Search results for: metaheuristic algorithms

Authors: S. Guruprasad, M. Z. Kurian, H. N. Suma

Abstract:

Medical imaging modalities are becoming life saving components. These modalities are very much essential to doctors for proper diagnosis, treatment planning and follow up. Some modalities provide anatomical information such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), X-rays and some provides only functional information such as Positron Emission Tomography (PET). Therefore, single modality image does not give complete information. This paper presents the fusion of structural information in CT and functional information present in PET image. This fused image is very much essential in detecting the stages and location of abnormalities and in particular very much needed in oncology for improved diagnosis and treatment. We have implemented and compared image fusion techniques like pyramid, wavelet, and principal components fusion methods along with hybrid method of DWT and PCA. The performances of the algorithms are evaluated quantitatively and qualitatively. The system is implemented and tested by using MATLAB software. Based on the MSE, PSNR and ENTROPY analysis, PCA and DWT-PCA methods showed best results over all experiments.

Keywords: image fusion, pyramid, wavelets, principal component analysis

Procedia PDF Downloads 262

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 128

Authors: Sumit Banerjee, Chandan Kumar Chanda

Abstract:

The paper presents allocation of active power losses and energy losses to consumers connected to radial distribution networks in a deregulated environment for loads of composite types. A detailed comparison among four algorithms, namely quadratic loss allocation, proportional loss allocation, pro rata loss allocation and exact loss allocation methods are presented. Quadratic and proportional loss allocations are based on identifying the active and reactive components of current in each branch and the losses are allocated to each consumer, pro rata loss allocation method is based on the load demand of each consumer and exact loss allocation method is based on the actual contribution of active power loss by each consumer. The effectiveness of the proposed comparison among four algorithms for composite load is demonstrated through an example.

Keywords: composite type, deregulation, loss allocation, radial distribution networks

Procedia PDF Downloads 256

Authors: Farzaneh Rajabighamchi, Stan van Hoesel, Christof Defryn

Abstract:

The generalized traveling salesman problem (GTSP) is an extension of the traveling salesman problem (TSP) where the set of nodes is partitioned into clusters, and the salesman must visit exactly one node per cluster. In this research, we apply the definition of the GTSP to an order picker routing problem with multiple locations per product. As such, each product represents a cluster and its corresponding nodes are the locations at which the product can be retrieved. To pick a certain product item from the warehouse, the picker needs to visit one of these locations during its pick tour. As all products are scattered throughout the warehouse, the product clusters not separated geographically. We propose an exact LP model as well as heuristic and meta-heuristic solution algorithms for the order picking problem with multiple product locations.

Keywords: warehouse optimization, order picking problem, generalised travelling salesman problem, heuristic algorithm

Procedia PDF Downloads 84

Authors: Seyed Mehran Kazemi, Bahare Fatemi

Abstract:

Sudoku is a logic-based combinatorial puzzle game which is popular among people of different ages. Due to this popularity, computer softwares are being developed to generate and solve Sudoku puzzles with different levels of difficulty. Several methods and algorithms have been proposed and used in different softwares to efficiently solve Sudoku puzzles. Various search methods such as stochastic local search have been applied to this problem. Genetic Algorithm (GA) is one of the algorithms which have been applied to this problem in different forms and in several works in the literature. In these works, chromosomes with little or no information were considered and obtained results were not promising. In this paper, we propose a new way of applying GA to this problem which uses more-informed chromosomes than other works in the literature. We optimize the parameters of our GA using puzzles with different levels of difficulty. Then we use the optimized values of the parameters to solve various puzzles and compare our results to another GA-based method for solving Sudoku puzzles.

Keywords: genetic algorithm, optimization, solving Sudoku puzzles, stochastic local search

Procedia PDF Downloads 392

Authors: Nihal Murali, Kunal Gupta, Surekha Bhanot

Abstract:

Training of artificial neural networks (ANNs) using reinforcement learning (RL) techniques is being widely discussed in the robot learning literature. The high model complexity of ANNs along with the model-free nature of RL algorithms provides a desirable combination for many robotics applications. There is a huge need for algorithms that generalize using raw sensory inputs, such as vision, without any hand-engineered features or domain heuristics. In this paper, the standard control problem of line following robot was used as a test-bed, and an ANN controller for the robot was trained on images from a live video feed using Q-learning. A virtual agent was first trained in simulation environment and then deployed onto a robot’s hardware. The robot successfully learns to traverse a wide range of curves and displays excellent generalization ability. Qualitative analysis of the evolution of policies, performance and weights of the network provide insights into the nature and convergence of the learning algorithm.

