Search results for: training algorithms

Authors: A. Kablan

Abstract:

The prediction of financial time series is a very complicated process. If the efficient market hypothesis holds, then the predictability of most financial time series would be a rather controversial issue, due to the fact that the current price contains already all available information in the market. This paper extends the Adaptive Neuro Fuzzy Inference System for High Frequency Trading which is an expert system that is capable of using fuzzy reasoning combined with the pattern recognition capability of neural networks to be used in financial forecasting and trading in high frequency. However, in order to eliminate unnecessary input in the training phase a new event based volatility model was proposed. Taking volatility and the scaling laws of financial time series into consideration has brought about the development of the Intraday Seasonality Observation Model. This new model allows the observation of specific events and seasonalities in data and subsequently removes any unnecessary data. This new event based volatility model provides the ANFIS system with more accurate input and has increased the overall performance of the system.

Keywords: Adaptive Neuro-fuzzy Inference system, High Frequency Trading, Intraday Seasonality Observation Model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3407

Authors: Davinder Singh, Jaimal Singh Khamba, Tarun Nanda

Abstract:

MSMEs are regarded as the sunrise sector of the Indian economy in view of its large potential for growth and likely socio economic impact specifically on employment and income generation. In today’s competitive business environment, global competition forces companies to continuously seek ways of improving their products and services. The pressure on organizations to adapt to new technologies and external threats requires resourcefulness, creativity and innovation. Market has become more open, competitive and customers more demanding. Without continuous technology innovation, no organization can ever remain competitive. Innovations reflect a critical way in which organizations respond to either technological or market challenges. The need of the market is to deliver high quality products through continuous changing in features in product, improve existing products, reduce their cost, and improve employee skills, training, technology infrastructure and financial policies. Therefore, the key factor of organization’s ability to change is innovation. The study presents a detailed review of literature on the role of technology innovation in improving manufacturing performance of industries.

Keywords: Competitive, Manufacturing performance, MSMEs, Technological Innovation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2919

Authors: Duy M. P. Vo, Cornelia Sennewald, Gerald Hoffmann, Chokri Cherif

Abstract:

The development of solutions to improve the resistance of buildings to short-term dynamic loads, particularly impact load, is driven by the urgent demand worldwide on securing human life and critical infrastructures. The research training group GRK 2250/1 aims to develop mineral-bonded composites that allow the fabrication of thin-layered strengthening layers providing available concrete members with enhanced impact resistance. This paper presents the development of 3D woven wire cellular structures that can be used as innovative reinforcement for targeted composites. 3D woven wire cellular structures are truss-like architectures that can be fabricated in an automatized process with a great customization possibility. The specific architecture allows this kind of structures to have good load bearing capability and forming behavior, which is of great potential to give strength against impact loading. An appropriate combination of topology and material enables an optimal use of thin-layered reinforcement in concrete constructions.

Keywords: 3D woven cellular structures, ductile behavior, energy absorption, fiber-based reinforced concrete, impact resistant.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 711

Authors: Ana Paula Lisboa Sohn, Filipa Dionísio Vieria, Nelson Casarotto, Idaulo José Cunha

Abstract:

This paper aims at identifying and analyzing the knowledge transmission channels in textile and clothing clusters located in Brazil and in Europe. Primary data was obtained through interviews with key individuals. The collection of primary data was carried out based on a questionnaire with ten categories of indicators of knowledge transmission. Secondary data was also collected through a literature review and through international organizations sites. Similarities related to the use of the main transmission channels of knowledge are observed in all cases. The main similarities are: influence of suppliers of machinery, equipment and raw materials; imitation of products and best practices; training promoted by technical institutions and businesses; and cluster companies being open to acquire new knowledge. The main differences lie in the relationship between companies, where in Europe the intensity of this relationship is bigger when compared to Brazil. The differences also occur in importance and frequency of the relationship with the government, with the cultural environment, and with the activities of research and development. It is also found factors that reduce the importance of geographical proximity in transmission of knowledge, and in generating trust and the establishment of collaborative behavior.

Keywords: Industrial clusters, interorganizational learning, knowledge transmission channels, textile and clothing industry.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2096

Authors: Anchana Khemphila, Veera Boonjing

Abstract:

In this study, the Multi-Layer Perceptron (MLP)with Back-Propagation learning algorithm are used to classify to effective diagnosis Parkinsons disease(PD).It-s a challenging problem for medical community.Typically characterized by tremor, PD occurs due to the loss of dopamine in the brains thalamic region that results in involuntary or oscillatory movement in the body. A feature selection algorithm along with biomedical test values to diagnose Parkinson disease.Clinical diagnosis is done mostly by doctor-s expertise and experience.But still cases are reported of wrong diagnosis and treatment. Patients are asked to take number of tests for diagnosis.In many cases,not all the tests contribute towards effective diagnosis of a disease.Our work is to classify the presence of Parkinson disease with reduced number of attributes.Original,22 attributes are involved in classify.We use Information Gain to determine the attributes which reduced the number of attributes which is need to be taken from patients.The Artificial neural networks is used to classify the diagnosis of patients.Twenty-Two attributes are reduced to sixteen attributes.The accuracy is in training data set is 82.051% and in the validation data set is 83.333%.

