International Science Index
International Journal of Computer and Systems Engineering
Application of Deep Learning in Colorization of LiDAR-Derived Intensity Images
Most aerial LiDAR systems have accompanying aerial cameras in order to capture not only the terrain of the surveyed area but also its true-color appearance. However, the presence of atmospheric clouds, poor lighting conditions, and aerial camera problems during an aerial survey may cause absence of aerial photographs. These leave areas having terrain information but lacking aerial photographs. Intensity images can be derived from LiDAR data but they are only grayscale images. A deep learning model is developed to create a complex function in a form of a deep neural network relating the pixel values of LiDAR-derived intensity images and true-color images. This complex function can then be used to predict the true-color images of a certain area using intensity images from LiDAR data. The predicted true-color images do not necessarily need to be accurate compared to the real world. They are only intended to look realistic so that they can be used as base maps.
Multi-Temporal Cloud Detection and Removal in Satellite Imagery for Land Resources Investigation
Clouds are inevitable contaminants in optical satellite imagery, and prevent the satellite imaging systems from acquiring clear view of the earth surface. The presence of clouds in satellite imagery bring negative influences for remote sensing land resources investigation. As a consequence, detecting the locations of clouds in satellite imagery is an essential preprocessing step, and further remove the existing clouds is crucial for the application of imagery. In this paper, a multi-temporal based satellite imagery cloud detection and removal method is proposed, which will be used for large-scale land resource investigation. The proposed method is mainly composed of four steps. First, cloud masks are generated for cloud contaminated images by single temporal cloud detection based on multiple spectral features. Then, a cloud-free reference image of target areas is synthesized by weighted averaging time-series images in which cloud pixels are ignored. Thirdly, the refined cloud detection results are acquired by multi-temporal analysis based on the reference image. Finally, detected clouds are removed via multi-temporal linear regression. The results of a case application in Hubei province indicate that the proposed multi-temporal cloud detection and removal method is effective and promising for large-scale land resource investigation.
A Geosynchronous Orbit Synthetic Aperture Radar Simulator for Moving Ship Targets
Ship detection is of great significance for both military and civilian applications. Synthetic aperture radar (SAR) with all-day, all-weather, ultra-long-range characteristics, has been used widely. In view of the low time resolution of low orbit SAR and the needs for high time resolution SAR data, GEO (Geosynchronous orbit) SAR is getting more and more attention. Since GEO SAR has short revisiting period and large coverage area, it is expected to be well utilized in marine ship targets monitoring. However, the height of the orbit increases the time of integration by almost two orders of magnitude. For moving marine vessels, the utility and efficacy of GEO SAR are still not sure. This paper attempts to find the feasibility of GEO SAR by giving a GEO SAR simulator of moving ships. This presented GEO SAR simulator is a kind of geometrical-based radar imaging simulator, which focus on geometrical quality rather than high radiometric. Inputs of this simulator are 3D ship model (.obj format, produced by most 3D design software, such as 3D Max), ship's velocity, and the parameters of satellite orbit and SAR platform. Its outputs are simulated GEO SAR raw signal data and SAR image. This simulating process is accomplished by the following four steps. (1) Reading 3D model, including the ship rotations (pitch, yaw, and roll) and velocity (speed and direction) parameters, extract information of those little primitives (triangles) which is visible from the SAR platform. (2) Computing the radar scattering from the ship with physical optics (PO) method. In this step, the vessel is sliced into many little rectangles primitives along the azimuth. The radiometric calculation of each primitive is carried out separately. Since this simulator only focuses on the complex structure of ships, only single-bounce reflection and double-bounce reflection are considered. (3) Generating the raw data with GEO SAR signal modeling. Since the normal ‘stop and go’ model is not available for GEO SAR, the range model should be reconsidered. (4) At last, generating GEO SAR image with improved Range Doppler method. Numerical simulation of fishing boat and cargo ship will be given. GEO SAR images of different posture, velocity, satellite orbit, and SAR platform will be simulated. By analyzing these simulated results, the effectiveness of GEO SAR for the detection of marine moving vessels is evaluated.
