Open Science Research Excellence

Open Science Index

Commenced in January 2007 Frequency: Monthly Edition: International Abstract Count: 56400

Computer and Information Engineering

Multimodal Convolutional Neural Network for Musical Instrument Recognition
The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.
An Improved Data Aided Channel Estimation Technique Using Genetic Algorithm for Massive Multi-Input Multiple-Output
With the increasing rate of wireless devices and high bandwidth operations, wireless networking and communications are becoming over crowded. To cope with such crowdy and messy situation, massive MIMO is designed to work with hundreds of low costs serving antennas at a time as well as improve the spectral efficiency at the same time. TDD has been used for gaining beamforming which is a major part of massive MIMO, to gain its best improvement to transmit and receive pilot sequences. All the benefits are only possible if the channel state information or channel estimation is gained properly. The common methods to estimate channel matrix used so far is LS, MMSE and a linear version of MMSE also proposed in many research works. We have optimized these methods using genetic algorithm to minimize the mean squared error and finding the best channel matrix from existing algorithms with less computational complexity. Our simulation result has shown that the use of GA worked beautifully on existing algorithms in a Rayleigh slow fading channel and existence of Additive White Gaussian Noise. We found that the GA optimized LS is better than existing algorithms as GA provides optimal result in some few iterations in terms of MSE with respect to SNR and computational complexity.
Non-Targeted Adversarial Object Detection Attack: Fast Gradient Sign Method
Today, there are many applications that are using computer vision models, such as face recognition, image classification, and object detection. The accuracy of these models is very important for the performance of these applications. One challenge that facing the computer vision models is the adversarial examples attack. In computer vision, the adversarial example is an image that is intentionally designed to cause the machine learning model to misclassify it. One of very well-known method that is used to attack the Convolution Neural Network (CNN) is Fast Gradient Sign Method (FGSM). The goal of this method is to find the perturbation that can fool the CNN using the gradient of the cost function of CNN. In this paper, we introduce a novel model that can attack Regional-Convolution Neural Network (R-CNN) that use FGSM. We first extract the regions that are detected by R-CNN, and then we resize these regions into the size of regular images. Then, we find the best perturbation of the regions that can fool CNN using FGSM. Next, we add the resulted perturbation to the attacked region to get a new region image that looks similar to the original image to human eyes. Finally, we placed the regions back to the original image and test the R-CNN with the attacked images. Our model could drop the accuracy of the R-CNN when we tested with Pascal VOC 2012 dataset.
A Comprehensive Study and Evaluation on Image Fashion Features Extraction
Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.
Enhancing the Performance of Bug Reporting System by Handling Duplicate Reporting Reports: Artificial Intelligence Based Mantis
Bug reporting systems are most important tool that guides regarding different maintenance activities in software engineering. Duplicate bug reports which describe the bugs and issues in bug reporting system repository increases processing time of bug triage that monitors all such activities and software programmers who are working and spending time on reports which were assigned by triage. These reports can reveal imperfections and degrade software quality. As there is a number of the potential duplicate bug reports increases, the number of bug reports in bug repository increases. Identifying duplicate bug reports help in decreasing development work load in fixing defects. However, it is difficult to manually identify all possible duplicates because of the huge number of already reported bug reports. In this paper, an artificial intelligence based system using Mantis is proposed to automatically detect duplicate bug reports. When new bugs are submitted to repository triages will mark it with a tag. It will investigate that whether it is a duplicate of an existing bug report by matching or not. Reports with duplicate tags will be eliminated from the repository which not only will improve the performance of the system but can also save cost and effort waste on bug triage and finding the duplicate bug.
Efficient Frequent Itemset Mining Methods over Real-Time Spatial Big Data
In recent years, there is a huge increase in the use of spatio-temporal applications where data and queries are continuously moving. As a result, the need to process real-time spatio-temporal data seems clear and real-time stream data management becomes a hot topic. Sliding window model and frequent itemset mining over dynamic data are the most important problems in the context of data mining. Thus, sliding window model for frequent itemset mining is a widely used model for data stream mining due to its emphasis on recent data and its bounded memory requirement. These methods use the traditional transaction-based sliding window model where the window size is based on a fixed number of transactions. Actually, this model supposes that all transactions have a constant rate which is not suited for real-time applications. And the use of this model in such applications endangers their performance. Based on these observations, this paper relaxes the notion of window size and proposes the use of a timestamp-based sliding window model. In our proposed frequent itemset mining algorithm, support conditions are used to differentiate frequents and infrequent patterns. Thereafter, a tree is developed to incrementally maintain the essential information. We evaluate our contribution. The preliminary results are quite promising.