Keywords: Data mining, classification, Parkinson disease, artificial neural networks, feature selection, information gain.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3796

Authors: Hadi Veisi, Mansour Jamzad

Abstract:

In this paper we present an adaptive method for image compression that is based on complexity level of the image. The basic compressor/de-compressor structure of this method is a multilayer perceptron artificial neural network. In adaptive approach different Back-Propagation artificial neural networks are used as compressor and de-compressor and this is done by dividing the image into blocks, computing the complexity of each block and then selecting one network for each block according to its complexity value. Three complexity measure methods, called Entropy, Activity and Pattern-based are used to determine the level of complexity in image blocks and their ability in complexity estimation are evaluated and compared. In training and evaluation, each image block is assigned to a network based on its complexity value. Best-SNR is another alternative in selecting compressor network for image blocks in evolution phase which chooses one of the trained networks such that results best SNR in compressing the input image block. In our evaluations, best results are obtained when overlapping the blocks is allowed and choosing the networks in compressor is based on the Best-SNR. In this case, the results demonstrate superiority of this method comparing with previous similar works and JPEG standard coding.

Keywords: Adaptive image compression, Image complexity, Multi-layer perceptron neural network, JPEG Standard, PSNR.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2233

Authors: Getaye F. Gelaw

Abstract:

The purpose of this study was to investigate the key physical and physiological characteristics of 16 elite male Ethiopian national team distance runners, who have an average age of 28.1 ± 4.3 years, a height of 175.0 ± 5.6 cm, a weight of 59.1 ± 3.9 kg, a BMI of 19.6 ± 1.5, and training age of 10.1 ± 5.1 yrs. The average weekly distance is 196.3 ± 13.8 km, the average 10,000 m time is 27:14 ± 0.5 min sec, the average half marathon time is 59:30 ± 0.6 min sec, the average marathon time is 2:04:20 ± 2.7 hr min ss. In addition, the average Cooper test (12-minute run test) is 4525.4 ± 139.7 meters, and the average VO2 max is 90.8 ± 3.1 ml/kg/m. All athletes have a high profile and compete on the international label, and according to the World Athletics athletes' ranking system in 2021, 56.3% of the 16 participants were platinum label status, while the remaining 43.7% were gold label status-completed an incremental treadmill test for the assessment of VO2peak, submaximal running, lactate threshold and test during which they ran continuously at 21 km/h. The laboratory determined VO2peak was 91.4 ± 1.7 mL/kg/min with anaerobic threshold of 74.2 ± 1.6 mL/min/kg and VO2 max 81%. The speed at the Anaerobic Threshold (AT) is 15.9 ± 0.6 kmh and the altitude is 4.0%. The Respiratory Compensation Point (RCP) was reached at 88.7 ± 1.1 mL/min/kg and 97% of VO2 max. On RCP, the speed is 17.6 ± 0.4 km/h and the altitude/slope are 5.5%, and the speed at Maximum effort is 19.5 ± 1.5 and the elevation is 6.0%. The data also suggest that Ethiopian distance top athletes have considerably higher VO2 max values than those found in earlier research.

Keywords: Long-distance running, Ethiopians, VO2 max, World Athletics, Anthropometric.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 290

Authors: Wenguang Ding, Yang Wu, Xia Wang, Yayu Gao

Abstract:

Biogas, a clean renewable energy, is attracting a growing concern of researchers and professionals in many fields. Based on the natural and climatic conditions in semi-arid regions of northwestern China, the present study introduces a specifically-designed family-size biogas plant (with a digester of 10m3) with manure and urine of animals and humanity as raw materials. The biogas plant is applicable to areas with altitudes of more than 2000 meters in northwestern China. In addition to the installation cost, a little operational expenditure, structure, characteristics, benefits of this small-scale biogas plant, this article introduces a wide range of specific popularization methods such as training, financial support, guided tour to the biogas plant, community-based group study and delivery of operational manuals. The feasibility of the biogas plant is explored on the basis of the availability of the raw materials. Simple operations contained in the current work increase the possibility of the wide use of this small-scale biogas plant in similar regions of the world.