Computerized Scoring System: A Stethoscope to Understand Consumer's Emotion through His or Her Feedback
Most companies pay careful attention to consumer feedback collection, so it is popular to find the ‘feedback’ button of all kinds of mobile apps. Yet it is much more changeling to analyze these feedback texts and to catch the true feelings of a consumer regarding either a problem or a complimentary of consumers who hands out the feedback. Especially to the Chinese content, it is possible that; in one context the Chinese feedback expresses positive feedback, but in the other context, the same Chinese feedback may be a negative one. For example, in Chinese, the feedback 'operating with loudness' works well with both refrigerator and stereo system. Apparently, this feedback towards a refrigerator shows negative feedback; however, the same feedback is positive towards a stereo system. By introducing Bradley, M. and Lang, P.'s Affective Norms for English Text (ANET) theory and Bucci W.’s Referential Activity (RA) theory, we, usability researchers at Pingan, are able to decipher the feedback and to find the hidden feelings behind the content. We subtract 2 disciplines ‘valence’ and ‘dominance’ out of 3 of ANET and 2 disciplines ‘concreteness’ and ‘specificity’ out of 4 of RA to organize our own rating system with a scale of 1 to 5 points. This rating system enables us to judge the feelings/emotion behind each feedback, and it works well with both single word/phrase and a whole paragraph. The result of the rating reflects the strength of the feeling/emotion of the consumer when he/she is typing the feedback. In our daily work, we first require a consumer to answer the net promoter score (NPS) before writing the feedback, so we can determine the feedback is positive or negative. Secondly, we code the feedback content according to company problematic list, which contains 200 problematic items. In this way, we are able to collect the data that how many feedbacks left by the consumer belong to one typical problem. Thirdly, we rate each feedback based on the rating system mentioned above to illustrate the strength of the feeling/emotion when our consumer writes the feedback. In this way, we actually obtain two kinds of data 1) the portion, which means how many feedbacks are ascribed into one problematic item and 2) the severity, how strong the negative feeling/emotion is when the consumer is writing this feedback. By crossing these two, and introducing the portion into X-axis and severity into Y-axis, we are able to find which typical problem gets the high score in both portion and severity. The higher the score of a problem has, the more urgent a problem is supposed to be solved as it means more people write stronger negative feelings in feedbacks regarding this problem. Moreover, by introducing hidden Markov model to program our rating system, we are able to computerize the scoring system and are able to process thousands of feedback in a short period of time, which is efficient and accurate enough for the industrial purpose.
Microservices: A Review of the Costs and the Benefits
This work is concerned with analyzing the merits and the costs of microservices. Following the hype associated with a new technology may result in more problems rather than solutions. The microservices approach offers many benefits in terms of flexibility and scalability. But rushing to microservices without balancing the situation may add unnecessary complexity, and there is a possibility that the costs may outweigh the benefits. The key question is: are Microservices in their current form solving more problems than they create? In this paper, we analyze the benefits and the costs of switching to microservices. The aim is to support the decision of whether to move to microservices or not based on the evaluation of the advantages and the disadvantages. The main contribution of this work is the provision of a clear picture of the costs and benefits of the technology to help decide if and when a switch to Microservices is the correct choice.
Monitoring Future Climate Changes Pattern over Major Cities in Ghana Using Coupled Modeled Intercomparison Project Phase 5, Support Vector Machine, and Random Forest Modeling
Climate change is recently gaining the attention of many countries across the world. Climate change, which is also known as global warming, referring to the increasing in average surface temperature has been a concern to the Environmental Protection Agency of Ghana. Recently, Ghana has become vulnerable to the effect of the climate change as a result of the dependence of the majority of the population on agriculture. The clearing down of trees to grow crops and burning of charcoal in the country has been a contributing factor to the rise in temperature nowadays in the country as a result of releasing of carbon dioxide and greenhouse gases into the air. Recently, petroleum stations across the cities have been on fire due to this climate changes and which have position Ghana in a way not able to withstand this climate event. As a result, the significant of this research paper is to project how the rise in the average surface temperature will be like at the end of the mid-21st century when agriculture and deforestation are allowed to continue for some time in the country. This study uses the Coupled Modeled Intercomparison Project phase 5 (CMIP5) experiment RCP 8.5 model output data to monitor the future climate changes from 2041-2050, at the end of the mid-21st century over the ten (10) major cities (Accra, Bolgatanga, Cape Coast, Koforidua, Kumasi, Sekondi-Takoradi, Sunyani, Ho, Tamale, Wa) in Ghana. In the models, Support Vector Machine and Random forest, where the cities as a function of heat wave metrics (minimum temperature, maximum temperature, mean temperature, heat wave duration and number of heat waves) assisted to provide more than 50% accuracy to predict and monitor the pattern of the surface air temperature. The findings identified were that the near-surface air temperature will rise between 1°C-2°C (degrees Celsius) over the coastal cities (Accra, Cape Coast, Sekondi-Takoradi). The temperature over Kumasi, Ho and Sunyani by the end of 2050 will rise by 1°C. In Koforidua, it will rise between 1°C-2°C. The temperature will rise in Bolgatanga, Tamale and Wa by 0.5°C by 2050. This indicates how the coastal and the southern part of the country are becoming hotter compared with the north, even though the northern part is the hottest. During heat waves from 2041-2050, Bolgatanga, Tamale, and Wa will experience the highest mean daily air temperature between 34°C-36°C. Kumasi, Koforidua, and Sunyani will experience about 34°C. The coastal cities (Accra, Cape Coast, Sekondi-Takoradi) will experience below 32°C. Even though, the coastal cities will experience the lowest mean temperature, they will have the highest number of heat waves about 62. Majority of the heat waves will last between 2 to 10 days with the maximum 30 days. The surface temperature will continue to rise by the end of the mid-21st century (2041-2050) over the major cities in Ghana and so needs to be addressed to the Environmental Protection Agency in Ghana in order to mitigate this problem.
