Search results for: Hash algorithm

Authors: R. S. Remya, U. S. Sethulekshmi

Abstract:

Detecting the authenticity of a video is an important issue in digital forensics as Video is used as a silent evidence in court such as in child pornography, movie piracy cases, insurance claims, cases involving scientific fraud, traffic monitoring etc. The biggest threat to video data is the availability of modern open video editing tools which enable easy editing of videos without leaving any trace of tampering. In this paper, we propose an efficient passive method for inter-frame video tampering detection, its type and location by estimating the optical flow of wavelet features of adjacent frames and thresholding the variation in the estimated feature. The performance of the algorithm is compared with the z-score thresholding and achieved an efficiency above 95% on all the tested databases. The proposed method works well for videos with dynamic (forensics) as well as static (surveillance) background.

Keywords: discrete wavelet transform, optical flow, optical flow variation, video tampering

Procedia PDF Downloads 339

Authors: Lilah Inzelberg, David Rand, Stanislav Steinberg, Moshe David Pur, Yael Hanein

Abstract:

Human facial expressions carry important psychological and neurological information. Facial expressions involve the co-activation of diverse muscles. They depend strongly on personal affective interpretation and on social context and vary between spontaneous and voluntary activations. Smiling, as a special case, is among the most complex facial emotional expressions, involving no fewer than 7 different unilateral muscles. Despite their ubiquitous nature, smiles remain an elusive and debated topic. Smiles are associated with happiness and greeting on one hand and anger or disgust-masking on the other. Accordingly, while high-resolution recording of muscle activation patterns, in a non-interfering setting, offers exciting opportunities, it remains an unmet challenge, as contemporary surface facial electromyography (EMG) methodologies are cumbersome, restricted to the laboratory settings, and are limited in time and resolution. Here we present a wearable and non-invasive method for objective mapping of facial muscle activation and demonstrate its application in a natural setting. The technology is based on a recently developed dry and soft electrode array, specially designed for surface facial EMG technique. Eighteen healthy volunteers (31.58 ± 3.41 years, 13 females), participated in the study. Surface EMG arrays were adhered to participant left and right cheeks. Participants were instructed to imitate three facial expressions: closing the eyes, wrinkling the nose and smiling voluntary and to watch a funny video while their EMG signal is recorded. We focused on muscles associated with 'enjoyment', 'social' and 'masked' smiles; three categories with distinct social meanings. We developed a customized independent component analysis algorithm to construct the desired facial musculature mapping. First, identification of the Orbicularis oculi and the Levator labii superioris muscles was demonstrated from voluntary expressions. Second, recordings of voluntary and spontaneous smiles were used to locate the Zygomaticus major muscle activated in Duchenne and non-Duchenne smiles. Finally, recording with a wireless device in an unmodified natural work setting revealed expressions of neutral, positive and negative emotions in face-to-face interaction. The algorithm outlined here identifies the activation sources in a subject-specific manner, insensitive to electrode placement and anatomical diversity. Our high-resolution and cross-talk free mapping performances, along with excellent user convenience, open new opportunities for affective processing and objective evaluation of facial expressivity, objective psychological and neurological assessment as well as gaming, virtual reality, bio-feedback and brain-machine interface applications.

Keywords: affective expressions, affective processing, facial EMG, high-resolution electromyography, independent component analysis, wireless electrodes

Procedia PDF Downloads 226

Authors: Shih-Kuei Lin

Abstract:

The characterization of the arbitrage-free dynamics of interest rates is developed in this study under the presence of Markov jump risks, when the term structure of the interest rates is modeled through simple forward rates. We consider Markov jump risks by allowing randomness in jump sizes, independence between jump sizes and jump times. The Markov jump diffusion model is used to capture empirical phenomena and to accurately describe interest jump risks in a financial market. We derive the arbitrage-free model of simple forward rates under the spot measure. Moreover, the analytical pricing formulas for a cap and a floor are derived under the forward measure when the jump size follows a lognormal distribution. In our empirical analysis, we find that the LIBOR market model with Markov jump risk better accounts for changes from/to different states and different rates.