Arabic Character Recognition Using Regression Curves with the Expectation Maximization Algorithm
In this paper, we demonstrate how regression curves can be used to recognize 2D non-rigid handwritten shapes. Each shape is represented by a set of non-overlapping uniformly distributed landmarks. The underlying models utilize 2nd order of polynomials to model shapes within a training set. To estimate the regression models, we need to extract the required coefficients which describe the variations for a set of shape class. Hence, a least square method is used to estimate such modes. We then proceed by training these coefficients using the apparatus Expectation Maximization algorithm. Recognition is carried out by finding the least error landmarks displacement with respect to the model curves. Handwritten isolated Arabic characters are used to evaluate our approach.
Exploring Transitions between Communal- and Market-Based Knowledge Sharing
Markets and communities are often cast as alternative forms of knowledge sharing, but an open question is how and why people dynamically transition between them. To study these transitions, we design a technology that allows geographically distributed participants to either buy knowledge (using virtual points) or request it for free. We use a data-driven, inductive approach, studying 550 members in over 5000 interactions, during nine months. Because the technology offered participants choices between market or community forms, we can document both individual and collective transitions that emerge as people cycle between these forms. Our inductive analysis revealed that uncertainties endemic to knowledge sharing were the impetus for these transitions. Communities evoke uncertainties about knowledge sharing’s costs and benefits, which markets resolve by quantifying explicit prices. However, if people manipulate markets, they create uncertainties about the validity of those prices, allowing communities to reemerge to establish certainty via identity-based validation.
Virtual Team Performance: A Transactive Memory System Perspective
Virtual teams (VT) initiatives, in which teams are geographically dispersed and communicate via modern computer-driven technologies, have attracted increasing attention from researchers and professionals. The growing need to examine how to balance and optimize VT is particularly important given the exposure experienced by companies when their employees encounter globalization and decentralization pressures to monitor VT performance. Hence, organization is regularly limited due to misalignment between the behavioral capabilities of the team’s dispersed competences and knowledge capabilities and how trust issues interplay and influence these VT dimensions and the effects of such exchanges. In fact, the future success of business depends on the extent to which VTs are managing efficiently their dispersed expertise, skills and knowledge to stimulate VT creativity. Transactive memory system (TMS) may enhance VT creativity using its three dimensons: knowledge specialization, credibility and knowledge coordination. TMS can be understood as a composition of both a structural component residing of individual knowledge and a set of communication processes among individuals. The individual knowledge is shared while being retrieved, applied and the learning is coordinated. TMS is driven by the central concept that the system is built on the distinction between internal and external memory encoding. A VT learns something new and catalogs it in memory for future retrieval and use. TMS uses the role of information technology to explain VT behaviors by offering VT members the possibility to encode, store, and retrieve information. TMS considers the members of a team as a processing system in which the location of expertise both enhances knowledge coordination and builds trust among members over time. We build on TMS dimensions to hypothesize the effects of specialization, coordination, and credibility on VT creativity. In fact, VTs consist of dispersed expertise, skills and knowledge that can positively enhance coordination and collaboration. Ultimately, this team composition may lead to recognition of both who has expertise and where that expertise is located; over time, the team composition may also build trust among VT members over time developing the ability to coordinate their knowledge which can stimulate creativity. We also assess the reciprocal relationship between TMS dimensions and VT creativity. We wish to use TMS to provide researchers with a theoretically driven model that is empirically validated through survey evidence. We propose that TMS provides a new way to enhance and balance VT creativity. This study also provides researchers insight into the use of TMS to influence positively VT creativity. In addition to our research contributions, we provide several managerial insights into how TMS components can be used to increase performance within dispersed VTs.
Ontology-Based Systemizing of the Science Information Devoted to Waste Utilizing by Methanogenesis
Over the past decades, amount of scientific information has been growing exponentially. It became more complicated to process and systemize this amount of data. The approach to systematization of scientific information on the production of biogas based on the ontological IT platform "T.O.D.O.S." has been developed. It has been proposed to select semantic characteristics of each work for their further introduction into the IT platform "T.O.D.O.S.". An ontological graph with a ranking function for previous scientific research and for a system of selection of microorganisms has been worked out. These systems provide high performance of information management of scientific information.
