Search results for: Algorithms

Authors: Vytautas Galvanauskas, Rimvydas Simutis, Donatas Levisauskas, Vykantas Grincas, Renaldas Urniezius

Abstract:

The paper deals with model-based development and implementation of efficient control strategies for recombinant protein synthesis in fed-batch E.coli cultivation processes. Based on experimental data, a kinetic dynamic model for cultivation process was developed. This model was used to determine substrate feeding strategies during the cultivation. The proposed feeding strategy consists of two phases – biomass growth phase and recombinant protein production phase. In the first process phase, substrate-limited process is recommended when the specific growth rate of biomass is about 90-95% of its maximum value. This ensures reduction of glucose concentration in the medium, improves process repeatability, reduces the development of secondary metabolites and other unwanted by-products. The substrate limitation can be enhanced to satisfy restriction on maximum oxygen transfer rate in the bioreactor and to guarantee necessary dissolved carbon dioxide concentration in culture media. In the recombinant protein production phase, the level of substrate limitation and specific growth rate are selected within the range to enable optimal target protein synthesis rate. To account for complex process dynamics, to efficiently exploit the oxygen transfer capability of the bioreactor, and to maintain the required dissolved oxygen concentration, adaptive control algorithms for dissolved oxygen control have been proposed. The developed model-based control strategies are useful in scale-up of cultivation processes and accelerate implementation of innovative biotechnological processes for industrial applications.

Keywords: adaptive algorithms, model-based control, recombinant E. coli, scale-up of bioprocesses

Procedia PDF Downloads 239

Authors: John Pomerat, Aviv Segev

Abstract:

Recently, deep learning has had many theoretical breakthroughs. For deep learning to be successful in the industry, however, there need to be practical algorithms capable of handling many real-world hiccups preventing the immediate application of a learning algorithm. Although AI promises to revolutionize the healthcare industry, getting access to patient data in order to train learning algorithms has not been easy. One proposed solution to this is data- sharing. In this paper, we propose an alternative protocol, based on multi-party computation, to train deep learning models while maintaining both the privacy and security of training data. We examine three methods of training neural networks in this way: Transfer learning, average ensemble learning, and series network learning. We compare these methods to the equivalent model obtained through data-sharing across two different experiments. Additionally, we address the security concerns of this protocol. While the motivating example is healthcare, our findings regarding multi-party computation of neural network training are purely theoretical and have use-cases outside the domain of healthcare.

Keywords: neural network aggregation, multi-party computation, transfer learning, average ensemble learning

Procedia PDF Downloads 140

Authors: Iannick Gagnon, Alain April

Abstract:

The performance of a metaheuristic on a given problem class depends on the class itself and the choice of parameters. Parameter tuning is the most time-consuming phase of the optimization process after the main calculations and it often nullifies the speed advantage of metaheuristics over traditional optimization algorithms. Several off-the-shelf parameter tuning algorithms are available, but when the objective function is expensive to evaluate, these can be prohibitively expensive to use. This paper presents a surrogate-like method for finding adequate parameters using fitness landscape analysis on simple benchmark functions and real-world objective functions. The result is a simple compound similarity metric based on the empirical correlation coefficient and a measure of convexity. It is then used to find the best benchmark functions to serve as surrogates. The near-optimal parameter set is then found using fractional factorial design. The real-world problem of NACA airfoil lift coefficient maximization is used as a preliminary proof of concept. The overall aim of this research is to reduce the computational overhead of metaheuristics parameterization.

Keywords: metaheuristics, stochastic optimization, particle swarm optimization, exploratory landscape analysis

Procedia PDF Downloads 131

Authors: Nadia Masood Khan, Muhammad Salman Khan, Gul Muhammad Khan

Abstract:

Cardiologists perform cardiac auscultation to detect abnormalities in heart sounds. Since accurate auscultation is a crucial first step in screening patients with heart diseases, there is a need to develop computer-aided detection/diagnosis (CAD) systems to assist cardiologists in interpreting heart sounds and provide second opinions. In this paper different algorithms are implemented for automated heart sound classification using unsegmented phonocardiogram (PCG) signals. Support vector machine (SVM), artificial neural network (ANN) and cartesian genetic programming evolved artificial neural network (CGPANN) without the application of any segmentation algorithm has been explored in this study. The signals are first pre-processed to remove any unwanted frequencies. Both time and frequency domain features are then extracted for training the different models. The different algorithms are tested in multiple scenarios and their strengths and weaknesses are discussed. Results indicate that SVM outperforms the rest with an accuracy of 73.64%.