Keywords: arbitrage-free, cap and floor, Markov jump diffusion model, simple forward rate model, volatility smile, EM algorithm

Procedia PDF Downloads 407

Authors: Akella Amarendra Babu, Rama Devi Yellasiri, Natukula Sainath

Abstract:

Where human beings can easily learn and adopt pronunciation variations, machines need training before put into use. Also humans keep minimum vocabulary and their pronunciation variations are stored in front-end of their memory for ready reference, while machines keep the entire pronunciation dictionary for ready reference. Supervised methods are used for preparation of pronunciation dictionaries which take large amounts of manual effort, cost, time and are not suitable for real time use. This paper presents an unsupervised adaptation model for building agile and dynamic pronunciation dictionaries online. These methods mimic human approach in learning the new pronunciations in real time. A new algorithm for measuring sound distances called Dynamic Phone Warping is presented and tested. Performance of the system is measured using an adaptation model and the precision metrics is found to be better than 86 percent.

Keywords: pronunciation variations, dynamic programming, machine learning, natural language processing

Procedia PDF Downloads 154

Authors: Amir Lotfi, Seyed Hamid Hosseini

Abstract:

During the recent past, due to the increase of electrical energy demand and governmental resources constraints in creating additional capacity in the generation, transmission, and distribution, privatization, and restructuring in electrical industry have been considered. So, in most of the countries, different parts of electrical industry like generation, transmission, and distribution have been separated in order to create competition. Considering these changes, environmental issues, energy growth, investment of private equity in energy generation units and difficulties of transmission lines expansion, distributed generation (DG) units have been used in power systems. Moreover, reduction in the need for transmission and distribution, the increase of reliability, improvement of power quality, and reduction of power loss have caused DG to be placed in power systems. On the other hand, considering low liquidity need, private investors tend to spend their money for DGs. In this project, the main goal is to offer an algorithm for planning and placing DGs in order to reduce the need for transmission and distribution network.

Keywords: planning, transmission, distributed generation, power security, power systems

Procedia PDF Downloads 462

Authors: Indrajit Ghosh, Jaydip Roy

Abstract:

Literacy rate is an important indicator for measurement of human development. But this is not a good one to capture the qualitative dimension of educational attainment of an individual or a society. The overall educational level of an area is an important issue beyond the literacy rate. The overall educational level can be thought of as an outcome of the educational levels of individuals. But there is no well-defined algorithm and mathematical model available to measure the overall educational level of an area. A heuristic approach based on accumulated experience of experts is effective one. It is evident that fuzzy logic offers a natural and convenient framework in modeling various concepts in social science domain. This work suggests the implementation of fuzzy logic to develop a mathematical model for measurement of educational attainment of an area in terms of Education Index. The contribution of the study is two folds: conceptualization of “Education Profile” and proposing a new mathematical model to measure educational attainment in terms of “Education Index”.

Keywords: education index, education profile, fuzzy logic, literacy

Procedia PDF Downloads 299

Authors: Zdzislaw Piotr Kaminski, Miroslaw Wendeker, Zbigniew Czyz

Abstract:

The paper describes the geometric model, calculation algorithm and results of the CFD simulation of the airflow around a rotor in the vertical axis wind turbine (VAWT) with the ANSYS Fluent computational solver. The CFD method enables creating aerodynamic characteristics of forces acting on rotor working surfaces and determining parameters such as torque or power generated by the rotor assembly. The object of the research was a rotor whose construction is based on patent no.PL219985. The conducted tests enabled a mathematical model with a description of the generation of aerodynamic forces acting on each rotor blade. Additionally, this model was compared to the results of the wind tunnel tests. The analysis also focused on the influence of the blade angle on turbine power and the TSR. The research has shown that the turbine blade angle has a significant impact on the optimal value of the TSR.

Keywords: computational fluid dynamics, numerical analysis, renewable energy, wind turbine

Procedia PDF Downloads 134

Authors: Hoi-Lam Ma, Sai-Ho Chung

Abstract:

In traditional integrated berth allocation with quay crane assignment models, time dimension is usually assumed in hourly based. However, nowadays, transshipment becomes the main business to many container terminals, especially in Southeast Asia (e.g. Hong Kong and Singapore). In these terminals, vessel arrivals are usually very frequent with small handling volume and very short staying time. Therefore, the traditional hourly-based modeling approach may cause significant berth and quay crane idling, and consequently cannot meet their practical needs. In this connection, a 15-minute-based modeling approach is requested by industrial practitioners. Accordingly, a Three-level Genetic Algorithm (3LGA) with Quay Crane (QC) shifting heuristics is designed to fulfill the research gap. The objective function here is to minimize the total service time. Preliminary numerical results show that the proposed 15-minute-based approach can reduce the berth and QC idling significantly.