Keywords: artificial neural networks, q-learning, reinforcement learning, robot learning

Procedia PDF Downloads 351

Authors: Farid Khosravikia, Patricia Clayton

Abstract:

The main objective of this study is to evaluate the advantages and disadvantages of various machine learning techniques in forecasting ground-motion intensity measures given source characteristics, source-to-site distance, and local site condition. Intensity measures such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Estimating these variables for future earthquake events is a key step in seismic hazard assessment and potentially subsequent risk assessment of different types of structures. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as a statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The algorithms are adjusted to quantify event-to-event and site-to-site variability of the ground motions by implementing them as random effects in the proposed models to reduce the aleatory uncertainty. All the algorithms are trained using a selected database of 4,528 ground-motions, including 376 seismic events with magnitude 3 to 5.8, recorded over the hypocentral distance range of 4 to 500 km in Oklahoma, Kansas, and Texas since 2005. The main reason of the considered database stems from the recent increase in the seismicity rate of these states attributed to petroleum production and wastewater disposal activities, which necessities further investigation in the ground motion models developed for these states. Accuracy of the models in predicting intensity measures, generalization capability of the models for future data, as well as usability of the models are discussed in the evaluation process. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available.

Keywords: artificial neural network, ground-motion models, machine learning, random forest, support vector machine

Procedia PDF Downloads 97

Authors: Abdul Basit Kiani, Maryam Kiani

Abstract:

Online developers now have the tools necessary to create online apps that are not only reliable but also highly interactive, thanks to the introduction of JavaScript frameworks and APIs. The objective is to give a broad overview of the recent advances in the area. The fusion of machine learning (ML) and artificial intelligence (AI) has expanded the possibilities for web development. Modern websites now include chatbots, clever recommendation systems, and customization algorithms built in. In the rapidly evolving landscape of modern websites, it has become increasingly apparent that user engagement and personalization are key factors for success. To meet these demands, websites now incorporate a range of innovative technologies. One such technology is chatbots, which provide users with instant assistance and support, enhancing their overall browsing experience. These intelligent bots are capable of understanding natural language and can answer frequently asked questions, offer product recommendations, and even help with troubleshooting. Moreover, clever recommendation systems have emerged as a powerful tool on modern websites. By analyzing user behavior, preferences, and historical data, these systems can intelligently suggest relevant products, articles, or services tailored to each user's unique interests. This not only saves users valuable time but also increases the chances of conversions and customer satisfaction. Additionally, customization algorithms have revolutionized the way websites interact with users. By leveraging user preferences, browsing history, and demographic information, these algorithms can dynamically adjust the website's layout, content, and functionalities to suit individual user needs. This level of personalization enhances user engagement, boosts conversion rates, and ultimately leads to a more satisfying online experience. In summary, the integration of chatbots, clever recommendation systems, and customization algorithms into modern websites is transforming the way users interact with online platforms. These advanced technologies not only streamline user experiences but also contribute to increased customer satisfaction, improved conversions, and overall website success.

Keywords: Javascript, machine learning, artificial intelligence, web development

Procedia PDF Downloads 40

Authors: V. Rashtchi, H. Bizhani, F. R. Tatari

Abstract:

This paper presents a new online loss minimization for an induction motor drive. Among the many loss minimization algorithms (LMAs) for an induction motor, a particle swarm optimization (PSO) has the advantages of fast response and high accuracy. However, the performance of the PSO and other optimization algorithms depend on the accuracy of the modeling of the motor drive and losses. In the development of the loss model, there is always a trade off between accuracy and complexity. This paper presents a new online optimization to determine an optimum flux level for the efficiency optimization of the vector-controlled induction motor drive. An induction motor (IM) model in d-q coordinates is referenced to the rotor magnetizing current. This transformation results in no leakage inductance on the rotor side, thus the decomposition into d-q components in the steady-state motor model can be utilized in deriving the motor loss model. The suggested algorithm is simple for implementation.