Keywords: pattern recognition, machine learning, computer aided diagnosis, heart sound classification, and feature extraction

Procedia PDF Downloads 237

Authors: Rashmi Yadawad, Disha Narayanan, Ravi Gautam

Abstract:

Through Wall Radar Imaging is gaining increasing importance now a days in the field of Defense and one of the most important criteria that forms the basis for the image quality obtained is the Cross-Range resolution of the image. In this research paper, the Bilateral Back projection algorithm has been implemented for Through Wall Radar Imaging. The sole purpose is to enhance the resolution in the cross range direction of the obtained Back projection image. Synthetic Data is generated for two targets which are placed at various locations in a room of dimensions 8 m by 6m. Two algorithms namely, simple back projection and Bilateral Back projection have been implemented, images are obtained and the obtained images are compared. Numerical simulations have been coded in MATLAB and experimental results of the two algorithms have been shown. Based on the comparison between the two images, it can be clearly seen that the ringing effect and chess board effect have been heavily reduced in the bilaterally back projected image and hence promising results are obtained giving a relatively sharper image with relatively well defined edges.

Keywords: through wall radar imaging, bilateral back projection, cross range resolution, synthetic data

Procedia PDF Downloads 322

Authors: Xiongxiong You, Zhanwen Niu

Abstract:

Choosing an appropriate surrogate model plays an important role in surrogates-assisted evolutionary algorithms (SAEAs) since there are many types and different kernel functions in the surrogate model. In this paper, an adaptive selection of the best suitable surrogate model method is proposed to solve different kinds of expensive optimization problems. Firstly, according to the prediction residual error sum of square (PRESS) and different model selection strategies, the excellent individual surrogate models are integrated into multiple ensemble models in each generation. Then, based on the minimum root of mean square error (RMSE), the best suitable surrogate model is selected dynamically. Secondly, two methods with dynamic number of models and selection strategies are designed, which are used to show the influence of the number of individual models and selection strategy. Finally, some compared studies are made to deal with several commonly used benchmark problems, as well as a rotor system optimization problem. The results demonstrate the accuracy and robustness of the proposed method.

Keywords: adaptive selection, expensive optimization, rotor system, surrogates assisted evolutionary algorithms

Procedia PDF Downloads 129

Authors: Kaoutar Ben Azzou, Hanaa Talei

Abstract:

Human resources is an important department in each organization as it manages the life cycle of employees from recruitment training to retirement or termination of contracts. The recruitment process starts with a job opening, followed by a selection of the best-fit candidates from all applicants. Matching the best profile for a job position requires a manual way of looking at many CVs, which requires hours of work that can sometimes lead to choosing not the best profile. The work presented in this paper aims at reducing the workload of HR personnel by automating the preliminary stages of the candidate screening process, thereby fostering a more streamlined recruitment workflow. This tool introduces an automated system designed to help with the recruitment process by scanning candidates' CVs, extracting pertinent features, and employing machine learning algorithms to decide the most fitting job profile for each candidate. Our work employs natural language processing (NLP) techniques to identify and extract key features from unstructured text extracted from a CV, such as education, work experience, and skills. Subsequently, the system utilizes these features to match candidates with job profiles, leveraging the power of classification algorithms.

Keywords: automated recruitment, candidate screening, machine learning, human resources management

Procedia PDF Downloads 35

Authors: Muhammad Bilal, Adil Ahmed

Abstract:

This study proposes an approach for electrical induction motor predictive maintenance utilizing machine learning algorithms. On the basis of a study of temperature data obtained from sensors put on the motor, the goal is to predict motor failures. The proposed models are trained to identify whether a motor is defective or not by utilizing machine learning algorithms like Support Vector Machines (SVM) and K-Nearest Neighbors (KNN). According to a thorough study of the literature, earlier research has used motor current signature analysis (MCSA) and vibration data to forecast motor failures. The temperature signal methodology, which has clear advantages over the conventional MCSA and vibration analysis methods in terms of cost-effectiveness, is the main subject of this research. The acquired results emphasize the applicability and effectiveness of the temperature-based predictive maintenance strategy by demonstrating the successful categorization of defective motors using the suggested machine learning models.