Keywords: transshipment, integrated berth allocation, variable-in-time quay crane assignment, quay crane assignment

Procedia PDF Downloads 154

Authors: Rakesh Prajapati, Purvik Patel, Hardik Patel

Abstract:

Productivity and quality are two important aspects that have become great concerns in today’s competitive global market. Every production/manufacturing unit mainly focuses on these areas in relation to the process, as well as the product developed. The electrical discharge machining (EDM) process, even now it is an experience process, wherein the selected parameters are still often far from the maximum, and at the same time selecting optimization parameters is costly and time consuming. Material Removal Rate (MRR) during the process has been considered as a productivity estimate with the aim to maximize it, with an intention of minimizing surface roughness taken as most important output parameter. These two opposites in nature requirements have been simultaneously satisfied by selecting an optimal process environment (optimal parameter setting). Objective function is obtained by Regression Analysis and Analysis of Variance. Then objective function is optimized using Genetic Algorithm technique. The model is shown to be effective; MRR and Surface Roughness improved using optimized machining parameters.

Keywords: MMR, TWR, OC, DOE, ANOVA, minitab

Procedia PDF Downloads 306

Authors: George Kurian, Hongmei Chi

Abstract:

Digital evidence plays a critical role in most legal investigations. In many cases, thumbnail databases show important information in that investigation. The probability of having digital evidence retrieved from a computer or smart device has increased, even though the previous user removed data and deleted apps on those devices. Due to the increase in digital forensics, the ability to store residual information from various thumbnail applications has improved. This paper will focus on investigating thumbnail information from Windows 10. Thumbnail images of interest in forensic investigations may be intact even when the original pictures have been deleted. It is our research goal to recover useful information from thumbnails. In this research project, we use various forensics tools to collect left thumbnail information from deleted videos or pictures. We examine and describe the various thumbnail sources in Windows and propose a methodology for thumbnail collection and analysis from laptops or desktops. A machine learning algorithm is adopted to help speed up content from thumbnail pictures.

Keywords: digital forensic, forensic tools, soundness, thumbnail, machine learning, OCR

Procedia PDF Downloads 109

Authors: Zikiou Nadia, Lahdir Mourad, Ameur Soltane

Abstract:

In this paper, we propose an effective method for image compression based on SVM Regression (SVR), with three different kernels, and biorthogonal 2D Discrete Wavelet Transform. SVM regression could learn dependency from training data and compressed using fewer training points (support vectors) to represent the original data and eliminate the redundancy. Biorthogonal wavelet has been used to transform the image and the coefficients acquired are then trained with different kernels SVM (Gaussian, Polynomial, and Linear). Run-length and Arithmetic coders are used to encode the support vectors and its corresponding weights, obtained from the SVM regression. The peak signal noise ratio (PSNR) and their compression ratios of several test images, compressed with our algorithm, with different kernels are presented. Compared with other kernels, Gaussian kernel achieves better image quality. Experimental results show that the compression performance of our method gains much improvement.

Keywords: image compression, 2D discrete wavelet transform (DWT-2D), support vector regression (SVR), SVM Kernels, run-length, arithmetic coding

Procedia PDF Downloads 365

Authors: Arun Prasad Jaganathan

Abstract:

In today's dynamic industrial world, traditional task allocation methods often fall short in adapting to evolving operational conditions. This paper introduces an AI-enabled task assistance and management system designed to overcome the limitations of conventional approaches. By using artificial intelligence (AI) and machine learning (ML), the system intelligently interprets user instructions, analyzes tasks, and allocates resources based on real-time data and environmental factors. Additionally, geolocation tracking enables proactive identification of potential delays, ensuring timely interventions. With its transparent reporting mechanisms, the system provides stakeholders with clear insights into task progress, fostering accountability and informed decision-making. The paper presents a comprehensive overview of the system architecture, algorithm, and implementation, highlighting its potential to revolutionize task management across diverse industries.