Keywords: biogas, family-size biogas plant, northwestern China, popularization

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2783

Authors: V Krishnaveni, S Jayaraman, A Gunasekaran, K Ramadoss

Abstract:

The ElectroEncephaloGram (EEG) is useful for clinical diagnosis and biomedical research. EEG signals often contain strong ElectroOculoGram (EOG) artifacts produced by eye movements and eye blinks especially in EEG recorded from frontal channels. These artifacts obscure the underlying brain activity, making its visual or automated inspection difficult. The goal of ocular artifact removal is to remove ocular artifacts from the recorded EEG, leaving the underlying background signals due to brain activity. In recent times, Independent Component Analysis (ICA) algorithms have demonstrated superior potential in obtaining the least dependent source components. In this paper, the independent components are obtained by using the JADE algorithm (best separating algorithm) and are classified into either artifact component or neural component. Neural Network is used for the classification of the obtained independent components. Neural Network requires input features that exactly represent the true character of the input signals so that the neural network could classify the signals based on those key characters that differentiate between various signals. In this work, Auto Regressive (AR) coefficients are used as the input features for classification. Two neural network approaches are used to learn classification rules from EEG data. First, a Polynomial Neural Network (PNN) trained by GMDH (Group Method of Data Handling) algorithm is used and secondly, feed-forward neural network classifier trained by a standard back-propagation algorithm is used for classification and the results show that JADE-FNN performs better than JADEPNN.

Keywords: Auto Regressive (AR) Coefficients, Feed Forward Neural Network (FNN), Joint Approximation Diagonalisation of Eigen matrices (JADE) Algorithm, Polynomial Neural Network (PNN).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1894

Authors: H. Shayeghi, A. Jalili

Abstract:

This paper presents a new Hybrid Fuzzy (HF) PID type controller based on Genetic Algorithms (GA-s) for solution of the Automatic generation Control (AGC) problem in a deregulated electricity environment. In order for a fuzzy rule based control system to perform well, the fuzzy sets must be carefully designed. A major problem plaguing the effective use of this method is the difficulty of accurately constructing the membership functions, because it is a computationally expensive combinatorial optimization problem. On the other hand, GAs is a technique that emulates biological evolutionary theories to solve complex optimization problems by using directed random searches to derive a set of optimal solutions. For this reason, the membership functions are tuned automatically using a modified GA-s based on the hill climbing method. The motivation for using the modified GA-s is to reduce fuzzy system effort and take large parametric uncertainties into account. The global optimum value is guaranteed using the proposed method and the speed of the algorithm-s convergence is extremely improved, too. This newly developed control strategy combines the advantage of GA-s and fuzzy system control techniques and leads to a flexible controller with simple stricture that is easy to implement. The proposed GA based HF (GAHF) controller is tested on a threearea deregulated power system under different operating conditions and contract variations. The results of the proposed GAHF controller are compared with those of Multi Stage Fuzzy (MSF) controller, robust mixed H2/H∞ and classical PID controllers through some performance indices to illustrate its robust performance for a wide range of system parameters and load changes.

Keywords: AGC, Hybrid Fuzzy Controller, Deregulated Power System, Power System Control, GAs.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1743

Authors: Vijaya K. Srivastava, Davide Spinello

Abstract:

This paper presents performance of two robust gradient-based heuristic optimization procedures based on 3ⁿ enumeration and tunneling approach to seek global optimum of constrained integer problems. Both these procedures consist of two distinct phases for locating the global optimum of integer problems with a linear or non-linear objective function subject to linear or non-linear constraints. In both procedures, in the first phase, a local minimum of the function is found using the gradient approach coupled with hemstitching moves when a constraint is violated in order to return the search to the feasible region. In the second phase, in one optimization procedure, the second sub-procedure examines 3ⁿ integer combinations on the boundary and within hypercube volume encompassing the result neighboring the result from the first phase and in the second optimization procedure a tunneling function is constructed at the local minimum of the first phase so as to find another point on the other side of the barrier where the function value is approximately the same. In the next cycle, the search for the global optimum commences in both optimization procedures again using this new-found point as the starting vector. The search continues and repeated for various step sizes along the function gradient as well as that along the vector normal to the violated constraints until no improvement in optimum value is found. The results from both these proposed optimization methods are presented and compared with one provided by popular MS Excel solver that is provided within MS Office suite and other published results.

Keywords: Constrained integer problems, enumerative search algorithm, Heuristic algorithm, tunneling algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 808

Authors: First G.M. Karthik, Second Ramachandra.V.Pujeri, Dr.