A Machine Learning Approach for Assessment of Tremor: A Neurological Movement Disorder
With the changing lifestyle and environment around us, the prevalence of the critical and incurable disease has proliferated. One such condition is the neurological disorder which is rampant among the old age population and is increasing at an unstoppable rate. Most of the neurological disorder patients suffer from some movement disorder affecting the movement of their body parts. Tremor is the most common movement disorder which is prevalent in such patients that infect the upper or lower limbs or both extremities. The tremor symptoms are commonly visible in Parkinson’s disease patient, and it can also be a pure tremor (essential tremor). The patients suffering from tremor face enormous trouble in performing the daily activity, and they always need a caretaker for assistance. In the clinics, the assessment of tremor is done through a manual clinical rating task such as Unified Parkinson’s disease rating scale which is time taking and cumbersome. Neurologists have also affirmed a challenge in differentiating a Parkinsonian tremor with the pure tremor which is essential in providing an accurate diagnosis. Therefore, there is a need to develop a monitoring and assistive tool for the tremor patient that keep on checking their health condition by coordinating them with the clinicians and caretakers for early diagnosis and assistance in performing the daily activity. In our research, we focus on developing a system for automatic classification of tremor which can accurately differentiate the pure tremor from the Parkinsonian tremor using a wearable accelerometer-based device, so that adequate diagnosis can be provided to the correct patient. In this research, a study was conducted in the neuro-clinic to assess the upper wrist movement of the patient suffering from Pure (Essential) tremor and Parkinsonian tremor using a wearable accelerometer-based device. Four tasks were designed in accordance with Unified Parkinson’s disease motor rating scale which is used to assess the rest, postural, intentional and action tremor in such patient. Various features such as time-frequency domain, wavelet-based and fast-Fourier transform based cross-correlation were extracted from the tri-axial signal which was used as input feature vector space for the different supervised and unsupervised learning tools for quantification of severity of tremor. A minimum covariance maximum correlation energy comparison index was also developed which was used as the input feature for various classification tools for distinguishing the PT and ET tremor types. An automatic system for efficient classification of tremor was developed using feature extraction methods, and superior performance was achieved using K-nearest neighbors and Support Vector Machine classifiers respectively.
Shedding Light on the Black Box: Explaining Deep Neural Network Prediction of Clinical Outcome
Deep neural network (DNN) models are being explored in the clinical domain, following the recent success in other domains such as image recognition. For clinical adoption, outcome prediction models require explanation, but due to the multiple non-linear inner transformations, DNN models are viewed by many as a black box. In this study, we developed a deep neural network model for predicting 1-year mortality of patients who underwent major cardio vascular procedures (MCVPs), using temporal image representation of past medical history as input. The dataset was obtained from the electronic medical data warehouse administered by Veteran Affairs Information and Computing Infrastructure (VINCI). We identified 21,355 veterans who had their first MCVP in 2014. Features for prediction included demographics, diagnoses, procedures, medication orders, hospitalizations, and frailty measures extracted from clinical notes. Temporal variables were created based on the patient history data in the 2-year window prior to the index MCVP. A temporal image was created based on these variables for each individual patient. To generate the explanation for the DNN model, we defined a new concept called impact score, based on the presence/value of clinical conditions’ impact on the predicted outcome. Like (log) odds ratio reported by the logistic regression (LR) model, impact scores are continuous variables intended to shed light on the black box model. For comparison, a logistic regression model was fitted on the same dataset. In our cohort, about 6.8% of patients died within one year. The prediction of the DNN model achieved an area under the curve (AUC) of 78.5% while the LR model achieved an AUC of 74.6%. A strong but not perfect correlation was found between the aggregated impact scores and the log odds ratios (Spearman’s rho = 0.74), which helped validate our explanation.