Impacts of Urbanization on Forest and Agriculture Areas in Savannakhet Province, Lao People's Democratic Republic
The current increased population pushes increasing demands for natural resources and living space. In Laos, urban areas have been expanding rapidly in recent years. The rapid urbanization can have negative impacts on landscapes, including forest and agriculture lands. The primary objective of this research were to map current urban areas in a large city in Savannakhet province, in Laos, 2) to compare changes in urbanization between 1990 and 2018, and 3) to estimate forest and agriculture areas lost due to expansions of urban areas during the last over twenty years within study area. Landsat 8 data was used and existing GIS data was collected including spatial data on rivers, lakes, roads, vegetated areas and other land use/land covers). GIS data was obtained from the government sectors. Object based classification (OBC) approach was applied in ECognition for image processing and analysis of urban area using. Historical data from other Landsat instruments (Landsat 5 and 7) were used to allow us comparing changes in urbanization in 1990, 2000, 2010 and 2018 in this study area. Only three main land cover classes were focused and classified, namely forest, agriculture and urban areas. Change detection approach was applied to illustrate changes in built-up areas in these periods. Our study shows that the overall accuracy of map was 95% assessed, kappa~ 0.8. It is found that that there is an ineffective control over forest and land-use conversions from forests and agriculture to urban areas in many main cities across the province. A large area of agriculture and forest has been decreased due to this conversion. Uncontrolled urban expansion and inappropriate land use planning can lead to creating a pressure in our resource utilisation. As consequence, it can lead to food insecurity and national economic downturn in a long term.
Modified InVEST for Whatsapp Messages Forensic Triage and Search through Visualization
WhatsApp as the most popular mobile messaging app has been used as evidence in many criminal cases. As the use of mobile messages generates large amounts of data, forensic investigation faces the challenge of large data problems. The hardest part of finding this important evidence is because current practice utilizes tools and technique that require manual analysis to check all messages. That way, analyze large sets of mobile messaging data will take a lot of time and effort. Our work offers methodologies based on forensic triage to reduce large data to manageable sets resulting easier to do detailed reviews, then show the results through interactive visualization to show important term, entities and relationship through intelligent ranking using Term Frequency-Inverse Document Frequency (TF-IDF) and Latent Dirichlet Allocation (LDA) Model. By implementing this methodology, investigators can improve investigation processing time and result's accuracy.
A Pilot Study on Cloud Enterprise Application Provider Selection Model for the Small and Medium Enterprise
Enterprise Applications (EAs) aid the organizations achieve operational excellence and competitive advantage. Over time, most Small and Medium Enterprises (SMEs) which are known to be the major drivers of most thriving global economies use the costly on-premise versions of these applications thereby making business difficult to competitively thrive in the same market environment with their large enterprise counterparts. The Advent of Cloud computing presents the SMEs affordable offer and great opportunities as such EAs can be cloud-hosted and rented on a pay-per-use basis which does not require huge initial capital. However, as there are numerous Cloud Service Providers (CSPs) offering EAs as Software-as-a-Service (SaaS), there is a challenge of choosing a suitable provider with Quality of Service (QoS) that meet the organizations’ customized requirements. The proposed model takes care of that and goes a step further to select the most affordable among a selected few of the CSPs. In the earlier stage, before developing the instrument and conducting the pilot test, the researchers conducted a structured interview with 3 experts to validate the proposed model. In conclusion, the validity and reliability of the instrument were tested through experts, typical respondents, and analyzed with SPSS 22. Results confirmed the validity of the proposed model and the validity and reliability of the instrument.
The Suitability of Agile Practices in Healthcare Industry with Regard to Healthcare Regulations
Nowadays, medical devices rely completely on software whether as whole software or as embedded software, therefore, the organization that develops medical device software can benefit from adopting agile practices. Using agile practices in healthcare software development industries would bring benefits such as producing a product of a high-quality with low cost and in short period. However, medical device software development companies faced challenges in adopting agile practices. These due to the gaps that exist between agile practices and the requirements of healthcare regulations such as documentation, traceability, and formality. This research paper will conduct a study to investigate the adoption rate of agile practice in medical device software development, and they will extract and outline the requirements of healthcare regulations such as Food and Drug Administration (FDA), Health Insurance Portability and Accountability Act (HIPAA), and Medical Device Directive (MDD) that affect directly or indirectly on software development life cycle. Moreover, this research paper will evaluate the suitability of using agile practices in healthcare industries by analyzing the most popular agile practices such as eXtream Programming (XP), Scrum, and Feature-Driven Development (FDD) from healthcare industry point of view and in comparison with the requirements of healthcare regulations. Finally, the authors propose an agile mixture model that consists of different practices from different agile methods. As result, the adoption rate of agile practices in healthcare industries still low and agile practices should enhance with regard to requirements of the healthcare regulations in order to be used in healthcare software development organizations. Therefore, the proposed agile mixture model may assist in minimizing the gaps existing between healthcare regulations and agile practices and increase the adoption rate in the healthcare industry. As this research paper part of the ongoing project, an evaluation of agile mixture model will be conducted in the near future.