Abstract:

Generalized Center String (GCS) problem are generalized from Common Approximate Substring problem and Common substring problems. GCS are known to be NP-hard allowing the problems lies in the explosion of potential candidates. Finding longest center string without concerning the sequence that may not contain any motifs is not known in advance in any particular biological gene process. GCS solved by frequent pattern-mining techniques and known to be fixed parameter tractable based on the fixed input sequence length and symbol set size. Efficient method known as Bpriori algorithms can solve GCS with reasonable time/space complexities. Bpriori 2 and Bpriori 3-2 algorithm are been proposed of any length and any positions of all their instances in input sequences. In this paper, we reduced the time/space complexity of Bpriori algorithm by Constrained Based Frequent Pattern mining (CBFP) technique which integrates the idea of Constraint Based Mining and FP-tree mining. CBFP mining technique solves the GCS problem works for all center string of any length, but also for the positions of all their mutated copies of input sequence. CBFP mining technique construct TRIE like with FP tree to represent the mutated copies of center string of any length, along with constraints to restraint growth of the consensus tree. The complexity analysis for Constrained Based FP mining technique and Bpriori algorithm is done based on the worst case and average case approach. Algorithm's correctness compared with the Bpriori algorithm using artificial data is shown.

Keywords: Constraint Based Mining, FP tree, Data mining, GCS problem, CBFP mining technique.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1710

Authors: L. Parisi

Abstract:

In this research work, neural networks were applied to classify two types of hip joint implants based on the relative hip joint implant side speed and three components of each ground reaction force. The condition of walking gait at normal velocity was used and carried out with each of the two hip joint implants assessed. Ground reaction forces’ kinetic temporal changes were considered in the first approach followed but discarded in the second one. Ground reaction force components were obtained from eighteen patients under such gait condition, half of which had a hip implant type I-II, whilst the other half had the hip implant, defined as type III by Orthoload®. After pre-processing raw gait kinetic data and selecting the time frames needed for the analysis, the ground reaction force components were used to train a MLP neural network, which learnt to distinguish the two hip joint implants in the abovementioned condition. Further to training, unknown hip implant side and ground reaction force components were presented to the neural networks, which assigned those features into the right class with a reasonably high accuracy for the hip implant type I-II and the type III. The results suggest that neural networks could be successfully applied in the performance assessment of hip joint implants.

Keywords: Kinemic gait data, Neural networks, Hip joint implant, Hip arthroplasty, Rehabilitation Engineering.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1808

Authors: Abdolghani Ebrahimi, Diego Klabjan, Chenxi Ge, Daniela Ladner, Parker Stride

Abstract:

In this work, we use machine learning and data analysis techniques to predict the one-year mortality of cirrhotic patients. Data from 2,322 patients with liver cirrhosis are collected at a single medical center. Different machine learning models are applied to predict one-year mortality. A comprehensive feature space including demographic information, comorbidity, clinical procedure and laboratory tests is being analyzed. A temporal pattern mining technic called Frequent Subgraph Mining (FSM) is being used. Model for End-stage liver disease (MELD) prediction of mortality is used as a comparator. All of our models statistically significantly outperform the MELD-score model and show an average 10% improvement of the area under the curve (AUC). The FSM technic itself does not improve the model significantly, but FSM, together with a machine learning technique called an ensemble, further improves the model performance. With the abundance of data available in healthcare through electronic health records (EHR), existing predictive models can be refined to identify and treat patients at risk for higher mortality. However, due to the sparsity of the temporal information needed by FSM, the FSM model does not yield significant improvements. Our work applies modern machine learning algorithms and data analysis methods on predicting one-year mortality of cirrhotic patients and builds a model that predicts one-year mortality significantly more accurate than the MELD score. We have also tested the potential of FSM and provided a new perspective of the importance of clinical features.

Keywords: machine learning, liver cirrhosis, subgraph mining, supervised learning

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 460

Authors: Ahmed El Azab, Amira M. Idrees, Mahmoud A. Mahmoud, Hesham Hefny

Abstract:

Social networking sites such as Twitter and Facebook attracts over 500 million users across the world, for those users, their social life, even their practical life, has become interrelated. Their interaction with social networking has affected their life forever. Accordingly, social networking sites have become among the main channels that are responsible for vast dissemination of different kinds of information during real time events. This popularity in Social networking has led to different problems including the possibility of exposing incorrect information to their users through fake accounts which results to the spread of malicious content during life events. This situation can result to a huge damage in the real world to the society in general including citizens, business entities, and others. In this paper, we present a classification method for detecting the fake accounts on Twitter. The study determines the minimized set of the main factors that influence the detection of the fake accounts on Twitter, and then the determined factors are applied using different classification techniques. A comparison of the results of these techniques has been performed and the most accurate algorithm is selected according to the accuracy of the results. The study has been compared with different recent researches in the same area; this comparison has proved the accuracy of the proposed study. We claim that this study can be continuously applied on Twitter social network to automatically detect the fake accounts; moreover, the study can be applied on different social network sites such as Facebook with minor changes according to the nature of the social network which are discussed in this paper.