Framework for Integrating Big Data and Thick Data: Understanding Customers Better
With the popularity of data-driven decision making on the rise, this study focuses on providing an alternative outlook towards the process of decision-making. Combining quantitative and qualitative methods rooted in the social sciences, an integrated framework is presented with a focus on delivering a much more robust and efficient approach towards the concept of data-driven decision-making with respect to not only Big data but also 'Thick data', a new form of qualitative data. In support of this, an example from the retail sector has been illustrated where the framework is put into action to yield insights and leverage business intelligence. An interpretive approach to analyze findings from both kinds of quantitative and qualitative data has been used to glean insights. Using traditional Point-of-sale data as well as an understanding of customer psychographics and preferences, techniques of data mining along with qualitative methods (such as grounded theory, ethnomethodology, etc.) are applied. This study’s final goal is to establish the framework as a basis for providing a holistic solution encompassing both the Big and Thick aspects of any business need. The proposed framework is a modified enhancement in lieu of traditional data-driven decision-making approach, which is mainly dependent on quantitative data for decision-making.
An Enhanced Particle Swarm Optimization Algorithm for Multiobjective Problems
Multiobjective Particle Swarm Optimization (MOPSO) has shown an effective performance for solving test functions and real-world optimization problems. However, this method has a premature convergence problem, which may lead to lack of diversity. In order to improve its performance, this paper presents a hybrid approach which embedded the MOPSO into the island model and integrated a local search technique, Variable Neighborhood Search, to enhance the diversity into the swarm. Experiments on two series of test functions have shown the effectiveness of the proposed approach. A comparison with other evolutionary algorithms shows that the proposed approach presented a good performance in solving multiobjective optimization problems.
Experiments on Weakly-Supervised Learning on Imperfect Data
Supervised predictive models require labeled data for training purposes. Complete and accurate labeled data, i.e., a ‘gold standard’, is not always available, and imperfectly labeled data may need to serve as an alternative. An important question is if the accuracy of the labeled data creates a performance ceiling for the trained model. In this study, we trained several models to recognize the presence of delirium in clinical documents using data with annotations that are not completely accurate (i.e., weakly-supervised learning). In the external evaluation, the support vector machine model with a linear kernel performed best, achieving an area under the curve of 89.3% and accuracy of 88%, surpassing the 80% accuracy of the training sample. We then generated a set of simulated data and carried out a series of experiments which demonstrated that models trained on imperfect data can (but do not always) outperform the accuracy of the training data, e.g., the area under the curve for some models is higher than 80% when trained on the data with an error rate of 40%. Our experiments also showed that the error resistance of linear modeling is associated with larger sample size, error type, and linearity of the data (all p-values < 0.001). In conclusion, this study sheds light on the usefulness of imperfect data in clinical research via weakly-supervised learning.
Semi-Supervised Outlier Detection Using a Generative and Adversary Framework
In many outlier detection tasks, only training data belonging to one class, i.e., the positive class, is available. The task is then to predict a new data point as belonging either to the positive class or to the negative class, in which case the data point is considered an outlier. For this task, we propose a novel corrupted Generative Adversarial Network (CorGAN). In the adversarial process of training CorGAN, the Generator generates outlier samples for the negative class, and the Discriminator is trained to distinguish the positive training data from the generated negative data. The proposed framework is evaluated using an image dataset and a real-world network intrusion dataset. Our outlier-detection method achieves state-of-the-art performance on both tasks.
An Automated System for the Detection of Citrus Greening Disease Based on Visual Descriptors
Citrus greening is a bacterial disease that causes considerable damage to citrus fruits worldwide. Efficient method for this disease detection must be carried out to minimize the production loss. This paper presents a pattern recognition system that comprises three stages for the detection of citrus greening from Orange leaves: segmentation, feature extraction and classification. Image segmentation is accomplished by adaptive thresholding. The feature extraction stage comprises of three visual descriptors i.e. shape, color and texture. From shape feature we have used asymmetry index, from color feature we have used histogram of Cb component from YCbCr domain and from texture feature we have used local binary pattern. Classification was done using support vector machines and k nearest neighbors. The best performances of the system is Accuracy = 88.02% and AUROC = 90.1% was achieved by automatic segmented images. Our experiments validate that: (1). Segmentation is an imperative preprocessing step for computer assisted diagnosis of citrus greening, and (2). The combination of shape, color and texture features form a complementary set towards the identification of citrus greening disease.
