite that most of the waste generated is organic. This paper aims to examine the viability of composting for organic waste management in the emerging city of Addis Ababa, Ethiopia by addressing the composting practice, quality of compost and application of compost in urban agriculture. The study collects data using compost laboratory testing and urban farm households’ survey and uses descriptive analysis on the state of compost production and application, physicochemical analysis of the compost samples, and regression analysis on the urban farmer’s willingness to pay for compost. The findings of the study indicated that there is composting practice at a small scale, most of the producers use unsorted feedstock materials, aerobic composting is dominantly used and the maturation period ranged from four to 10 weeks. The carbon content of the compost ranges from 30.8 to 277.1 due to the type of feedstock applied and this surpasses the ideal proportions for C:N ratio. The total nitrogen, pH, organic matter and moisture content are relatively optimal. The levels of heavy metals measured for Mn, Cu, Pb, Cd and Cr6+ in the compost samples are also insignificant. In the urban agriculture sector, chemical fertilizer is the dominant type of soil input in crop productions but vegetable producers use a combination of both fertilizer and other organic inputs including compost. The willingness to pay for compost depends on income, household size, gender, type of soil inputs, monitoring soil fertility, the main product of the farm, farming method and farm ownership. Finally, this study recommends the need for collaboration among stakeholders along the value chain of waste, awareness creation on the benefits of composting and addressing challenges faced by both compost producers and users.
Abstract: Anticholinergic medication has been associated with events such as falls, delirium, and cognitive impairment in older patients. To further assess this, anticholinergic burden scores have been developed to quantify risk. A risk model based on clustering was deployed in a healthcare management system to cluster patients into multiple risk groups according to anticholinergic burden scores of multiple medicines prescribed to patients to facilitate clinical decision-making. To do so, anticholinergic burden scores of drugs were extracted from the literature which categorizes the risk on a scale of 1 to 3. Given the patients’ prescription data on the healthcare database, a weighted anticholinergic risk score was derived per patient based on the prescription of multiple anticholinergic drugs. This study was conducted on 300,000 records of patients currently registered with a major regional UK-based healthcare provider. The weighted risk scores were used as inputs to an unsupervised learning algorithm (mean-shift clustering) that groups patients into clusters that represent different levels of anticholinergic risk. This work evaluates the association between the average risk score and measures of socioeconomic status (index of multiple deprivation) and health (index of health and disability). The clustering identifies a group of 15 patients at the highest risk from multiple anticholinergic medication. Our findings show that this group of patients is located within more deprived areas of London compared to the population of other risk groups. Furthermore, the prescription of anticholinergic medicines is more skewed to female than male patients, suggesting that females are more at risk from this kind of multiple medication. The risk may be monitored and controlled in a healthcare management system that is well-equipped with tools implementing appropriate techniques of artificial intelligence.
Abstract: This study aims to provide sub-hourly streamflow predictions and associated rating curves for small catchments of intermittent and torrential flow regime characterized by flash floods occurring especially during April and November. The methodology entails two lumped conceptual hydrological models which work in series. The total model is based upon eleven parameters and shows good flexibility in handling different input sets. Runoff Coefficient has contributed to improving the model’s performances and has been treated as an additional parameter; while Sensitivity Analysis has highlighted how slight changes in the model’s input can lead to changes in model’s output. The adopted procedure is steady and useful to give very practical engineering information at the expense of a parsimonious request both in input data and in the number of adopted parameters. According to the obtained results, the authors encourage the test of this combined procedure on different hydrological scenarios in order to provide information for poorly monitored catchments and not updated sites.
Abstract: This article describes the Random Subspace Neural Classifier (RSC) for the recognition of meteors in the night sky. We used images of meteors entering the atmosphere at night between 8:00 p.m.-5: 00 a.m. The objective of this project is to classify meteor and star images (with stars as the image background). The monitoring of the sky and the classification of meteors are made for future applications by scientists. The image database was collected from different websites. We worked with RGB-type images with dimensions of 220x220 pixels stored in the BitMap Protocol (BMP) format. Subsequent window scanning and processing were carried out for each image. The scan window where the characteristics were extracted had the size of 20x20 pixels with a scanning step size of 10 pixels. Brightness, contrast and contour orientation histograms were used as inputs for the RSC. The RSC worked with two classes and classified into: 1) with meteors and 2) without meteors. Different tests were carried out by varying the number of training cycles and the number of images for training and recognition. The percentage error for the neural classifier was calculated. The results show a good RSC classifier response with 89% correct recognition. The results of these experiments are presented and discussed.
Abstract: Price fluctuation contract is imperative and of paramount essence in the construction industry as it provides adequate relief and cushioning for changes in the prices of input resources during construction. As a result, several methods have been devised to better help in arriving at fair recompense in the event of price changes. However, stakeholders often appear not to be satisfied with the existing methods of fluctuation evaluation, ostensibly because of the challenges associated with them. The aim of this study was to identify the challenges to usage of building construction cost indices in Ghana. Data were gathered from contractors and quantity surveying firms. The study utilized survey questionnaire approach to elicit responses from the contractors and the consultants. Data gathered were analyzed scientifically, using the Relative Importance Index (RII) to rank the problems associated with the existing methods. The findings revealed the following among others: late release of data; inadequate recovery of costs; and work items of interest not included in the published indices as the main challenges of the existing methods. Findings provided useful lessons for policy makers and practitioners in decision making towards the usage and improvement of available indices.
Abstract: Early detection of anomalies in data centers is important to reduce downtimes and the costs of periodic maintenance. However, there is little research on this topic and even fewer on the fusion of sensor data for the detection of abnormal events. The goal of this paper is to propose a method for anomaly detection in data centers by combining sensor data (temperature, humidity, power) and deep learning models. The model described in the paper uses one autoencoder per sensor to reconstruct the inputs. The auto-encoders contain Long-Short Term Memory (LSTM) layers and are trained using the normal samples of the relevant sensors selected by correlation analysis. The difference signal between the input and its reconstruction is then used to classify the samples using feature extraction and a random forest classifier. The data measured by the sensors of a data center between January 2019 and May 2020 are used to train the model, while the data between June 2020 and May 2021 are used to assess it. Performances of the model are assessed a posteriori through F1-score by comparing detected anomalies with the data center’s history. The proposed model outperforms the state-of-the-art reconstruction method, which uses only one autoencoder taking multivariate sequences and detects an anomaly with a threshold on the reconstruction error, with an F1-score of 83.60% compared to 24.16%.