Keywords: Fake accounts detection, classification algorithms, twitter accounts analysis, features based techniques.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5853

Authors: Yang Meng, James L. Patnao

Abstract:

With the rapid development of information technology and the impact of the epidemic environment, the traditional teaching model can no longer meet the requirements of the development of the times. The development of teaching mechanisms is the inevitable trend of the future development of higher education. We must further promote the informatization of higher education in teaching practice, let modern information technology penetrate and practice in classroom teaching, and provide promising opportunities for the high-quality development of higher education. This article, mainly through the distribution of questionnaires to teachers of colleges and universities, aims to understand the degree of informatization in the teaching of colleges and universities. And on the basis of domestic and foreign scholars' research on higher education informatization, it analyzes the existing problems, and finds the optimal solution based on the needs of education and teaching development. According to the survey results, most college teachers will use information technology in teaching practice, but the information technology teaching tools used by teachers are relatively simple, and most of them use only slides. In addition, backward informatization infrastructure and less informatization training are the main challenges facing the current teaching informatization construction. If colleges and universities can make good use of information technology and multimedia technology and combine it with traditional teaching, it will definitely promote the development of college education and further promote the modernization and informatization of higher education.

Keywords: Higher education, teaching practice, informatization construction, e-education.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 251

Authors: Yi-Shiang Shiu, Meng-Lung Lin, Kang-Tsung Chang, Tzu-How Chu

Abstract:

Most paddy rice fields in East Asia are small parcels, and the weather conditions during the growing season are usually cloudy. FORMOSAT-2 multi-spectral images have an 8-meter resolution and one-day recurrence, ideal for mapping paddy rice fields in East Asia. To map rice fields, this study first determined the transplanting and the most active tillering stages of paddy rice and then used multi-temporal images to distinguish different growing characteristics between paddy rice and other ground covers. The unsupervised ISODATA (iterative self-organizing data analysis techniques) and supervised maximum likelihood were both used to discriminate paddy rice fields, with training areas automatically derived from ten-year cultivation parcels in Taiwan. Besides original bands in multi-spectral images, we also generated normalized difference vegetation index and experimented with object-based pre-classification and post-classification. This paper discusses results of different image classification methods in an attempt to find a precise and automatic solution to mapping paddy rice in Taiwan.

Keywords: paddy rice fields; multi-temporal; FORMOSAT-2images, normalized difference vegetation index, object-basedclassification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1802

Authors: Lo Kar Yin, Law Ka Mei

Abstract:

Architectural, engineering, construction and operations (AECO) industry practitioners have been well adapting to the dynamic construction market from the fundamental training of its disciplines. As further triggered by the pandemic since 2019, great steps are taken in virtual environment and the best collaboration is strived with project teams without boundaries. With adoption of Building Information Modeling-based approach and qualitative analysis, this paper is to review quantity take-off (QTO) and cost estimation process through modeling techniques in liaison with suppliers, fabricators, subcontractors, contractors, designers, consultants and services providers in the construction industry value chain for automatic project cost budgeting, project cost control and cost evaluation on design options of in-situ reinforced-concrete construction and Modular Integrated Construction (MiC) at design stage, variation of works and cash flow/spending analysis at construction stage as far as practicable, with a view to sharing the findings for enhancing mutual trust and co-operation among AECO industry practitioners. It is to foster development through a common prototype of design and build project delivery method in NEC4 Engineering and Construction Contract (ECC) Options A and C.

Keywords: Building Information Modeling, cost estimation, quantity take-off, modeling techniques.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 732

Authors: Sergey Kopysov, Nikita Nedozhogin, Leonid Tonkov

Abstract:

The article presents a parallel iterative solver for large sparse linear systems which can be used on a heterogeneous platform. Traditionally, the problem of solving linear systems do not scale well on cluster containing multiple Central Processing Units (multi-CPUs cluster) or cluster containing multiple Graphics Processing Units (multi-GPUs cluster). For example, most of the attempts to implement the classical conjugate gradient method were at best counted in the same amount of time as the problem was enlarged. The paper proposes the pipelined variant of the conjugate gradient method (PCG), a formulation that is potentially better suited for hybrid CPU/GPU computing since it requires only one synchronization point per one iteration, instead of two for standard CG (Conjugate Gradient). The standard and pipelined CG methods need the vector entries generated by current GPU and other GPUs for matrix-vector product. So the communication between GPUs becomes a major performance bottleneck on miltiGPU cluster. The article presents an approach to minimize the communications between parallel parts of algorithms. Additionally, computation and communication can be overlapped to reduce the impact of data exchange. Using pipelined version of the CG method with one synchronization point, the possibility of asynchronous calculations and communications, load balancing between the CPU and GPU for solving the large linear systems allows for scalability. The algorithm is implemented with the combined use of technologies: MPI, OpenMP and CUDA. We show that almost optimum speed up on 8-CPU/2GPU may be reached (relatively to a one GPU execution). The parallelized solver achieves a speedup of up to 5.49 times on 16 NVIDIA Tesla GPUs, as compared to one GPU.