Keywords: induction machine, loss minimization, magnetizing current, particle swarm optimization

Procedia PDF Downloads 610

Authors: Alica Höpken, Hergen Pargmann

Abstract:

The resource-efficient planning of the complex production planning processes in the meat industry and the reduction of food waste is a permanent challenge. The complexity of the production planning process occurs in every part of the supply chain, from agriculture to the end consumer. It arises from long and uncertain planning phases. Uncertainties such as stochastic yields, fluctuations in demand, and resource variability are part of this process. In the meat industry, waste mainly relates to incorrect storage, technical causes in production, or overproduction. The high amount of food waste along the complex supply chain in the meat industry could not be reduced by simple solutions until now. Therefore, resource-efficient production planning by conventional methods is currently only partially feasible. The realization of intelligent, automated production planning is basically possible through the application of machine learning algorithms, such as those of reinforcement learning. By applying the adapted design thinking method, machine learning methods (especially reinforcement learning algorithms) are used for the complex production planning process in the meat industry. This method represents a concretization to the application area. A resource-efficient production planning process is made available by adapting the design thinking method. In addition, the complex processes can be planned efficiently by using this method, since this standardized approach offers new possibilities in order to challenge the complexity and the high time consumption. It represents a tool to support the efficient production planning in the meat industry. This paper shows an elegant adaption of the design thinking method to apply the reinforcement learning method for a resource-efficient production planning process in the meat industry. Following, the steps that are necessary to introduce machine learning algorithms into the production planning of the food industry are determined. This is achieved based on a case study which is part of the research project ”REIF - Resource Efficient, Economic and Intelligent Food Chain” supported by the German Federal Ministry for Economic Affairs and Climate Action of Germany and the German Aerospace Center. Through this structured approach, significantly better planning results are achieved, which would be too complex or very time consuming using conventional methods.

Keywords: change management, design thinking method, machine learning, meat industry, reinforcement learning, resource-efficient production planning

Procedia PDF Downloads 106

Authors: Sudeepti Dayal, Neelesh Gupta

Abstract:

Image abbreviation is utilized for reducing the size of a file without demeaning the quality of the image to an objectionable level. The depletion in file size permits more images to be deposited in a given number of spaces. It also minimizes the time necessary for images to be transferred. Storage of medical images is a most researched area in the current scenario. To store a medical image, there are two parameters on which the image is divided, regions of interest and non-regions of interest. The best way to store an image is to compress it in such a way that no important information is lost. Compression can be done in two ways, namely lossy, and lossless compression. Under that, several compression algorithms are applied. In the paper, two algorithms are used which are, discrete cosine transform, applied to non-region of interest (lossy), and discrete wavelet transform, applied to regions of interest (lossless). The paper introduces SPIHT (set partitioning hierarchical tree) algorithm which is applied onto the wavelet transform to obtain good compression ratio from which an image can be stored efficiently.

Keywords: Compression ratio, DWT, SPIHT, DCT

Procedia PDF Downloads 327

Authors: L. S. Chathurika

Abstract:

Early prediction of student performance is an important factor to be gained academic excellence. Whatever the study stream in secondary education, students lay the foundation for higher studies during the first year of their degree or diploma program in Sri Lanka. The information technology (IT) field has certain improvements in the education domain by selecting specialization areas to show the talents and skills of students. These specializations can be software engineering, network administration, database administration, multimedia design, etc. After completing the first-year, students attempt to select the best path by considering numerous factors. The purpose of this experiment is to predict the best study path using machine learning algorithms. Five classification algorithms: decision tree, support vector machine, artificial neural network, Naïve Bayes, and logistic regression are selected and tested. The support vector machine obtained the highest accuracy, 82.4%. Then affecting features are recognized to select the best study path.

Keywords: algorithm, classification, evaluation, features, testing, training

Procedia PDF Downloads 101

Authors: Celina Pestano-Gabino, Concepcion Gonzalez-Concepcion, M. Candelaria Gil-Fariña

Abstract:

Some estimation methods for VARMA models, and Multivariate Time Series Models in general, rely on the use of a Hankel matrix. It is known that if the data sample is populous enough and the dimension of the Hankel matrix is unnecessarily large, this may result in an unnecessary number of computations as well as in numerical problems. In this sense, the aim of this paper is two-fold. First, we provide some theoretical results for these matrices which translate into a lower dimension for the matrices normally used in the algorithms. This contribution thus serves to improve those methods from a numerical and, presumably, statistical point of view. Second, we have chosen an estimation algorithm to illustrate in practice our improvements. The results we obtained in a simulation of VARMA models show that an increase in the size of the Hankel matrix beyond the theoretical bound proposed as valid does not necessarily lead to improved practical results. Therefore, for future research, we propose conducting similar studies using any of the linear system estimation methods that depend on Hankel matrices.