Keywords: artificial intelligence, machine learning, task allocation, operational efficiency, resource optimization

Procedia PDF Downloads 28

Authors: Z. Kourdi, B. Bouazza, M. Khaouani, A. Guen-Bouazza, Z. Djennati, A. Boursali

Abstract:

We present a simulation of a HEMT (high electron mobility transistor) structure with and without a field plate. We extract the device characteristics through the analysis of DC, AC and high frequency regimes, as shown in this paper. This work demonstrates the optimal device with a gate length of 15 nm, InAlN/GaN heterostructure and field plate structure, making it superior to modern HEMTs when compared with otherwise equivalent devices. This improves the ability to bear the burden of the current density passes in the channel. We have demonstrated an excellent current density, as high as 2.05 A/mm, a peak extrinsic transconductance of 590 mS/mm at VDS=2 V, and cutting frequency cutoffs of 638 GHz in the first HEMT and 463 GHz for Field plate HEMT., maximum frequency of 1.7 THz, maximum efficiency of 73%, maximum breakdown voltage of 400 V, DIBL=33.52 mV/V and an ON/OFF current density ratio higher than 1 x 1010. These values were determined through the simulation by deriving genetic and Monte Carlo algorithms that optimize the design and the future of this technology.

Keywords: HEMT, Silvaco, field plate, genetic algorithm, quantum

Procedia PDF Downloads 450

Authors: Pardeep Singh, Kamlesh Dutta

Abstract:

Many languages have fixed sentence structure and others are free word order. The accuracy of anaphora resolution of syntax based algorithm depends on structure of the sentence. So, it is important to analyze the structure of any language before implementing these algorithms. In this study, we analyzed the sentence structure exploiting the case marker in Hindi as well as some special tag for subject and object. We also investigated the word order for Hindi. Word order typology refers to the study of the order of the syntactic constituents of a language. We analyzed 165 news items of Ranchi Express from EMILEE corpus of plain text. It consisted of 1745 sentences. Eight file of dialogue based from the same corpus has been analyzed which will have 1521 sentences. The percentages of subject object verb structure (SOV) and object subject verb (OSV) are 66.90 and 33.10, respectively.

Keywords: anaphora resolution, free word order languages, SOV, OSV

Procedia PDF Downloads 456

Authors: Muhammad Shahzad Shifa, Amna Shifa, Muhammad Omar, Aamir Shahzad, Rahmat Ali Khan

Abstract:

Many applications of remote sensing to vegetation and crop response depend on spectral properties of individual leaves and plants. Vegetation indices are usually determined to estimate crop biophysical parameters like crop canopies and crop leaf area indices with the help of remote sensing. Cotton crops assessment is performed with the help of vegetative indices. Remotely sensed images from an optical multispectral radiometer MSR5 are used in this study. The interpretation is based on the fact that different materials reflect and absorb light differently at different wavelengths. Non-normalized and normalized forms of these datasets are analyzed using two complementary data mining algorithms; K-means and K-nearest neighbor (KNN). Our analysis shows that the use of normalized reflectance data and vegetative indices are suitable for an automated assessment and decision making.

Keywords: cotton, condition assessment, KNN algorithm, clustering, MSR5, vegetation indices

Procedia PDF Downloads 314

Authors: Ines Gasmi, Salima Smiti, Makram Soui, Khaled Ghedira

Abstract:

Credit risk is considered as one of the important issues for financial institutions. It provokes great losses for banks. To this objective, numerous methods for credit risk evaluation have been proposed. Many evaluation methods are black box models that cannot adequately reveal information hidden in the data. However, several works have focused on building transparent rules-based models. For credit risk assessment, generated rules must be not only highly accurate, but also highly interpretable. In this paper, we aim to build both, an accurate and transparent credit risk evaluation model which proposes a set of classification rules. In fact, we consider the credit risk evaluation as an optimization problem which uses a genetic programming (GP) algorithm, where the goal is to maximize the accuracy of generated rules. We evaluate our proposed approach on the base of German and Australian credit datasets. We compared our finding with some existing works; the result shows that the proposed GP outperforms the other models.