Keywords: Conjugate Gradient, GPU, parallel programming, pipelined algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 393

Authors: Hamidreza Hoshyarmanesh, Hosein Madieh, Sanju Lama, Yaser Maddahi, Garnette R. Sutherland, Kourosh Zareinia

Abstract:

This paper illustrates the concept of an algorithm to register specified markers on the neuroArm surgical manipulators, an image-guided MR-compatible tele-operated robot for microsurgery and stereotaxy. Two range-finding algorithms, namely time-of-flight and phase-shift, are evaluated for registration and supervisory control. The time-of-flight approach is implemented in a semi-field experiment to determine the precise position of a tiny retro-reflective moving object. The moving object simulates a surgical tool tip. The tool is a target that would be connected to the neuroArm end-effector during surgery inside the magnet bore of the MR imaging system. In order to apply flight approach, a 905-nm pulsed laser diode and an avalanche photodiode are utilized as the transmitter and receiver, respectively. For the experiment, a high frequency time to digital converter was designed using a field-programmable gate arrays. In the phase-shift approach, a continuous green laser beam with a wavelength of 530 nm was used as the transmitter. Results showed that a positioning error of 0.1 mm occurred when the scanner-target point distance was set in the range of 2.5 to 3 meters. The effectiveness of this non-contact approach exhibited that the method could be employed as an alternative for conventional mechanical registration arm. Furthermore, the approach is not limited by physical contact and extension of joint angles.

Keywords: 3D laser scanner, intraoperative MR imaging, neuroArm, real time registration, robot-assisted surgery, supervisory control.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1076

Authors: Konstantinos Perifanos, Eirini Florou, Dionysis Goutsos

Abstract:

This paper presents and benchmarks a number of end-to-end Deep Learning based models for metaphor detection in Greek. We combine Convolutional Neural Networks and Recurrent Neural Networks with representation learning to bear on the metaphor detection problem for the Greek language. The models presented achieve exceptional accuracy scores, significantly improving the previous state-of-the-art results, which had already achieved accuracy 0.82. Furthermore, no special preprocessing, feature engineering or linguistic knowledge is used in this work. The methods presented achieve accuracy of 0.92 and F-score 0.92 with Convolutional Neural Networks (CNNs) and bidirectional Long Short Term Memory networks (LSTMs). Comparable results of 0.91 accuracy and 0.91 F-score are also achieved with bidirectional Gated Recurrent Units (GRUs) and Convolutional Recurrent Neural Nets (CRNNs). The models are trained and evaluated only on the basis of training tuples, the related sentences and their labels. The outcome is a state-of-the-art collection of metaphor detection models, trained on limited labelled resources, which can be extended to other languages and similar tasks.

Keywords: Metaphor detection, deep learning, representation learning, embeddings.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 572

Authors: Paolo Pilloni, Fabrizio Mulas, Salvatore Carta

Abstract:

Nowadays, several research studies point up that an active lifestyle is essential for physical and mental health benefits. Mobile phones have greatly influenced people’s habits and attitudes also in the way they exercise. Our research work is mainly focused on investigating how to exploit mobile technologies to favour people’s exertion experience. To this end, we developed an exertion framework users can exploit through a real world mobile application, called EverywhereSport Run (EWRun), designed to act as a virtual personal trainer to support runners during their trainings. In this work, inspired by both previous findings in the field of interaction design for people with visual impairments, feedback gathered from real users of our framework, and positive results obtained from two experimentations, we present some new interaction facilities we designed to enhance the interaction experience during a training. The positive obtained results helped us to derive some interaction design recommendations we believe will be a valid support for designers of future mobile systems conceived to be used in circumstances where there are limited possibilities of interaction.

Keywords: Human Computer Interaction, Interaction Design Guidelines, Persuasive Mobile Technologies for Sport and Health.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1954

Authors: Sarah K. Taylor, Miratun M. Saharuddin, Zabri A. Talib

Abstract:

Cybercrime is on the rise, and yet many Law Enforcement Agencies (LEAs) in Malaysia have no Digital Forensics Laboratory (DFL) to assist them in the attrition and analysis of digital evidence. From the estimated number of 30 LEAs in Malaysia, sadly, only eight of them owned a DFL. All of the DFLs are concentrated in the capital of Malaysia and none at the state level. LEAs are still depending on the national DFL (CyberSecurity Malaysia) even for simple and straightforward cases. A survey was conducted among LEAs in Malaysia owning a DFL to understand their history of establishing the DFL, the challenges that they faced and the significance of the DFL to their case investigation. The results showed that the while some LEAs faced no challenge in establishing a DFL, some of them took seven to 10 years to do so. The reason was due to the difficulty in convincing their management because of the high costs involved. The results also revealed that with the establishment of a DFL, LEAs were better able to get faster forensic result and to meet agency’s timeline expectation. It is also found that LEAs were also able to get more meaningful forensic results on cases that require niche expertise, compared to sending off cases to the national DFL. Other than that, cases are getting more complex, and hence, a continuous stream of budget for equipment and training is inevitable. The result derived from the study is hoped to be used by other LEAs in justifying to their management the benefits of establishing an in-house DFL.