A Fuzzy-Rough Feature Selection Based on Binary Shuffled Frog Leaping Algorithm
Feature selection and attribute reduction are crucial problems, and widely used techniques in the field of machine learning, data mining and pattern recognition to overcome the well-known phenomenon of the Curse of Dimensionality. This paper presents a feature selection method that efficiently carries out attribute reduction, thereby selecting the most informative features of a dataset. It consists of two components: 1) a measure for feature subset evaluation, and 2) a search strategy. For the evaluation measure, we have employed the fuzzy-rough dependency degree (FRFDD) of the lower approximation-based fuzzy-rough feature selection (L-FRFS) due to its effectiveness in feature selection. As for the search strategy, a modified version of a binary shuffled frog leaping algorithm is proposed (B-SFLA). The proposed feature selection method is obtained by hybridizing the B-SFLA with the FRDD. Nine classifiers have been employed to compare the proposed approach with several existing methods over twenty two datasets, including nine high dimensional and large ones, from the UCI repository. The experimental results demonstrate that the B-SFLA approach significantly outperforms other metaheuristic methods in terms of the number of selected features and the classification accuracy.
FPGA Implementation of the BB84 Protocol
The development of a quantum key distribution (QKD) system on a field-programmable gate array (FPGA) platform is the subject of this paper. A quantum cryptographic protocol is designed based on the properties of quantum information and the characteristics of FPGAs. The proposed protocol performs key extraction, reconciliation, error correction, and privacy amplification tasks to generate a perfectly secret final key. We modeled the presence of the spy in our system with a strategy to reveal some of the exchanged information without being noticed. Using an FPGA card with a 100 MHz clock frequency, we have demonstrated the evolution of the error rate as well as the amounts of mutual information (between the two interlocutors and that of the spy) passing from one step to another in the key generation process.
Cells Detection and Recognition in Bone Marrow Examination with Deep Learning Method
In this paper, deep learning methods are applied in bio-medical field to detect and count different types of cells in an automatic way instead of manual work in medical practice, specifically in bone marrow examination. The process is mainly composed of two steps, detection and recognition. Mask-Region-Convolutional Neural Networks (Mask-RCNN) was used for detection and image segmentation to extract cells and then Convolutional Neural Networks (CNN), as well as Deep Residual Network (ResNet) was used to classify. Result of cell detection network shows high efficiency to meet application requirements. For the cell recognition network, two networks are compared and the final system is fully applicable.
Facial Recognition and Landmark Detection in Fitness Assessment and Performance Improvement
For physical therapy, exercise prescription, athlete training, and regular fitness training, it is crucial to perform health assessments or fitness assessments periodically. An accurate assessment is propitious for tracking recovery progress, preventing potential injury and making long-range training plans. Assessments include necessary measurements, height, weight, blood pressure, heart rate, body fat, etc. and advanced evaluation, muscle group strength, stability-mobility, and movement evaluation, etc. In the current standard assessment procedures, the accuracy of assessments, especially advanced evaluations, largely depends on the experience of physicians, coaches, and personal trainers. And it is challenging to track clients’ progress in the current assessment. Unlike the tradition assessment, in this paper, we present a deep learning based face recognition algorithm for accurate, comprehensive and trackable assessment. Based on the result from our assessment, physicians, coaches, and personal trainers are able to adjust the training targets and methods. The system categorizes the difficulty levels of the current activity for the client or user, furthermore make more comprehensive assessments based on tracking muscle group over time using a designed landmark detection method. The system also includes the function of grading and correcting the form of the clients during exercise. Experienced coaches and personal trainer can tell the clients' limit based on their facial expression and muscle group movements, even during the first several sessions. Similar to this, using a convolution neural network, the system is trained with people’s facial expression to differentiate challenge levels for clients. It uses landmark detection for subtle changes in muscle groups movements. It measures the proximal mobility of the hips and thoracic spine, the proximal stability of the scapulothoracic region and distal mobility of the glenohumeral joint, as well as distal mobility, and its effect on the kinetic chain. This system integrates data from other fitness assistant devices, including but not limited to Apple Watch, Fitbit, etc. for a improved training and testing performance. The system itself doesn’t require history data for an individual client, but the history data of a client can be used to create a more effective exercise plan. In order to validate the performance of the proposed work, an experimental design is presented. The results show that the proposed work contributes towards improving the quality of exercise plan, execution, progress tracking, and performance.