Keywords: credit risk assessment, rule generation, genetic programming, feature selection

Procedia PDF Downloads 327

Authors: M. H. Fazel Zarandi, N. Moshahedi

Abstract:

The hub location problem appears in a variety of applications such as medical centers, firefighting facilities, cargo delivery systems and telecommunication network design. The location of service centers has a strong influence on the congestion at each of them, and, consequently, on the quality of service. This paper presents a fuzzy maximal hub covering location problem (FMCHLP) in which travel costs between any pair of nodes is considered as a fuzzy variable. In order to consider the quality of service, we model each hub as a queue. Arrival rate follows Poisson distribution and service rate follows Erlang distribution. In this paper, at first, a nonlinear mathematical programming model is presented. Then, we convert it to the linear one. We solved the linear model using GAMS software up to 25 nodes and for large sizes due to the complexity of hub covering location problems, and simulated annealing algorithm is developed to solve and test the model. Also, we used possibilistic c-means clustering method in order to find an initial solution.

Keywords: fuzzy modeling, location, possibilistic clustering, queuing

Procedia PDF Downloads 378

Authors: Mohammed A. M. Ibrahim

Abstract:

This paper investigate the utilization of wire and wireless networks to be platform for distributed educational monitoring system. Universities in developing countries suffer from a lot of shortages(staff, equipment, and finical budget) and optimal utilization of the wire and wireless network, so universities can mitigate some of the mentioned problems and avoid the problems that maybe humble the education processes in many universities by using our implementation of the examinations system as a test-bed to utilize the network as a solution to the shortages for academic staff in Taiz University. This paper selects a two areas first one quizzes activities is only a test bed application for wireless network learning environment system to be distributed among students. Second area is the features and the security of wireless, our tested application implemented in a promising area which is the use of WLAN in higher education for leering environment.

Keywords: networking wire and wireless technology, wireless network security, distributed computing, algorithm, encryption and decryption

Procedia PDF Downloads 319

Authors: An-Jui. Li, Kelvin Lim, Chien-Kuo Chiu, Benson Hsiung

Abstract:

This paper details the utilization of artificial intelligence (AI) in the field of slope stability whereby quick and convenient solutions can be obtained using the developed tool. The AI tool used in this study is the artificial neural network (ANN), while the slope stability analysis methods are the finite element limit analysis methods. The developed tool allows for the prompt prediction of the safety factors of fill slopes and their corresponding probability of failure (depending on the degree of variation of the soil parameters), which can give the practicing engineer a reasonable basis in their decision making. In fact, the successful use of the Extreme Learning Machine (ELM) algorithm shows that slope stability analysis is no longer confined to the conventional methods of modeling, which at times may be tedious and repetitive during the preliminary design stage where the focus is more on cost saving options rather than detailed design. Therefore, similar ANN-based tools can be further developed to assist engineers in this aspect.

Keywords: landslide, limit analysis, artificial neural network, soil properties

Procedia PDF Downloads 184

Authors: Mefteh Bouhalleb

Abstract:

With the recent progress in nanotechnology, nanofluids have excellent potentiality in many modern engineering processes, particularly for solar systems such as concentrated solar power plants (CSP). In this context, a numerical simulation is performed to investigate laminar natural convection nanofluids in an inclined rectangular enclosure. Mass conservation, momentum, and energy equations are numerically solved by the finite volume element method using the SIMPLER algorithm for pressure-velocity coupling. In this work, we tested the acting factors on the system response time, such as the particle volume fraction of nanoparticles, particle material, particle size, an inclination angle of enclosure and Rayleigh number. The results show that the diameter of solid particles and Rayleigh number plays an important role in the system response time. The orientation angle of the cavity affects the system response time. A phenomenon of hysteresis appears when the system does not return to its initial state.

Keywords: nanofluid, nanoparticles, heat transfer, time response

Procedia PDF Downloads 80

Authors: Yue Hu, Xilu Zhao, Takao Yamaguchi, Manabu Sasajima, Yoshio Koike, Akira Hara

Abstract:

This study tried optimization of design parameter of a cone loudspeaker unit as an example of the high flexibility of the products design. We developed an acoustic analysis software program that considers the impact of damping caused by air viscosity. In sound reproduction, it is difficult to each design the parameter of the loudspeaker. To overcome the limitation of the design problem in practice, this paper proposes a new an acoustic analysis algorithm to optimize design the parameter of the loudspeaker. The material character of cone paper and the loudspeaker edge was the design parameter, and the vibration displacement of the cone paper was the objective function. The results of the analysis were compared with the predicted value. They had high accuracy to the predicted value. These results suggest that, though the parameter design is difficult by experience and intuition, it can be performed comparatively easily using the optimization design by the developed acoustic analysis software.