The Effect of Computerized Systems of Office Automation on Employees' Productivity Efficiency
One of the factors that can play an important role in increasing productivity is the optimal use of information technology, which in this area today has a significant role to play in computer systems of office automation in organizations and companies. Therefore, this research has been conducted with the aim of investigating the effect of the relationship between computerized systems of office automation and the productivity of employees in the municipality of Ilam city. The statistical population of this study was 110 people. Using Cochran formula, the minimum sample size is 78 people. The present research is a descriptive-looking research in terms of the type of objective view. A questionnaire was used to collect data. To assess the reliability of variables, Cornbrash’s alpha coefficient was used, which was equal to 0.85; SPSS19 and Pearson test was used to analyze the data and test the hypothesis of the research. In this research, three hypotheses of the relationship between office automation with efficiency, performance, and effectiveness were investigated. The results showed a direct and positive relationship between the office automation system and the increase in the efficiency, effectiveness, and efficiency of employees, and there was no reason to reject these hypotheses.
Anomaly Detection of Log Analysis using Data Visualization Techniques for Digital Forensics Audit and Investigation
In common digital forensics cases, investigation may rely on the analysis conducted on specific and relevant exhibits involved. Usually the investigation officer may define and advise digital forensic analyst about the goals and objectives to be achieved in reconstructing the trail of evidence while maintaining the specific scope of investigation. With the technology growth, people are starting to realize the importance of cyber security to their organization and this new perspective creates awareness that digital forensics auditing must come in place in order to measure possible threat or attack to their cyber-infrastructure. Instead of performing investigation on incident basis, auditing may broaden the scope of investigation to the level of anomaly detection in daily operation of organization’s cyber space. While handling a huge amount of data such as log files, performing digital forensics audit for large organization proven to be onerous task for the analyst either to analyze the huge files or to translate the findings in a way where the stakeholder can clearly understand. Data visualization can be emphasized in conducting digital forensic audit and investigation to resolve both needs. This study will identify the important factors that should be considered to perform data visualization techniques in order to detect anomaly that meet the digital forensic audit and investigation objectives.
Digital Forensics Analysis Focusing on the Onion Router Browser Artifacts in Windows 10
The Onion Router (Tor) browser is a well-known tool and widely used by people who seeking for web anonymity when browsing the internet. Criminals are taking this advantage to be anonymous over the internet. Accessing the dark web could be the significant reason for the criminal in order for them to perform illegal activities while maintaining their anonymity. For a digital forensic analyst, it is crucial to extract the trail of evidence in proving that the criminal’s computer has used Tor browser to conduct such illegal activities. By applying the digital forensic methodology, several techniques could be performed including application analysis, memory analysis, and registry analysis. Since Windows 10 is the latest operating system released by Microsoft Corporation, this study will use Windows 10 as the operating system platform that running Tor browser. From the analysis, significant artifacts left by Tor browser were discovered such as the execution date, application installation date and browsing history that can be used as an evidence. Although Tor browser was designed to achieved anonymity, there is still some trail of evidence can be found in Windows 10 platform that can be useful for investigation.
A Coevolutionary Framework of Business-IT Alignment through the Lens of Enterprise Architecture
The major challenges for sustainable business-IT alignment (BITA) in a company root in its volatile external competitive environment, increasingly complex internal relationships, and subversive IT roles. Failure to adequately address BITA results in wasting organizational resources, losing competitive advantages, and failing to produce adequate returns on investments. The coevolution is more suitable to describe the dynamic relationships of business and IT and has received certain attention in recent years. Multiple mechanisms for achieving BITC (e.g., sharing domain knowledge, modular design) were obtained. However, instead of a complete managing process, BITC achievement is still hard to operate in practice. This study emphasizes what the BITC management process looks like and how to execute this coevolution step-by-step. A practical coevolutionary framework that combines the enterprise architecture (EA) method with misalignment analysis is proposed in this paper. It contains steps of EA design, misalignment detection, misalignment correction, and EA management /misalignment prevention. The step of misalignment correction is especially discussed at length. This study also evaluates the proposed framework by comparing the characteristics, principles, and approaches of coevolution in the literature.