An Automatic Generating Unified Modelling Language Use Case Diagram and Test Cases Based on Classification Tree Method
The processes in software development by Object Oriented methodology have many stages those take time and high cost. The inconceivable error in system analysis process will affect to the design and the implementation process. The unexpected output causes the reason why we need to revise the previous process. The more rollback of each process takes more expense and delayed time. Therefore, the good test process from the early phase, the implemented software is efficient, reliable and also meet the user’s requirement. Unified Modelling Language (UML) is the tool which uses symbols to describe the work process in Object Oriented Analysis (OOA). This paper presents the approach for automatically generated UML use case diagram and test cases. UML use case diagram is generated from the event table and test cases are generated from use case specifications and Graphic User Interfaces (GUI). Test cases are derived from the Classification Tree Method (CTM) that classify data to a node present in the hierarchy structure. Moreover, this paper refers to the program that generates use case diagram and test cases. As the result, it can reduce work time and increase efficiency work.
A Recommender System for Dynamic Selection of Undergraduates' Elective Courses
The task of selecting a few elective courses from a variety of available elective courses has been a difficult one for many students over the years. In many higher institutions, guidance and counselors or level advisers are usually employed to assist the students in picking the right choice of courses. In reality, these counselors and advisers are most times overloaded with too many students to attend to, and sometimes they do not have enough time for the students. Most times, the academic strength of the student based on past results are not considered in the new choice of electives. Recommender systems implement advanced data analysis techniques to help users find the items of their interest by producing a predicted likeliness score or a list of top recommended items for a given active user. Therefore, in this work, a collaborative filtering-based recommender system that will dynamically recommend elective courses to undergraduate students based on their past grades in related courses was developed. This approach employed the use of the k-nearest neighbor algorithm to discover hidden relationships between the related courses passed by students in the past and the currently available elective courses. Real students’ results dataset was used to build and test the recommendation model. The developed system will not only improve the academic performance of students, but it will also help reduce the workload on the level advisers and school counselors.
Cost Effective Real-Time Image Processing Based Optical Mark Reader
In this modern era of automation, most of the academic exams and competitive exams are Multiple Choice Questions (MCQ). The responses of these MCQ based exams are recorded in the Optical Mark Reader (OMR) sheet. Evaluation of the OMR sheet requires separate specialized machines for scanning and marking. The sheets used by these machines are special and costs more than a normal sheet. Available process is non-economical and dependent on paper thickness, scanning quality, paper orientation, special hardware and customized software. This study tries to tackle the problem of evaluating the OMR sheet without any special hardware and making the whole process economical. We propose an image processing based algorithm which can be used to read and evaluate the scanned OMR sheets with no special hardware required. It will eliminate the use of special OMR sheet. Responses recorded in normal sheet is enough for evaluation. The proposed system takes care of color, brightness, rotation, little imperfections in the OMR sheet images.
Housing Price Prediction Using Machine Learning Algorithms: The Case of Melbourne City, Australia
House price forecasting is a main topic in the real estate market research. Effective house price prediction models could not only allow home buyers and real estate agents to make better data-driven decisions but may also be beneficial for the property policymaking process. This study investigates the housing market by using machine learning techniques to analyze real historical house sale transactions in Australia. It seeks useful models which could be deployed as an application for house buyers and sellers. Data analytics show a high discrepancy between the house price in the most expensive suburbs and the most affordable suburbs in the city of Melbourne. In addition, experiments demonstrate that the combination of Stepwise and Support Vector Machine (SVM), based on the Mean Squared Error (MSE) measurement, consistently outperforms other models in terms of prediction accuracy.
Consolidating Service Engineering Ontologies Building Service Ontology from SOA Modeling Language (SoaML)
As a term for characterizing a process of devising a service system, the term &lsquo;service engineering&rsquo; is still regarded as an &lsquo;open&rsquo; research challenge due to unspecified details and conflicting perspectives. This paper presents consolidated service engineering ontologies in collecting, specifying and defining relationship between components pertinent within the context of service engineering. The ontologies are built by way of literature surveys from the collected conceptual works by collating various concepts into an integrated ontology. Two ontologies are produced: general service ontology and software service ontology. The software-service ontology is drawn from the informatics domain, while the generalized ontology of a service system is built from both a business management and the information system perspective. The produced ontologies are verified by exercising conceptual operationalizations of the ontologies in adopting several service orientation features and service system patterns. The proposed ontologies are demonstrated to be sufficient to serve as a basis for a service engineering framework.