Keywords: covariances Hankel matrices, Kronecker indices, system identification, VARMA models

Procedia PDF Downloads 217

Authors: G. M. Komba

Abstract:

The Internet Gateway (IGW) has extra ability than a simple Mesh Router (MR) and the responsibility to route mostly the all traffic from Mesh Clients (MCs) to the Internet backbone however, IGWs are more expensive. Choosing strategic locations for the Internet Gateways (IGWs) best location in Backbone Wireless Mesh (BWM) precarious to the Wireless Mesh Network (WMN) and the location of IGW can improve a quantity of performance related problem. In this paper, we propose a novel algorithm, namely New Gateway Location Algorithm (NGLA), which aims to achieve four objectives, decreasing the network cost effective, minimizing delay, optimizing the throughput capacity, Different from existing algorithms, the NGLA increasingly recognizes IGWs, allocates mesh routers (MRs) to identify IGWs and promises to find a feasible IGW location and install minimum as possible number of IGWs while regularly conserving the all Quality of Service (QoS) requests. Simulation results showing that the NGLA outperforms other different algorithms by comparing the number of IGWs with a large margin and it placed 40% less IGWs and 80% gain of throughput. Furthermore the NGLA is easy to implement and could be employed for BWM.

Keywords: Wireless Mesh Network, Gateway Location Algorithm, Quality of Service, BWM

Procedia PDF Downloads 339

Authors: S. H. Borghei, E. Teymourian, M. Mobin, G. M. Komaki, S. Sheikh

Abstract:

Imperialist competitive algorithm (ICA) is a recent meta-heuristic method that is inspired by the social evolutions for solving NP-Hard problems. The ICA is a population based algorithm which has achieved a great performance in comparison to other meta-heuristics. This study is about developing enhanced ICA approach to solve the cell formation problem (CFP) using sequence data. In addition to the conventional ICA, an enhanced version of ICA, namely EICA, applies local search techniques to add more intensification aptitude and embed the features of exploration and intensification more successfully. Suitable performance measures are used to compare the proposed algorithms with some other powerful solution approaches in the literature. In the same way, for checking the proficiency of algorithms, forty test problems are presented. Five benchmark problems have sequence data, and other ones are based on 0-1 matrices modified to sequence based problems. Computational results elucidate the efficiency of the EICA in solving CFP problems.

Keywords: cell formation problem, group technology, imperialist competitive algorithm, sequence data

Procedia PDF Downloads 433

Authors: Cemil Turan, Mohammad Shukri Salman

Abstract:

The convergence rate of the least-mean-square (LMS) algorithm deteriorates if the input signal to the filter is correlated. In a system identification problem, this convergence rate can be improved if the signal is white and/or if the system is sparse. We recently proposed a sparse transform domain LMS-type algorithm that uses a variable step-size for a sparse system identification. The proposed algorithm provided high performance even if the input signal is highly correlated. In this work, we investigate the performance of the proposed TD-LMS algorithm for a large number of filter tap which is also a critical issue for standard LMS algorithm. Additionally, the optimum value of the most important parameter is calculated for all experiments. Moreover, the convergence analysis of the proposed algorithm is provided. The performance of the proposed algorithm has been compared to different algorithms in a sparse system identification setting of different sparsity levels and different number of filter taps. Simulations have shown that the proposed algorithm has prominent performance compared to the other algorithms.

Keywords: adaptive filtering, sparse system identification, TD-LMS algorithm, VSSLMS algorithm

Procedia PDF Downloads 329

Authors: Azene Zenebe

Abstract:

Deep learning is a subset of machine learning which incorporates techniques for the construction of artificial neural networks and found to be useful for modeling complex problems with large dataset. Deep learning requires a very high power computational and longer time for training. By aggregating computing power, high performance computer (HPC) has emerged as an approach to resolving advanced problems and performing data-driven research activities. Cyber threat intelligence (CIT) is actionable information or insight an organization or individual uses to understand the threats that have, will, or are currently targeting the organization. Results of review of literature will be presented along with results of experimental study that compares the performance of tree-based and function-base machine learning including deep learning algorithms using secondary dataset collected from darknet.