Keywords: Digital forensics, digital forensics laboratory, digital evidence, law enforcement agency.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1514

Authors: Vitomir Struc, Nikola Pavesic

Abstract:

Over the past decades, automatic face recognition has become a highly active research area, mainly due to the countless application possibilities in both the private as well as the public sector. Numerous algorithms have been proposed in the literature to cope with the problem of face recognition, nevertheless, a group of methods commonly referred to as appearance based have emerged as the dominant solution to the face recognition problem. Many comparative studies concerned with the performance of appearance based methods have already been presented in the literature, not rarely with inconclusive and often with contradictory results. No consent has been reached within the scientific community regarding the relative ranking of the efficiency of appearance based methods for the face recognition task, let alone regarding their susceptibility to appearance changes induced by various environmental factors. To tackle these open issues, this paper assess the performance of the three dominant appearance based methods: principal component analysis, linear discriminant analysis and independent component analysis, and compares them on equal footing (i.e., with the same preprocessing procedure, with optimized parameters for the best possible performance, etc.) in face verification experiments on the publicly available XM2VTS database. In addition to the comparative analysis on the XM2VTS database, ten degraded versions of the database are also employed in the experiments to evaluate the susceptibility of the appearance based methods on various image degradations which can occur in "real-life" operating conditions. Our experimental results suggest that linear discriminant analysis ensures the most consistent verification rates across the tested databases.

Keywords: Biometrics, face recognition, appearance based methods, image degradations, the XM2VTS database.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2293

Authors: Sarah Beaver, Renee Bryce

Abstract:

Audiograms detect hearing impairment, but missing values pose problems. This work explores imputations in an attempt to improve accuracy. This work implements Linear Regression, Lasso, Linear Support Vector Regression, Bayesian Ridge, K Nearest Neighbors (KNN), and Random Forest machine learning techniques to impute audiogram frequencies ranging from 125 Hz to 8000 Hz. The data contain patients who had or were candidates for cochlear implants. Accuracy is compared across two different Nested Cross-Validation k values. Over 4000 audiograms were used from 800 unique patients. Additionally, training on data combines and compares left and right ear audiograms versus single ear side audiograms. The accuracy achieved using Root Mean Square Error (RMSE) values for the best models for Random Forest ranges from 4.74 to 6.37. The R2 values for the best models for Random Forest ranges from .91 to .96. The accuracy achieved using RMSE values for the best models for KNN ranges from 5.00 to 7.72. The R2 values for the best models for KNN ranges from .89 to .95. The best imputation models received R2 between .89 to .96 and RMSE values less than 8dB. We also show that the accuracy of classification predictive models performed better with our imputation models versus constant imputations by a two percent increase.

Keywords: Machine Learning, audiograms, data imputations, single imputations.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 189

Authors: G. Meyer, M. Klewer, P. Nyhuis

Abstract:

Constantly changing economic conditions require companies to design their production to be more economical, innovative, and flexible. Since workers have a decisive influence on cost, time, and quality, e.g. by monitoring indicators that determine quality, by developing processes more resistant to disturbances, or by monitoring environmental standards, a focus on personnel as a production factor is needed. This presupposes the efficient use and systematic enhancement of employees’ existing competences since greater consideration of these aspects in work planning will help to enhance competitiveness. The aim of the research project ‘Integrated Technology- and Competence-based Work Planning in Socio-Technical Systems’ is to develop a new work planning method that combines technology with work science by incorporating employees’ skills as a quality indicator. For employee competences to increase competitiveness, it is first of all necessary to assess how competences affect cost, time, and quality. A model for deriving predictions about the effects of competence-based parameters on these strategic business objectives is developed in this paper.

Keywords: Competence management, education and training, employee competences, one-factor-at-a-time method, work planning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1550

Authors: Roberto Sabatini, Alessandro Gardi, Mark A. Richardson

Abstract:

The availability of powerful eye-safe laser sources and the recent advancements in electro-optical and mechanical beam-steering components have allowed laser-based Light Detection and Ranging (LIDAR) to become a promising technology for obstacle warning and avoidance in a variety of manned and unmanned aircraft applications. LIDAR outstanding angular resolution and accuracy characteristics are coupled to its good detection performance in a wide range of incidence angles and weather conditions, providing an ideal obstacle avoidance solution, which is especially attractive in low-level flying platforms such as helicopters and small-to-medium size Unmanned Aircraft (UA). The Laser Obstacle Avoidance Marconi (LOAM) system is one of such systems, which was jointly developed and tested by SELEX-ES and the Italian Air Force Research and Flight Test Centre. The system was originally conceived for military rotorcraft platforms and, in this paper, we briefly review the previous work and discuss in more details some of the key development activities required for integration of LOAM on UA platforms. The main hardware and software design features of this LOAM variant are presented, including a brief description of the system interfaces and sensor characteristics, together with the system performance models and data processing algorithms for obstacle detection, classification and avoidance. In particular, the paper focuses on the algorithm proposed for optimal avoidance trajectory generation in UA applications.