Reconfigurable Device for 3D Visualization of Three Dimensional Surfaces
The article refers to the development of an augmented reality 3D display, through the control of servo motors and projection of image with aid of video projector on the model. Augmented Reality is a branch that explores multiple approaches to increase real-world view by viewing additional information along with the real scene. The article presents the broad use of electrical, electronic, mechanical and industrial automation for geospatial visualizations, applications in mathematical models with the visualization of functions and 3D surface graphics and volumetric rendering that are currently seen in 2D layers. Application as a 3D display for representation and visualization of Digital Terrain Model (DTM) and Digital Surface Models (DSM), where it can be applied in the identification of canyons in the marine area of the Campos Basin, Rio de Janeiro, Brazil. The same can execute visualization of regions subject to landslides, as in Serra do Mar - Agra dos Reis and Serranas cities both in the State of Rio de Janeiro. From the foregoing, loss of human life and leakage of oil from pipelines buried in these regions may be anticipated in advance. The physical design consists of a table consisting of a 9 x 16 matrix of servo motors, totalizing 144 servos, a mesh is used on the servo motors for visualization of the models projected by a retro projector. Each model for by an image pre-processing, is sent to a server to be converted and viewed from a software developed in C # Programming Language.
Measuring Banks’ Antifragile via Fuzzy Logic
Analysing the world banking sector, we realize that traditional risk measurement methodologies no longer reflect the real scenario with uncertainty and leave out events that can change the dynamics of markets. Considering this, regulators and financial institutions began to look for more realistic models. The aim is to include external influences and interdependencies between agents, to portray and measure the operationalization of these complex systems and their risks in a more realistic way. Within this context, X-Events are more frequent than assumed and with uncertainties and constant changes, and the concept of antifragility starts to gain great prominence in comparison to others methodologies of risk management. It is very useful to analyse whether a system succumbs (fragile), resists (robust) or gets benefits (antifragile) from disorder and stress. Thus, this work proposes the creation of the Banking Antifragility Index (BAI), which is based on the calculation of a triangular fuzzy number – to "quantify" qualitative criteria linked to antifragility.
A Cloud-Based Federated Identity Management in Europe
Currently, there is a so called ‘identity crisis’ in cybersecurity caused by the substantial security, privacy and usability shortcomings encountered in existing systems for identity management. Federated Identity Management (FIM) could be solution for this crisis, as it is a method that facilitates management of identity processes and policies among collaborating entities without enforcing a global consistency, that is difficult to achieve when there are ID legacy systems. To cope with this problem, the Connecting Europe Facility (CEF) initiative proposed in 2014 a federated solution in anticipation of the adoption of the Regulation (EU) N°910/2014, the so-called eIDAS Regulation. At present, a network of eIDAS Nodes is being deployed at European level to allow that every citizen recognized by a member state is to be recognized within the trust network at European level, enabling the consumption of services in other member states that, until now were not allowed, or whose concession was tedious. This is a very ambitious approach, since it tends to enable cross-border authentication of Member States citizens without the need to unify the authentication method (eID Scheme) of the member state in question. However, this federation is currently managed by member states and it is initially applied only to citizens and public organizations. The goal of this paper is to present the results of a European Project, named eID@Cloud, that focuses on the integration of eID in 5 cloud platforms belonging to authentication service providers of different EU Member States to act as Service Providers (SP) for private entities. We propose an initiative based on a private eID Scheme both for natural and legal persons. The methodology followed in the eID@Cloud project is that each Identity Provider (IdP) is subscribed to an eIDAS Node Connector, requesting for authentication, that is subscribed to an eIDAS Node Proxy Service, issuing authentication assertions. To cope with high loads, load balancing is supported in the eIDAS Node. The eID@Cloud project is still going on, but we already have some important outcomes. First, we have deployed the federation identity nodes and tested it from the security and performance point of view. The pilot prototype has shown the feasibility of deploying this kind of systems, ensuring good performance due to the replication of the eIDAS nodes and the load balance mechanism. Second, our solution avoids the propagation of identity data out of the native domain of the user or entity being identified, which avoids problems well known in cybersecurity due to network interception, man in the middle attack, etc. Last, but not least, this system allows to connect any country or collectivity easily, providing incremental development of the network and avoiding difficult political negotiations to agree on a single authentication format (which would be a major stopper).
Semi-Supervised Learning for Spanish Speech Recognition Using Deep Neural Networks
Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.
Modelling the Impact of Installation of Heat Cost Allocators in District Heating Systems Using Machine Learning
Following the regulation of EU Directive on Energy Efficiency, specifically Article 9, individual metering in district heating systems has to be introduced by the end of 2016. These directions have been implemented in member state’s legal framework, Croatia is one of these states. The directive allows installation of both heat metering devices and heat cost allocators. Mainly due to bad communication and PR, the general public false image was created that the heat cost allocators are devices that save energy. Although this notion is wrong, the aim of this work is to develop a model that would precisely express the influence of installation heat cost allocators on potential energy savings in each unit within multifamily buildings. At the same time, in recent years, a science of machine learning has gain larger application in various fields, as it is proven to give good results in cases where large amounts of data are to be processed with an aim to recognize a pattern and correlation of each of the relevant parameter as well as in the cases where the problem is too complex for a human intelligence to solve. A special method of machine learning, decision tree method, has proven an accuracy of over 92% in prediction general building consumption. In this paper, a machine learning algorithms will be used to isolate the sole impact of installation of heat cost allocators on a single building in multifamily houses connected to district heating systems. Special emphasises will be given regression analysis, logistic regression, support vector machines, decision trees and random forest method.