Keywords: deep-learning, cyber security, cyber threat modeling, tree-based machine learning, function-based machine learning, data science

Procedia PDF Downloads 126

Authors: Misrak Assefa

Abstract:

The Amharic language is the second most widely spoken Semitic language in the world. There are several new overloaded on the web. Searching some useful documents from the web on a specific topic, which is written in the Amharic language, is a challenging task. Hence, document categorization is required for managing and filtering important information. In the classification of Amharic text news, there is still a gap in the domain of information that needs to be launch. This study attempts to design an automatic Amharic news classification using a supervised learning mechanism on four un-touch classes. To achieve this research, 4,182 news articles were used. Naive Bayes (NB) and Decision tree (j48) algorithms were used to classify the given Amharic dataset. In this paper, k-fold cross-validation is used to estimate the accuracy of the classifier. As a result, it shows those algorithms can be applicable in Amharic news categorization. The best average accuracy result is achieved by j48 decision tree and naïve Bayes is 95.2345 %, and 94.6245 % respectively using three categories. This research indicated that a typical decision tree algorithm is more applicable to Amharic news categorization.

Keywords: text categorization, supervised machine learning, naive Bayes, decision tree

Procedia PDF Downloads 166

Authors: Bashar Albaalbaki, Roger E. Khayat

Abstract:

This work is a comparative study on the effect of Delaunay triangulation algorithms on discretization error for conduction-convection conservation problems. A structured triangulation and many unstructured Delaunay triangulations using three popular algorithms for node placement strategies are used. The numerical method employed is the vertex-centered finite volume method. It is found that when the computational domain can be meshed using a structured triangulation, the discretization error is lower for structured triangulations compared to unstructured ones for only low Peclet number values, i.e. when conduction is dominant. However, as the Peclet number is increased and convection becomes more significant, the unstructured triangulations reduce the discretization error. Also, no statistical correlation between triangulation angle extremums and the discretization error is found using 200 samples of randomly generated Delaunay and non-Delaunay triangulations. Thus, the angle extremums cannot be an indicator of the discretization error on their own and need to be combined with other triangulation quality measures, which is the subject of further studies.

Keywords: conduction-convection problems, Delaunay triangulation, discretization error, finite volume method

Procedia PDF Downloads 72

Authors: Daniel Bulanda, Janusz A. Starzyk, Adrian Horzyk

Abstract:

The electrocardiogram (ECG) is one of the most commonly used medical tests, essential for correct diagnosis and treatment of the patient. While ECG devices generate a huge amount of data, only a small part of them carries valuable medical information. To deal with this problem, many compression algorithms and filters have been developed over the past years. However, the rapid development of new machine learning techniques poses new challenges. To address this class of problems, we created the FlexPoints algorithm that searches for characteristic points on the ECG signal and ignores all other points that do not carry relevant medical information. The conducted experiments proved that the presented algorithm can significantly reduce the number of data points which represents ECG signal without losing valuable medical information. These sparse but essential characteristic points (flex points) can be a perfect input for some modern machine learning models, which works much better using flex points as an input instead of raw data or data compressed by many popular algorithms.

Keywords: characteristic points, electrocardiogram, ECG, machine learning, signal compression

Procedia PDF Downloads 139

Authors: Benalia Nadia, Zerzouri Nora, Ben Si Ali Nadia

Abstract:

Abstract: Among the many different modern heuristic optimization methods, genetic algorithms (GA) and the particle swarm optimization (PSO) technique have been attracting a lot of interest. The GA has gained popularity in academia and business mostly because to its simplicity, ability to solve highly nonlinear mixed integer optimization problems that are typical of complex engineering systems, and intuitiveness. The mechanics of the PSO methodology, a relatively recent heuristic search tool, are modeled after the swarming or cooperative behavior of biological groups. It is suitable to compare the performance of the two techniques since they both aim to solve a particular objective function but make use of distinct computing methods. In this article, PSO and GA optimization approaches are used for the parameter tuning of the power system stabilizer and Proportional integral derivative regulator. Load angle and rotor speed variations in the single machine infinite bus bar system is used to measure the performance of the suggested solution.