Analysis of Electrocardiography Survey Data for the Classification of Heart Diseases Using Artificial Neural Network
Due to the prevalence of heart diseases in Pakistan and the frequent deaths of patients, there is a need to predict the heart diseases timely so that the proper medication can be provided to the patients. The research is performed on the results of the survey (based on patient’s ECG) conducted in Pakistan. This study makes use of ECGs collected from the different areas of Pakistan. The readings of parameters P, Q, R, S, T and their intervals are extracted from these collected ECGs. The dataset contains 278 features which contain readings from all leads and also include patient’s name, age, location, etc. Firstly, feature reduction algorithm is applied to the dataset. The dataset features are reduced from 278 to 150 with the help of the dimensionality reduction algorithm principle component analysis. After reducing the features, a neural network algorithm is applied for predicting heart diseases. A feedforward multi-layer neural network (NN) with error back-propagation (BP) learning algorithm is applied to the dataset in order to predict the heart diseases from ECG signals. The best prediction rates obtained are 99.5% with one hidden layer. The results successfully showed that this framework is efficiently predicting heart diseases that can be used for improving the diagnosis of heart diseases in Pakistan and also used for educational purposes. By further analyzing the results, we can see that most common diseases in the collected dataset are with label: 10.
A Comparative Study of GTC and PSP Algorithms for Mining Sequential Patterns Embedded in Database with Time Constraints
This paper will consider the problem of sequential mining patterns embedded in a database by handling the time constraints as defined in the GSP algorithm (level wise algorithms). We will compare two previous approaches GTC and PSP, that resumes the general principles of GSP. Furthermore this paper will discuss PG-hybrid algorithm, that using PSP and GTC. The results show that PSP and GTC are more efficient than GSP. On the other hand, the GTC algorithm performs better than PSP. The PG-hybrid algorithm use PSP algorithm for the two first passes on the database, and GTC approach for the following scans. Experiments show that the hybrid approach is very efficient for short, frequent sequences.
An Integrated Web-Based Workflow System for Design of Computational Pipelines in the Cloud
With more and more workflow systems adopting cloud as their execution environment, it presents various challenges that need to be addressed in order to be utilized efficiently. This paper introduces a method for resource provisioning based on our previous research of dynamic allocation and its pipeline processes. We present an abstraction for workload scheduling in which independent tasks get scheduled among various available processors of distributed computing for optimization. We also propose an integrated web-based workflow designer by taking advantage of the HTML5 technology and chaining together multiple tools. In order to make the combination of multiple pipelines executing on the cloud in parallel, we develop a script translator and an execution engine for workflow management in the cloud. All information is known in advance by the workflow engine and tasks are allocated according to the prior knowledge in the repository. This proposed effort has the potential to provide support for process definition, workflow enactment and monitoring of workflow processes. Users would benefit from the web-based system that allows creation and execution of pipelines without scripting knowledge.
An Energy Efficient Spectrum Shaping Scheme for Substrate Integrated Waveguides Based on Spread Reshaping Code
In the microwave and millimeter-wave transmission region, substrate-integrated waveguide (SIW) is a very promising candidate for the development of circuits and components. It facilitates the transmission at the data rates in excess of 200 Gbit/s. An SIW mimics a rectangular waveguide by approximating the closed sidewalls with a via fence. This structure suppresses the low frequency components and makes the channel of the SIW a bandpass or high pass filter. This channel characteristic impedes the conventional baseband transmission using non-return-to-zero (NRZ) pulse shaping scheme. Therefore, mixers are commonly proposed to be used as carrier modulator and demodulator in order to facilitate a passband transmission. However, carrier modulation is not an energy efficient solution, because modulation and demodulation at high frequencies consume a lot of energy. For the first time to our knowledge, this paper proposes a spectrum shaping scheme of low complexity for the channel of SIW, namely spread reshaping code. It aims at matching the spectrum of the transmit signal to the channel frequency response. It facilitates the transmission through the SIW channel while it avoids using carrier modulation. In some cases, it even does not need equalization. Simulations reveal a good performance of this scheme, such that, as a result, eye opening is achieved without any equalization or modulation for the respective transmission channels.