Keywords: predictive maintenance, electrical induction motors, machine learning, temperature signal methodology, motor failures

Procedia PDF Downloads 89

Authors: Yasmeen A. S. Essawy, Khaled Nassar

Abstract:

With the rapid increase of complexity in the building industry, professionals in the A/E/C industry were forced to adopt Building Information Modeling (BIM) in order to enhance the communication between the different project stakeholders throughout the project life cycle and create a semantic object-oriented building model that can support geometric-topological analysis of building elements during design and construction. This paper presents a model that extracts topological relationships and geometrical properties of building elements from an existing fully designed BIM, and maps this information into a directed acyclic Elemental Graph Data Model (EGDM). The model incorporates BIM-based search algorithms for automatic deduction of geometrical data and topological relationships for each building element type. Using graph search algorithms, such as Depth First Search (DFS) and topological sortings, all possible construction sequences can be generated and compared against production and construction rules to generate an optimized construction sequence and its associated schedule. The model is implemented in a C# platform.

Keywords: building information modeling (BIM), elemental graph data model (EGDM), geometric and topological data models, graph theory

Procedia PDF Downloads 355

Authors: Mikiyas Adugna

Abstract:

Effort estimation is among the activities carried out in software development processes. An accurate model of estimation leads to project success. The method of agile effort estimation is a complex task because of the dynamic nature of software development. Researchers are still conducting studies on agile effort estimation to enhance prediction accuracy. Due to these reasons, we investigated and proposed a model on LASSO and Elastic Net regression to enhance estimation accuracy. The proposed model has major components: preprocessing, train-test split, training with default parameters, and cross-validation. During the preprocessing phase, the entire dataset is normalized. After normalization, a train-test split is performed on the dataset, setting training at 80% and testing set to 20%. We chose two different phases for training the two algorithms (Elastic Net and LASSO) regression following the train-test-split. In the first phase, the two algorithms are trained using their default parameters and evaluated on the testing data. In the second phase, the grid search technique (the grid is used to search for tuning and select optimum parameters) and 5-fold cross-validation to get the final trained model. Finally, the final trained model is evaluated using the testing set. The experimental work is applied to the agile story point dataset of 21 software projects collected from six firms. The results show that both Elastic Net and LASSO regression outperformed the compared ones. Compared to the proposed algorithms, LASSO regression achieved better predictive performance and has acquired PRED (8%) and PRED (25%) results of 100.0, MMRE of 0.0491, MMER of 0.0551, MdMRE of 0.0593, MdMER of 0.063, and MSE of 0.0007. The result implies LASSO regression algorithm trained model is the most acceptable, and higher estimation performance exists in the literature.

Keywords: agile software development, effort estimation, elastic net regression, LASSO

Procedia PDF Downloads 41

Authors: Oluwatosin Olofintoye, Josiah Adeyemo, Gbemileke Shomade

Abstract:

The relationship between rainfall and runoff is an important issue in surface water hydrology therefore the understanding and development of accurate rainfall-runoff models and their applications in water resources planning, management and operation are of paramount importance in hydrological studies. This paper reviews some of the previous works on the rainfall-runoff process modeling. The hydrologic applications of conceptual models and artificial neural networks (ANNs) for the precipitation-runoff process modeling were studied. Gradient training methods such as error back-propagation (BP) and evolutionary algorithms (EAs) are discussed in relation to the training of artificial neural networks and it is shown that application of EAs to artificial neural networks training could be an alternative to other training methods. Therefore, further research interest to exploit the abundant expert knowledge in the area of artificial intelligence for the solution of hydrologic and water resources planning and management problems is needed.

Keywords: artificial intelligence, artificial neural networks, evolutionary algorithms, gradient training method, rainfall-runoff model

Procedia PDF Downloads 425

Authors: B. Karabuluter, O. Karaduman

Abstract:

Traffic management and traffic planning are important issues, especially in big cities. Due to the increase of personal vehicles and the physical constraints of urban roads, the problem of transportation especially in crowded cities over time is revealed. This situation reduces the living standards, and it can put human life at risk because the vehicles such as ambulance, fire department are prevented from reaching their targets. Even if the city planners take these problems into account, emergency planning and traffic management are needed to avoid cases such as traffic congestion, intersections, traffic jams caused by traffic accidents or roadworks. In this study, in smart traffic management issues, proposed solutions using intelligent vehicles acting in cooperation with urban roads are examined. Traffic management is becoming more difficult due to factors such as fatigue, carelessness, sleeplessness, social behavior patterns, and lack of education. However, autonomous vehicles, which remove the problems caused by human weaknesses by providing driving control, are increasing the success of practicing the algorithms developed in city traffic management. Such intelligent vehicles have become an important solution in urban life by using 'swarm intelligence' algorithms and cooperative driving methods to provide traffic flow, prevent traffic accidents, and increase living standards. In this study, studies conducted in this area have been dealt with in terms of traffic jam, intersections, regulation of traffic flow, signaling, prevention of traffic accidents, cooperation and communication techniques of vehicles, fleet management, transportation of emergency vehicles. From these concepts, some taxonomies were made out of the way. This work helps to develop new solutions and algorithms for cities where intelligent vehicles that can perform cooperative driving can take place, and at the same time emphasize the trend in this area.

Keywords: intelligent traffic management, cooperative driving, smart driving, urban road, swarm intelligence, connected vehicles

Procedia PDF Downloads 313

Authors: Mokrani Mohamed Amine

Abstract:

In this paper, we have implemented successive interference cancellation algorithms in the 5G downlink. We have calculated the maximum throughput in Frequency Division Duplex (FDD) mode in the downlink, where we have obtained a value equal to 836932 b/ms. The transmitter is of type Multiple Input Multiple Output (MIMO) with eight transmitting and receiving antennas. Each antenna among eight transmits simultaneously a data rate of 104616 b/ms that contains the binary messages of the three users; in this case, the Cyclic Redundancy Check CRC is negligible, and the MIMO category is the spatial diversity. The technology used for this is called Non-Orthogonal Multiple Access (NOMA) with a Quadrature Phase Shift Keying (QPSK) modulation. The transmission is done in a Rayleigh fading channel with the presence of obstacles. The MIMO Successive Interference Cancellation (SIC) receiver with two transmitting and receiving antennas recovers its binary message without errors for certain values of transmission power such as 50 dBm, with 0.054485% errors when the transmitted power is 20dBm and with 0.00286763% errors for a transmitted power of 32 dBm(in the case of user 1) as well as with 0.0114705% errors when the transmitted power is 20 dBm also with 0.00286763% errors for a power of 24 dBm(in the case of user2) by applying the steps involved in SIC.

Keywords: 5G, NOMA, QPSK, TBS, LDPC, SIC, capacity

Procedia PDF Downloads 87

Authors: Carolina Silva, João Nuno Oliveira, Rui Sousa, João Paulo Silva

Abstract:

This study uses genetic algorithms to solve a job scheduling problem in a weaving factory. The underline problem regards an NP-Hard problem concerning unrelated parallel machines, with sequence-dependent setup times. This research uses real data regarding a weaving industry located in the North of Portugal, with a capacity of 96 looms and a production, on average, of 440000 meters of fabric per month. Besides, this study includes a high level of complexity once most of the real production constraints are applied, and several real data instances are tested. Topics such as data analyses and algorithm performance are addressed and tested, to offer a solution that can generate reliable and due date results. All the approaches will be tested in the operational environment, and the KPIs monitored, to understand the solution's impact on the production, with a particular focus on the total number of weeks of late deliveries to clients. Thus, the main goal of this research is to develop a solution that allows for the production of automatically optimized production plans, aiming to the tardiness minimizing.

Keywords: genetic algorithms, textile industry, job scheduling, optimization

Procedia PDF Downloads 138

Authors: Jorge A. Ruiz-Vanoye, Ocotlán Díaz-Parra, Alejandro Fuentes-Penna, José J. Hernández-Flores, Edith Olaco Garcia

Abstract:

The Project Scheduling Problem (PSP) is a generic name given to a whole class of problems in which the best form, time, resources and costs for project scheduling are necessary. The PSP is an application area related to the project management. This paper aims at being a guide to understand PSP by presenting a survey of the general parameters of PSP: the Resources (those elements that realize the activities of a project), and the Activities (set of operations or own tasks of a person or organization); the mathematical models of the main variants of PSP and the algorithms used to solve the variants of the PSP. The project scheduling is an important task in project management. This paper contains mathematical models, resources, activities, and algorithms of project scheduling problems. The project scheduling problem has attracted researchers of the automotive industry, steel manufacturer, medical research, pharmaceutical research, telecommunication, industry, aviation industry, development of the software, manufacturing management, innovation and technology management, construction industry, government project management, financial services, machine scheduling, transportation management, and others. The project managers need to finish a project with the minimum cost and the maximum quality.