Keywords: SMIB, genetic algorithm, PSO, transient stability, power system stabilizer, PID

Procedia PDF Downloads 51

Authors: Fahad Y. Al-dawish

Abstract:

The current trend by a large number of organizations and individuals to use cloud computing. Many consider it a significant shift in the field of computing. Cloud computing are distributed and parallel systems consisting of a collection of interconnected physical and virtual machines. With increasing request and profit of cloud computing infrastructure, diverse computing processes can be executed on cloud environment. Many organizations and individuals around the world depend on the cloud computing environments infrastructure to carry their applications, platform, and infrastructure. One of the major and essential issues in this environment related to allocating incoming tasks to suitable virtual machine (cloud task scheduling). Cloud task scheduling is classified as optimization problem, and there are several meta-heuristic algorithms have been anticipated to solve and optimize this problem. Good task scheduler should execute its scheduling technique on altering environment and the types of incoming task set. In this research project a cloud task scheduling methodology based on ant colony optimization ACO algorithm, we call it ACO-TS Ant Colony Optimization for Task Scheduling has been proposed and compared with different scheduling algorithms (Random, First Come First Serve FCFS, and Fastest Processor to the Largest Task First FPLTF). Ant Colony Optimization (ACO) is random optimization search method that will be used for assigning incoming tasks to available virtual machines VMs. The main role of proposed algorithm is to minimizing the makespan of certain tasks set and maximizing resource utilization by balance the load among virtual machines. The proposed scheduling algorithm was evaluated by using Cloudsim toolkit framework. Finally after analyzing and evaluating the performance of experimental results we find that the proposed algorithm ACO-TS perform better than Random, FCFS, and FPLTF algorithms in each of the makespaan and resource utilization.

Keywords: cloud Task scheduling, ant colony optimization (ACO), cloudsim, cloud computing

Procedia PDF Downloads 398

Authors: Zhao Gao, Eran Edirisinghe

Abstract:

The production of an accurate sketch of a suspect based on a verbal description obtained from a witness is an essential task for most criminal investigations. The criminal investigation system employs specifically trained professional artists to manually draw a facial image of the suspect according to the descriptions of an eyewitness for subsequent identification. Within the advancement of Deep Learning, Recurrent Neural Networks (RNN) have shown great promise in Natural Language Processing (NLP) tasks. Additionally, Generative Adversarial Networks (GAN) have also proven to be very effective in image generation. In this study, a trained GAN conditioned on textual features such as keywords automatically encoded from a verbal description of a human face using an RNN is used to generate photo-realistic facial images for criminal investigations. The intention of the proposed system is to map corresponding features into text generated from verbal descriptions. With this, it becomes possible to generate many reasonably accurate alternatives to which the witness can use to hopefully identify a suspect from. This reduces subjectivity in decision making both by the eyewitness and the artist while giving an opportunity for the witness to evaluate and reconsider decisions. Furthermore, the proposed approach benefits law enforcement agencies by reducing the time taken to physically draw each potential sketch, thus increasing response times and mitigating potentially malicious human intervention. With publically available 'CelebFaces Attributes Dataset' (CelebA) and additionally providing verbal description as training data, the proposed architecture is able to effectively produce facial structures from given text. Word Embeddings are learnt by applying the RNN architecture in order to perform semantic parsing, the output of which is fed into the GAN for synthesizing photo-realistic images. Rather than the grid search method, a metaheuristic search based on genetic algorithms is applied to evolve the network with the intent of achieving optimal hyperparameters in a fraction the time of a typical brute force approach. With the exception of the ‘CelebA’ training database, further novel test cases are supplied to the network for evaluation. Witness reports detailing criminals from Interpol or other law enforcement agencies are sampled on the network. Using the descriptions provided, samples are generated and compared with the ground truth images of a criminal in order to calculate the similarities. Two factors are used for performance evaluation: The Structural Similarity Index (SSIM) and the Peak Signal-to-Noise Ratio (PSNR). A high percentile output from this performance matrix should attribute to demonstrating the accuracy, in hope of proving that the proposed approach can be an effective tool for law enforcement agencies. The proposed approach to criminal facial image generation has potential to increase the ratio of criminal cases that can be ultimately resolved using eyewitness information gathering.