Keywords: air viscosity, loudspeaker, cone paper, edge, optimization

Procedia PDF Downloads 382

Authors: J. K. Alhassan, S. A. Lawal

Abstract:

This work is on decision tree-based classification for the disbursement of scholarship. Tree-based data mining classification technique is used in other to determine the generic rule to be used to disburse the scholarship. The system based on the defined rules from the tree is able to determine the class (status) to which an applicant shall belong whether Granted or Not Granted. The applicants that fall to the class of granted denote a successful acquirement of scholarship while those in not granted class are unsuccessful in the scheme. An algorithm that can be used to classify the applicants based on the rules from tree-based classification was also developed. The tree-based classification is adopted because of its efficiency, effectiveness, and easy to comprehend features. The system was tested with the data of National Information Technology Development Agency (NITDA) Abuja, a Parastatal of Federal Ministry of Communication Technology that is mandated to develop and regulate information technology in Nigeria. The system was found working according to the specification. It is therefore recommended for all scholarship disbursement organizations.

Keywords: classification, data mining, decision tree, scholarship

Procedia PDF Downloads 351

Authors: N. Beit Saeid

Abstract:

Cyclones are used in the field of air industrial gases pollution and control the pollution with centrifugal forces that is generated with spatial geometry of the cyclone. Their simple design, low capital and maintenance costs and adaptability to a wide range of operating conditions have made cyclones one of the most widely used industrial dust collectors. Their cost of operation is proportional to the fan energy required to overcome their pressure drop. Optimized geometry of outlet diffuser of the cyclones potentially could reduce exit pressure losses without affecting collection efficiency. Three rectangular outlets and a radial outlet with a variable opening had been analyzed on two cyclones. Pressure drop was investigated for inlet velocities from about 10 to 20 m s−1. The radial outlet reduced cyclone pressure drop by between 8.7 and 11.9 percent when its exit area was equal to the flow area of the cyclone vortex finder or gas exit. A simple payback based on avoided energy costs was estimated to be between 3600 and 5000 h, not including installation cost.

Keywords: cyclone, CFD, optimization, genetic algorithm

Procedia PDF Downloads 366

Authors: Ratun Rahman, Rafid Islam, Akin Ahmed, Kamrul Hasan, Hasan Mahmud

Abstract:

Malware is becoming more prevalent, and several threat categories have risen dramatically in recent years. This paper provides a bird's-eye view of the world of malware analysis. The efficiency of five different machine learning methods (Naive Bayes, K-Nearest Neighbor, Decision Tree, Random Forest, and TensorFlow Decision Forest) combined with features picked from the retrieval of Android permissions to categorize applications as harmful or benign is investigated in this study. The test set consists of 1,168 samples (among these android applications, 602 are malware and 566 are benign applications), each consisting of 948 features (permissions). Using the permission-based dataset, the machine learning algorithms then produce accuracy rates above 80%, except the Naive Bayes Algorithm with 65% accuracy. Of the considered algorithms TensorFlow Decision Forest performed the best with an accuracy of 90%.

Keywords: android malware detection, machine learning, malware, malware analysis

Procedia PDF Downloads 136

Authors: James Purnama, Riri Fitri Sari

Abstract:

We investigate the machine learning algorithm selection problem in the term of a depth image based eye gaze estimation, with respect to its essential difficulty in reducing the number of required training samples and duration time of training. Statistics based prediction accuracy are increasingly used to assess and evaluate prediction or estimation in gaze estimation. This article evaluates Root Mean Squared Error (RMSE) and R-Squared statistical analysis to assess machine learning methods on depth camera data for gaze estimation. There are 4 machines learning methods have been evaluated: Random Forest Regression, Regression Tree, Support Vector Machine (SVM), and Linear Regression. The experiment results show that the Random Forest Regression has the lowest RMSE and the highest R-Squared, which means that it is the best among other methods.

Keywords: gaze estimation, gaze tracking, eye tracking, kinect, regression model, orange python

Procedia PDF Downloads 521

Authors: E. E. EL-Kholy, Ahamed Kalas, Mahmoud Fauzy, M. El-Shahat Dessouki, Abdou M. El-refay, Mohammed El-Zefery

Abstract:

Direct Torque Control (DTC) is an AC drive control method especially designed to provide fast and robust responses. In this paper a progressive algorithm for direct torque control of three-phase induction drive system supplied by photovoltaic arrays using voltage source inverter to control motor torque and flux with maximum power point tracking at different level of insolation is presented. Experimental results of the new DTC method obtained by an experimental rapid prototype system for drives are presented. Simulation and experimental results confirm that the proposed system gives quick, robust torque and speed responses at constant switching frequencies.