Keywords: LIDAR, Low-Level Flight, Nap-of-the-Earth Flight, Near Infra-Red, Obstacle Avoidance, Obstacle Detection, Obstacle Warning System, Sense and Avoid, Trajectory Optimisation, Unmanned Aircraft.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7093

Authors: Mylena McCoggle, Shyra Wilson, Andrea Rivera, Rocio Alba-Flores, Valentin Soloiu

Abstract:

This work describes a system that uses electromyography (EMG) signals obtained from muscle sensors and an Artificial Neural Network (ANN) for signal classification and pattern recognition that is used to control a small unmanned aerial vehicle using specific arm movements. The main objective of this endeavor is the development of an intelligent interface that allows the user to control the flight of a drone beyond direct manual control. The sensor used were the MyoWare Muscle sensor which contains two EMG electrodes used to collect signals from the posterior (extensor) and anterior (flexor) forearm, and the bicep. The collection of the raw signals from each sensor was performed using an Arduino Uno. Data processing algorithms were developed with the purpose of classifying the signals generated by the arm’s muscles when performing specific movements, namely: flexing, resting, and motion of the arm. With these arm motions roll control of the drone was achieved. MATLAB software was utilized to condition the signals and prepare them for the classification. To generate the input vector for the ANN and perform the classification, the root mean square and the standard deviation were processed for the signals from each electrode. The neuromuscular information was trained using an ANN with a single 10 neurons hidden layer to categorize the four targets. The result of the classification shows that an accuracy of 97.5% was obtained. Afterwards, classification results are used to generate the appropriate control signals from the computer to the drone through a Wi-Fi network connection. These procedures were successfully tested, where the drone responded successfully in real time to the commanded inputs.

Keywords: Biosensors, electromyography, Artificial Neural Network, Arduino, drone flight control, machine learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 572

Authors: Blessing Ojeme, Frederick Quinn, Russell Karls, Shannon Quinn

Abstract:

The detection and segmentation of mitochondria from fluorescence microscopy is crucial for understanding the complex structure of the nervous system. However, the constant fission and fusion of mitochondria and image distortion in the background make the task of detection and segmentation challenging. Although there exists a number of open-source software tools and artificial intelligence (AI) methods designed for analyzing mitochondrial images, the availability of only a few combined expertise in the medical field and AI required to utilize these tools poses a challenge to its full adoption and use in clinical settings. Motivated by the advantages of automated methods in terms of good performance, minimum detection time, ease of implementation, and cross-platform compactibility, this study proposes a fully automated framework for the detection and segmentation of mitochondria using both image shape information and descriptive statistics. Using the low-cost, open-source Python and OpenCV library, the algorithms are implemented in three stages: pre-processing; image binarization; and coarse-to-fine segmentation. The proposed model is validated using the fluorescence mitochondrial dataset. Ground truth labels generated using Labkit were also used to evaluate the performance of our detection and segmentation model using precision, recall and rand index. The study produces good detection and segmentation results and reports the challenges encountered during the image analysis of mitochondrial morphology from the fluorescence mitochondrial dataset. A discussion on the methods and future perspectives of fully automated frameworks concludes the paper.

Keywords: 2D, Binarization, CLAHE, detection, fluorescence microscopy, mitochondria, segmentation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 494

Authors: Eric P. Jiang

Abstract:

As the Internet continues to grow at a rapid pace as the primary medium for communications and commerce and as telecommunication networks and systems continue to expand their global reach, digital information has become the most popular and important information resource and our dependence upon the underlying cyber infrastructure has been increasing significantly. Unfortunately, as our dependency has grown, so has the threat to the cyber infrastructure from spammers, attackers and criminal enterprises. In this paper, we propose a new machine learning based network intrusion detection framework for cyber security. The detection process of the framework consists of two stages: model construction and intrusion detection. In the model construction stage, a semi-supervised machine learning algorithm is applied to a collected set of network audit data to generate a profile of normal network behavior and in the intrusion detection stage, input network events are analyzed and compared with the patterns gathered in the profile, and some of them are then flagged as anomalies should these events are sufficiently far from the expected normal behavior. The proposed framework is particularly applicable to the situations where there is only a small amount of labeled network training data available, which is very typical in real world network environments.

Keywords: classification, data analysis and mining, network intrusion detection, semi-supervised learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1804