Keywords: RNN, GAN, NLP, facial composition, criminal investigation

Procedia PDF Downloads 139

Authors: K. Tejasri, K. Suvarna Vani, S. Prathyusha, S. Ramya

Abstract:

Proteins are chained sequences of amino acids which are brought together by the peptide bonds. Many varying formations of the chains are possible due to multiple combinations of amino acids and rotation in numerous positions along the chain. Protein structure prediction is one of the crucial goals worked towards by the members of bioinformatics and theoretical chemistry backgrounds. Among the four different structure levels in proteins, we emphasize mainly the secondary level structure. Generally, the secondary protein basically comprises alpha-helix and beta-sheets. Multi-class classification problem of data with disparity is truly a challenge to overcome and has to be addressed for the beta strands. Imbalanced data distribution constitutes a couple of the classes of data having very limited training samples collated with other classes. The secondary structure data is extracted from the protein primary sequence, and the beta-strands are predicted using suitable machine learning algorithms.

Keywords: proteins, secondary structure elements, beta-sheets, beta-strands, alpha-helices, machine learning algorithms

Procedia PDF Downloads 73

Authors: Thiago Ribeiro de Alencar, Jacyro Gramulia Junior, Patricia Teixeira Leite

Abstract:

The area of the electricity sector that deals with energy needs by the hydroelectric in a coordinated manner is called Operation Planning of Hydrothermal Power Systems (OPHPS). The purpose of this is to find a political operative to provide electrical power to the system in a given period, with reliability and minimal cost. Therefore, it is necessary to determine an optimal schedule of generation for each hydroelectric, each range, so that the system meets the demand reliably, avoiding rationing in years of severe drought, and that minimizes the expected cost of operation during the planning, defining an appropriate strategy for thermal complementation. Several optimization algorithms specifically applied to this problem have been developed and are used. Although providing solutions to various problems encountered, these algorithms have some weaknesses, difficulties in convergence, simplification of the original formulation of the problem, or owing to the complexity of the objective function. An alternative to these challenges is the development of techniques for simulation optimization and more sophisticated and reliable, it can assist the planning of the operation. Thus, this paper presents the development of a computational tool, namely Hydro-IA for solving optimization problem identified and to provide the User an easy handling. Adopted as intelligent optimization technique is Genetic Algorithm (GA) and programming language is Java. First made the modeling of the chromosomes, then implemented the function assessment of the problem and the operators involved, and finally the drafting of the graphical interfaces for access to the User. The results with the Genetic Algorithms were compared with the optimization technique nonlinear programming (NLP). Tests were conducted with seven hydroelectric plants interconnected hydraulically with historical stream flow from 1953 to 1955. The results of comparison between the GA and NLP techniques shows that the cost of operating the GA becomes increasingly smaller than the NLP when the number of hydroelectric plants interconnected increases. The program has managed to relate a coherent performance in problem resolution without the need for simplification of the calculations together with the ease of manipulating the parameters of simulation and visualization of output results.

Keywords: energy, optimization, hydrothermal power systems, artificial intelligence and genetic algorithms

Procedia PDF Downloads 395

Authors: Ionel Zagan, Vasile Gheorghita Gaitan

Abstract:

Minimizing the response time to asynchronous events in a real-time system is an important factor in increasing the speed of response and an interesting concept in designing equipment fast enough for the most demanding applications. The present article will present the results regarding the validation of the nMPRA (Multi Pipeline Register Architecture) architecture using the FPGA Virtex-7 circuit. The nMPRA concept is a hardware processor with the scheduler implemented at the processor level; this is done without affecting a possible bus communication, as is the case with the other CPU solutions. The implementation of static or dynamic scheduling operations in hardware and the improvement of handling interrupts and events by the real-time executive described in the present article represent a key solution for eliminating the overhead of the operating system functions. The nMPRA processor is capable of executing a preemptive scheduling, using various algorithms without a software scheduler. Therefore, we have also presented various scheduling methods and algorithms used in scheduling the real-time tasks.