Keywords: PSP, Combinatorial Optimization Problems, Project Management; Manufacturing Management, Technology Management.

Procedia PDF Downloads 395

Authors: Saeed M. Alshahrani

Abstract:

This research integrated blockchain technology to reform medical records management in healthcare informatics. It was aimed at resolving the limitations of centralized systems by establishing a secure, decentralized, and user-centric platform. The system was architected with a sophisticated three-tiered structure, integrating advanced cryptographic methodologies, consensus algorithms, and the Fast Healthcare Interoperability Resources (HL7 FHIR) standard to ensure data security, transaction validity, and semantic interoperability. The research has profound implications for healthcare delivery, patient care, legal compliance, operational efficiency, and academic advancements in blockchain technology and healthcare IT sectors. The methodology adapted in this research comprises of Preliminary Feasibility Study, Literature Review, Design and Development, Cryptographic Algorithm Integration, Modeling the data and testing the system. The research employed a permissioned blockchain with a Practical Byzantine Fault Tolerance (PBFT) consensus algorithm and Ethereum-based smart contracts. It integrated advanced cryptographic algorithms, role-based access control, multi-factor authentication, and RESTful APIs to ensure security, regulate access, authenticate user identities, and facilitate seamless data exchange between the blockchain and legacy healthcare systems. The research contributed to the development of a secure, interoperable, and decentralized system for managing medical records, addressing the limitations of the centralized systems that were in place. Future work will delve into optimizing the system further, exploring additional blockchain use cases in healthcare, and expanding the adoption of the system globally, contributing to the evolution of global healthcare practices and policies.

Keywords: healthcare informatics, blockchain, medical records management, decentralized architecture, data security, cryptographic algorithms

Procedia PDF Downloads 44

Authors: Fabio Fabris, Alex A. Freitas

Abstract:

Hierarchical classification is a special type of classification task where the class labels are organised into a hierarchy, with more generic class labels being ancestors of more specific ones. Meta-learning for classification-algorithm recommendation consists of recommending to the user a classification algorithm, from a pool of candidate algorithms, for a dataset, based on the past performance of the candidate algorithms in other datasets. Meta-learning is normally used in conventional, non-hierarchical classification. By contrast, this paper proposes a meta-learning approach for more challenging task of hierarchical classification, and evaluates it in a large number of bioinformatics datasets. Hierarchical classification is especially relevant for bioinformatics problems, as protein and gene functions tend to be organised into a hierarchy of class labels. This work proposes meta-learning approach for recommending the best hierarchical classification algorithm to a hierarchical classification dataset. This work’s contributions are: 1) proposing an algorithm for splitting hierarchical datasets into new datasets to increase the number of meta-instances, 2) proposing meta-features for hierarchical classification, and 3) interpreting decision-tree meta-models for hierarchical classification algorithm recommendation.

Keywords: algorithm recommendation, meta-learning, bioinformatics, hierarchical classification

Procedia PDF Downloads 291

Authors: Corentin Jankowiak, Aggelos Zacharopoulos, Caterina Brandoni

Abstract:

In a context of increasing stress put on the electricity network by the decarbonization of many sectors, energy storage is likely to be the key mitigating element, by acting as a buffer between production and demand. In particular, the highest potential for storage is when connected closer to the loads. Yet, low voltage storage struggles to penetrate the market at a large scale due to the novelty and complexity of the solution, and the competitive advantage of fossil fuel-based technologies regarding regulations. Strong and reliable numerical simulations are required to show the benefits of storage located near loads and promote its development. The present study was restrained from excluding aggregated control of storage: it is assumed that the storage units operate independently to one another without exchanging information – as is currently mostly the case. A computationally light battery model is presented in detail and validated by direct comparison with a domestic battery operating in real conditions. This model is then used to develop Peak-Shaving (PS) control strategies as it is the decentralized service from which beneficial impacts are most likely to emerge. The aggregation of flatter, peak- shaved consumption profiles is likely to lead to flatter and arbitraged profile at higher voltage layers. Furthermore, voltage fluctuations can be expected to decrease if spikes of individual consumption are reduced. The crucial part to achieve PS lies in the charging pattern: peaks depend on the switching on and off of appliances in the dwelling by the occupants and are therefore impossible to predict accurately. A performant PS strategy must, therefore, include a smart charge recovery algorithm that can ensure enough energy is present in the battery in case it is needed without generating new peaks by charging the unit. Three categories of PS algorithms are introduced in detail. First, using a constant threshold or power rate for charge recovery, followed by algorithms using the State Of Charge (SOC) as a decision variable. Finally, using a load forecast – of which the impact of the accuracy is discussed – to generate PS. A performance metrics was defined in order to quantitatively evaluate their operating regarding peak reduction, total energy consumption, and self-consumption of domestic photovoltaic generation. The algorithms were tested on load profiles with a 1-minute granularity over a 1-year period, and their performance was assessed regarding these metrics. The results show that constant charging threshold or power are far from optimal: a certain value is not likely to fit the variability of a residential profile. As could be expected, forecast-based algorithms show the highest performance. However, these depend on the accuracy of the forecast. On the other hand, SOC based algorithms also present satisfying performance, making them a strong alternative when the reliable forecast is not available.