Early Diagnosis of Myocardial Ischemia Based on Support Vector Machine and Gaussian Mixture Model by Using Features of ECG Recordings
Acute myocardial infarction is a major cause of death in the world. Therefore, its fast and reliable diagnosis is a major clinical need. ECG is the most important diagnostic methodology which is used to make decisions about the management of the cardiovascular diseases. In patients with acute myocardial ischemia, temporary chest pains together with changes in ST segment and T wave of ECG occur shortly before the start of myocardial infarction. In this study, a technique which detects changes in ST/T sections of ECG is developed for the early diagnosis of acute myocardial ischemia. For this purpose, a database of real ECG recordings that contains a set of records from 75 patients presenting symptoms of chest pain who underwent elective percutaneous coronary intervention (PCI) is constituted. 12-lead ECG’s of the patients were recorded before and during the PCI procedure. Two ECG epochs, which are the pre-inflation ECG which is acquired before any catheter insertion and the occlusion ECG which is acquired during balloon inflation, are analyzed for each patient. By using pre-inflation and occlusion recordings, ECG features that are critical in the detection of acute myocardial ischemia are identified and the most discriminative features for the detection of acute myocardial ischemia are extracted. A classification technique based on support vector machine (SVM) approach operating with linear and radial basis function (RBF) kernels to detect ischemic events by using ST-T derived joint features from non-ischemic and ischemic states of the patients is developed. The dataset is randomly divided into training and testing sets and the training set is used to optimize SVM hyperparameters by using grid-search method and 10fold cross-validation. SVMs are designed specifically for each patient by tuning the kernel parameters in order to obtain the optimal classification performance results. As a result of implementing the developed classification technique to real ECG recordings, it is shown that the proposed technique provides highly reliable detections of the anomalies in ECG signals. Furthermore, to develop a detection technique that can be used in the absence of ECG recording obtained during healthy stage, the detection of acute myocardial ischemia based on ECG recordings of the patients obtained during ischemia is also investigated. For this purpose, a Gaussian mixture model (GMM) is used to represent the joint pdf of the most discriminating ECG features of myocardial ischemia. Then, a Neyman-Pearson type of approach is developed to provide detection of outliers that would correspond to acute myocardial ischemia. Neyman – Pearson decision strategy is used by computing the average log likelihood values of ECG segments and comparing them with a range of different threshold values. For different discrimination threshold values and number of ECG segments, probability of detection and probability of false alarm values are computed, and the corresponding ROC curves are obtained. The results indicate that increasing number of ECG segments provide higher performance for GMM based classification. Moreover, the comparison between the performances of SVM and GMM based classification showed that SVM provides higher classification performance results over ECG recordings of considerable number of patients.
A Comprehensive Evaluation of Supervised Machine Learning for the Phase Identification Problem
Power distribution circuits undergo frequent network
topology changes that are often left undocumented. As a result, the
documentation of a circuit’s connectivity becomes inaccurate with
time. The lack of reliable circuit connectivity information is one of the
biggest obstacles to model, monitor, and control modern distribution
systems. To enhance the reliability and efficiency of electric power
distribution systems, the circuit’s connectivity information must be
updated periodically. This paper focuses on one critical component of
a distribution circuit’s topology - the secondary transformer to phase
association. This topology component describes the set of phase lines
that feed power to a given secondary transformer (and therefore a
given group of power consumers). Finding the documentation of this
component is call Phase Identification, and is typically performed
with physical measurements. These measurements can take time
lengths on the order of several months, but with supervised learning,
the time length can be reduced significantly. This paper compares
several such methods applied to Phase Identification for a large
range of real distribution circuits, describes a method of training
data selection, describes preprocessing steps unique to the Phase
Identification problem, and ultimately describes a method which
obtains high accuracy (> 96% in most cases, > 92% in the worst
case) using only 5% of the measurements typically used for Phase
An Intelligent Traffic Management System Based on the WiFi and Bluetooth Sensing
This paper introduces an automated clustering solution that applies to WiFi/Bluetooth sensing data and is later used for traffic management applications. The paper initially summarizes a number of clustering approaches and thereafter shows their performance for noise removal. In this context, clustering is used to recognize WiFi and Bluetooth MAC addresses that belong to passengers traveling by a public urban transit bus. The main objective is to build an intelligent system that automatically filters out MAC addresses that belong to persons located outside the bus for different routes in the city of Ottawa. The proposed intelligent system alleviates the need for defining restrictive thresholds that however reduces the accuracy as well as the range of applicability of the solution for different routes. This paper moreover discusses the performance benefits of the presented clustering approaches in terms of the accuracy, time and space complexity, and the ease of use. Note that results of clustering can further be used for the purpose of the origin-destination estimation of individual passengers, predicting the traffic load, and intelligent management of urban bus schedules.