Keywords: nMPRA architecture, pipeline processor, preemptive scheduling, real-time system

Procedia PDF Downloads 337

Authors: Daniel Fundi Murithi

Abstract:

Data from economic, social, clinical, and industrial studies are in some way incomplete or incorrect due to censoring. Such data may have adverse effects if used in the estimation problem. We propose the use of Maximum Likelihood Estimation (MLE) under a progressive type-II censoring scheme to remedy this problem. In particular, maximum likelihood estimates (MLEs) for the location (µ) and scale (λ) parameters of two Parameter Rayleigh distribution are realized under a progressive type-II censoring scheme using the Expectation-Maximization (EM) and the Newton-Raphson (NR) algorithms. These algorithms are used comparatively because they iteratively produce satisfactory results in the estimation problem. The progressively type-II censoring scheme is used because it allows the removal of test units before the termination of the experiment. Approximate asymptotic variances and confidence intervals for the location and scale parameters are derived/constructed. The efficiency of EM and the NR algorithms is compared given root mean squared error (RMSE), bias, and the coverage rate. The simulation study showed that in most sets of simulation cases, the estimates obtained using the Expectation-maximization algorithm had small biases, small variances, narrower/small confidence intervals width, and small root of mean squared error compared to those generated via the Newton-Raphson (NR) algorithm. Further, the analysis of a real-life data set (data from simple experimental trials) showed that the Expectation-Maximization (EM) algorithm performs better compared to Newton-Raphson (NR) algorithm in all simulation cases under the progressive type-II censoring scheme.

Keywords: expectation-maximization algorithm, maximum likelihood estimation, Newton-Raphson method, two-parameter Rayleigh distribution, progressive type-II censoring

Procedia PDF Downloads 133

Authors: Thanakrit Taweesoontorn, Sarucha Yanyong, Poom Konghuayrob

Abstract:

Robotic arms have gained popularity in various industries due to their accuracy and efficiency. This research proposes a method for bin-picking tasks using the Cobot, combining the YOLOv5 CNNs model for object detection and pose estimation with traditional feature detection (FAST), feature description (BRISK), and matching algorithms. By integrating these algorithms and utilizing a small-scale depth sensor camera for capturing depth and color images, the system achieves real-time object detection and accurate pose estimation, enabling the robotic arm to pick objects correctly in both position and orientation. Furthermore, the proposed method is implemented within the ROS framework to provide a seamless platform for robotic control and integration. This integration of robotics, cameras, and AI technology contributes to the development of industrial robotics, opening up new possibilities for automating challenging tasks and improving overall operational efficiency.

Keywords: robotic vision, image processing, applications of robotics, artificial intelligent

Procedia PDF Downloads 55

Authors: Hirofumi Miyajima, Kazuya Kishida, Noritaka Shigei, Hiromi Miyajima

Abstract:

Most of self-tuning fuzzy systems, which are automatically constructed from learning data, are based on the steepest descent method (SDM). However, this approach often requires a large convergence time and gets stuck into a shallow local minimum. One of its solutions is to use fuzzy rule modules with a small number of inputs such as DIRMs (Double-Input Rule Modules) and SIRMs (Single-Input Rule Modules). In this paper, we consider a (generalized) DIRMs model composed of double and single-input rule modules. Further, in order to reduce the redundant modules for the (generalized) DIRMs model, pruning and generative learning algorithms for the model are suggested. In order to show the effectiveness of them, numerical simulations for function approximation, Box-Jenkins and obstacle avoidance problems are performed.

Keywords: Box-Jenkins's problem, double-input rule module, fuzzy inference model, obstacle avoidance, single-input rule module

Procedia PDF Downloads 331

Authors: Heramb Somthankar, Anujoy Chakraborty

Abstract:

This paper presents a classification study on yield prediction in semiconductor production using machine learning approaches. A complicated semiconductor production process is generally monitored continuously by signals acquired from sensors and measurement sites. A monitoring system contains a variety of signals, all of which contain useful information, irrelevant information, and noise. In the case of each signal being considered a feature, "Feature Selection" is used to find the most relevant signals. The open-source UCI SECOM Dataset provides 1567 such samples, out of which 104 fail in quality assurance. Feature extraction and selection are performed on the dataset, and useful signals were considered for further study. Afterward, common machine learning algorithms were employed to predict whether the signal yields pass or fail. The most relevant algorithm is selected for prediction based on the accuracy and loss of the ML model.

Keywords: deep learning, feature extraction, feature selection, machine learning classification algorithms, semiconductor production monitoring, signal processing, time-series analysis

Procedia PDF Downloads 85