Keywords: photovoltaic (PV) array, direct torque control (DTC), constant switching frequency, induction motor, maximum power point tracking (MPPT)

Procedia PDF Downloads 465

Authors: Shuo Liu, Dan Liu

Abstract:

In the field of named entity recognition, most of the research has been conducted around simple entities. However, for nested named entities, which still contain entities within entities, it has been difficult to identify them accurately due to their boundary ambiguity. In this paper, a hierarchical recognition model is constructed based on the grammatical structure and semantic features of Chinese text for boundary calculation based on lexical features. The analysis is carried out at different levels in terms of granularity, semantics, and lexicality, respectively, avoiding repetitive work to reduce computational effort and using the semantic features of words to calculate the boundaries of entities to improve the accuracy of the recognition work. The results of the experiments carried out on web-based microblogging data show that the model achieves an accuracy of 86.33% and an F1 value of 89.27% in recognizing nested named entities, making up for the shortcomings of some previous recognition models and improving the efficiency of recognition of nested named entities.

Keywords: coarse-grained, nested named entity, Chinese natural language processing, word embedding, T-SNE dimensionality reduction algorithm

Procedia PDF Downloads 109

Authors: José A. Menezes, Giordano Cabral, Bruno T. Gomes

Abstract:

Choosing good features is an essential part of machine learning. Recent techniques aim to automate this process. For instance, feature learning intends to learn the transformation of raw data into a useful representation to machine learning tasks. In automatic audio classification tasks, this is interesting since the audio, usually complex information, needs to be transformed into a computationally convenient input to process. Another technique tries to generate features by searching a feature space. Genetic algorithms, for instance, have being used to generate audio features by combining or modifying them. We find this approach particularly interesting and, despite the undeniable advances of feature learning approaches, we wanted to take a step forward in the use of genetic algorithms to find audio features, combining them with more conventional methods, like PCA, and inserting search control mechanisms, such as constraints over a confusion matrix. This work presents the results obtained on particular audio classification problems.

Keywords: feature generation, feature learning, genetic algorithm, music information retrieval

Procedia PDF Downloads 415

Authors: Victor Onomza Waziri, John K. Alhassan, Idris Ismaila, Noel Dogonyara

Abstract:

This paper describes the problem of building secure computational services for encrypted information in the Cloud. Computing without decrypting the encrypted data; therefore, it meets the yearning of computational encryption algorithmic aspiration model that could enhance the security of big data for privacy or confidentiality, availability and integrity of the data and user’s security. The cryptographic model applied for the computational process of the encrypted data is the Fully Homomorphic Encryption Scheme. We contribute a theoretical presentations in a high-level computational processes that are based on number theory that is derivable from abstract algebra which can easily be integrated and leveraged in the Cloud computing interface with detail theoretic mathematical concepts to the fully homomorphic encryption models. This contribution enhances the full implementation of big data analytics based on cryptographic security algorithm.

Keywords: big data analytics, security, privacy, bootstrapping, Fully Homomorphic Encryption Scheme

Procedia PDF Downloads 459

Authors: Vincent Liu

Abstract:

Machine learning algorithms (ML) can achieve high accuracy in predicting outcomes compared to classical models. Metaheuristic, nature-inspired algorithms can enhance traditional ML algorithms by optimizing them such as by performing feature selection. We compare ten ML algorithms to predict 30-day hospital readmission rates for diabetes patients in the US using a dataset from UCI Machine Learning Repository with feature selection performed by Greywolf nature-inspired algorithm. The baseline accuracy for the initial random forest model was 65%. After performing feature engineering, SMOTE for class balancing, and Greywolf optimization, the machine learning algorithms showed better metrics, including F1 scores, accuracy, and confusion matrix with improvements ranging in 10%-30%, and a best model of XGBoost with an accuracy of 95%. Applying machine learning this way can improve patient outcomes as unnecessary rehospitalizations can be prevented by focusing on patients that are at a higher risk of readmission.

Keywords: diabetes, machine learning, 30-day readmission, metaheuristic

Procedia PDF Downloads 34