Sensor and Sensor System Design, Selection and Data Fusion Using Non-Deterministic Multi-Attribute Tradespace Exploration
The conceptualization and design phases of a system lifecycle consume a significant amount of the lifecycle budget in the form of direct tasking and capital, as well as the implicit costs associated with unforeseeable design errors that are only realized during downstream phases. Ad hoc or iterative approaches to generating system requirements oftentimes fail to consider the full array of feasible systems or product designs for a variety of reasons, including, but not limited to: initial conceptualization that oftentimes incorporates a priori or legacy features; the inability to capture, communicate and accommodate stakeholder preferences; inadequate technical designs and/or feasibility studies; and locally-, but not globally-, optimized subsystems and components. These design pitfalls can beget unanticipated developmental or system alterations with added costs, risks and support activities, heightening the risk for suboptimal system performance, premature obsolescence or forgone development. Supported by rapid advances in learning algorithms and hardware technology, sensors and sensor systems have become commonplace in both commercial and industrial products. The evolving array of hardware components (i.e. sensors, CPUs, modular / auxiliary access, etc…) as well as recognition, data fusion and communication protocols have all become increasingly complex and critical for design engineers during both concpetualization and implementation. This work seeks to develop and utilize a non-deterministic approach for sensor system design within the multi-attribute tradespace exploration (MATE) paradigm, a technique that incorporates decision theory into model-based techniques in order to explore complex design environments and discover better system designs. Developed to address the inherent design constraints in complex aerospace systems, MATE techniques enable project engineers to examine all viable system designs, assess attribute utility and system performance, and better align with stakeholder requirements. Whereas such previous work has been focused on aerospace systems and conducted in a deterministic fashion, this study addresses a wider array of system design elements by incorporating both traditional tradespace elements (e.g. hardware components) as well as popular multi-sensor data fusion models and techniques. Furthermore, statistical performance features to this model-based MATE approach will enable non-deterministic techniques for various commercial systems that range in application, complexity and system behavior, demonstrating a significant utility within the realm of formal systems decision-making.
The Intersection/Union Region Computation for Drosophila Brain Images Using Encoding Schemes Based on Multi-Core CPUs
With more and more Drosophila Driver and Neuron images, it is an important work to find the similarity relationships among them as the functional inference. There is a general problem that how to find a Drosophila Driver image, which can cover a set of Drosophila Driver/Neuron images. In order to solve this problem, the intersection/union region for a set of images should be computed at first, then a comparison work is used to calculate the similarities between the region and other images. In this paper, three encoding schemes, namely Integer, Boolean, Decimal, are proposed to encode each image as a one-dimensional structure. Then, the intersection/union region from these images can be computed by using the compare operations, Boolean operators and lookup table method. Finally, the comparison work is done as the union region computation, and the similarity score can be calculated by the definition of Tanimoto coefficient. The above methods for the region computation are also implemented in the multi-core CPUs environment with the OpenMP. From the experimental results, in the encoding phase, the performance by the Boolean scheme is the best than that by others; in the region computation phase, the performance by Decimal is the best when the number of images is large. The speedup ratio can achieve 12 based on 16 CPUs. This work was supported by the Ministry of Science and Technology under the grant MOST 106-2221-E-182-070.
Impact of Extended Enterprise Resource Planning in the Context of Cloud Computing on Industries and Organizations
The Extended Enterprise Resource Planning (ERPII)
system usually requires massive amounts of storage space, powerful
servers, and large upfront and ongoing investments to purchase and
manage the software and the related hardware which are not
affordable for organizations. In recent decades, organizations prefer
to adapt their business structures with new technologies for
remaining competitive in the world economy. Therefore, cloud
computing (which is one of the tools of information technology (IT))
is a modern system that reveals the next-generation application
architecture. Also, cloud computing has had some advantages that
reduce costs in many ways such as: lower upfront costs for all
computing infrastructure and lower cost of maintaining and
supporting. On the other hand, traditional ERPII is not responding for
huge amounts of data and relations between the organizations. In this
study, based on a literature study, ERPII is investigated in the context
of cloud computing where the organizations operate more efficiently.
Also, ERPII conditions have a response to needs of organizations in
large amounts of data and relations between the organizations.
CompleX-Machine: An Automated Testing Tool Using X-Machine Theory
This paper is aimed at creating an Automatic Java X-Machine testing tool for software development. The nature of software development is changing; thus, the type of software testing tools required is also changing. Software is growing increasingly complex and, in part due to commercial impetus for faster software releases with new features and value, increasingly in danger of containing faults. These faults can incur huge cost for software development organisations and users; Cambridge Judge Business School’s research estimated the cost of software bugs to the global economy is $312 billion. Beyond the cost, faster software development methodologies and increasing expectations on developers to become testers is driving demand for faster, automated, and effective tools to prevent potential faults as early as possible in the software development lifecycle. Using X-Machine theory, this paper will explore a new tool to address software complexity, changing expectations on developers, faster development pressures and methodologies, with a view to reducing the huge cost of fixing software bugs.
Reconstruction of Performace-Based Budgeting in Indonesian Local Government: Application of Soft Systems Methodology in Producing Guideline for Policy Implementation
Effective public policy creation required a strong budget system, both in terms of design and implementation. Performance-based Budget is an evolutionary approach with two substantial characteristics; first, the strong integration between budgeting and planning, and second, its existence as guidance so that all activities and expenditures refer to measurable performance targets. There are four processes in the government that should be followed in order to make the budget become performance-based. These four processes consist of the preparation of a vision according to the bold aspiration, the formulation of outcome, the determination of output based on the analysis of organizational resources, and the formulation of Value Creation Map that contains a series of programs and activities. This is consistent with the concept of logic model which revealed that the budget performance should be placed within a relational framework of resources, activities, outputs, outcomes and impacts. Through the issuance of Law 17/2003 regarding State Finance, local governments in Indonesia have to implement performance-based budget. Central Government then issued Government Regulation 58/2005 which contains the detail guidelines how to prepare local governments budget. After a decade, implementation of performance budgeting in local government is still not fully meet expectations, though the guidance is completed, socialization routinely performed, and trainings have also been carried out at all levels. Accordingly, this study views the practice of performance-based budget at local governments as a problematic situation. This condition must be approached with a system approach that allows the solutions from many point of views. Based on the fact that the infrastructure of budgeting has already settled, the study then considering the situation as complexity. Therefore, the intervention needs to be done in the area of human activity system. Using Soft Systems Methodology, this research will reconstruct the process of performance-based budget at local governments is area of human activity system. Through conceptual models, this study will invite all actors (central government, local government, and the parliament) for dialogue and formulate interventions in human activity systems that systematically desirable and culturally feasible. The result will direct central government in revise the guidance to local government budgeting process as well as a reference to build the capacity building strategy.
Comparison of Web Development Using Framework over Library
Over recent years, web development has changed significantly. Driven largely by the rise of trends like mobiles, the world of development is rapidly evolving. The rise of the Internet makes web applications crucial nowadays. The web application has been an interface for a company and one of the ways they present their portfolio to the client. On the other hand, the web has become part of the file management system which takes over the role of paper. Due to high demand in web applications, developers are required to develop a web application that are cost-effective, secure and well coded. A framework has been proposed to develop an application rather than using library style development. The framework is helping the developer in creating the structure of a web automatically. This paper will compare the advantages and disadvantages of web development using framework against library-style development. This comparison is based on a previous research paper focusing on two main indicators, which are the impact to management and impact to the developer.
Comparison of Authentication Methods in Internet of Things Technology
Internet of Things (IoT) is a powerful industry system, which end-devices are interconnected and automated, allowing the devices to analyze data and execute actions based on the analysis. The IoT technology leverages the technology of Radio-Frequency Identification (RFID) and Wireless Sensor Network (WSN), including mobile and sensor. These technologies contribute to the evolution of IoT. However, due to more devices are connected each other in the Internet, and data from various sources exchanged between things, confidentiality of the data becomes a major concern. This paper focuses on one of the major challenges in IoT; authentication, in order to preserve data integrity and confidentiality are in place. A few solutions are reviewed based on papers from the last few years. One of the proposed solutions is securing the communication between IoT devices and cloud servers with Elliptic Curve Cryptograhpy (ECC) based mutual authentication protocol. This solution focuses on Hyper Text Transfer Protocol (HTTP) cookies as security parameter. Next proposed solution is using keyed-hash scheme protocol to enable IoT devices to authenticate each other without the presence of a central control server. Another proposed solution uses Physical Unclonable Function (PUF) based mutual authentication protocol. It emphasizes on tamper resistant and resource-efficient technology, which equals a 3-way handshake security protocol.
Segmentation of Arabic Handwritten Numeral Strings Based on Watershed Approach
Arabic offline handwriting recognition systems are considered as one of the most challenging topics. Arabic Handwritten Numeral Strings are used to automate systems that deal with numbers such as postal code, banking account numbers and numbers on car plates. Segmentation of connected numerals is the main bottleneck in the handwritten numeral recognition system. This is, in turn, can increase the speed and efficiency of the recognition system. In this paper, we proposed powerful algorithms for automatic segmentation and feature extraction of Arabic handwritten numeral strings based on Watershed approach. The proposed algorithms have been designed and implemented to achieve the main goal of segmenting and extracting the string of numeral digits written by hand especially in a courtesy amount of bank checks. The segmentation algorithm partitions the string into multiple regions that can be associated with the properties of one or more criterion. The numeral extraction algorithm extracts the numeral string digits into separated individual digit. Both algorithms for segmentation and feature extraction have been tested successfully and efficiently for all types of numerals.