Keywords: decentralised control, domestic integrated batteries, electricity network performance, peak-shaving algorithm

Procedia PDF Downloads 102

Authors: Pooya Niksiar, Ali Ashrafizadeh, Mehrzad Shams, Amir Hossein Madani

Abstract:

There are many computationally demanding applications in science and engineering which need efficient algorithms implemented on high performance computers. Recently, Graphics Processing Units (GPUs) have drawn much attention as compared to the traditional CPU-based hardware and have opened up new improvement venues in scientific computing. One particular application area is Computational Fluid Dynamics (CFD), in which mature CPU-based codes need to be converted to GPU-based algorithms to take advantage of this new technology. In this paper, numerical solutions of two classes of discrete fluid flow models via both CPU and GPU are discussed and compared. Test problems include an Eulerian model of a two-dimensional incompressible laminar flow case and a Lagrangian model of a two phase flow field. The CUDA programming standard is used to employ an NVIDIA GPU with 480 cores and a C++ serial code is run on a single core Intel quad-core CPU. Up to two orders of magnitude speed up is observed on GPU for a certain range of grid resolution or particle numbers. As expected, Lagrangian formulation is better suited for parallel computations on GPU although Eulerian formulation represents significant speed up too.

Keywords: CFD, Eulerian formulation, graphics processing units, Lagrangian formulation

Procedia PDF Downloads 388

Authors: Shady Adib, Vladimir Vinogradov, Peter Gosling

Abstract:

Digital Twin (DT) technology is a new technology that appeared in the early 21st century. The DT is defined as the digital representation of living and non-living physical assets. By connecting the physical and virtual assets, data are transmitted smoothly, allowing the virtual asset to fully represent the physical asset. Although there are lots of studies conducted on the DT concept, there is still limited information about the ability of the DT models for monitoring and detecting unexpected changes in structural behaviour in real time. This is due to the large computational efforts required for the analysis and an excessively large amount of data transferred from sensors. This paper aims to develop the DT concept to be able to detect the abnormal changes in structural behaviour in real time using advanced modelling techniques, deep learning algorithms, and data acquisition systems, taking into consideration model uncertainties. finite element (FE) models were first developed offline to be used with a reduced basis (RB) model order reduction technique for the construction of low-dimensional space to speed the analysis during the online stage. The RB model was validated against experimental test results for the establishment of a DT model of a two-dimensional truss. The established DT model and deep learning algorithms were used to identify the location of damage once it has appeared during the online stage. Finally, the RB model was used again to identify the damage severity. It was found that using the RB model, constructed offline, speeds the FE analysis during the online stage. The constructed RB model showed higher accuracy for predicting the damage severity, while deep learning algorithms were found to be useful for estimating the location of damage with small severity.

Keywords: data acquisition system, deep learning, digital twin, model uncertainties, reduced basis, reduced order model

Procedia PDF Downloads 79

Authors: Yuanxiang Miao

Abstract:

Couchsurfing is a free homestay and social networking service accessible via the website and mobile app. Couchsurfers can directly request free accommodations from others and receive offers from each other. However, it is typically difficult for people to make a decision that accepts or declines a request when they receive it from Couchsurfers because they do not know each other at all. People are expected to meet up with some Couchsurfers who are kind, generous, and interesting while it is unavoidable to meet up with someone unfriendly. This paper utilized classification algorithms of Machine Learning to help people to find out the Good Couchsurfers and Not Good Couchsurfers on the Couchsurfing website. By knowing the prior experience, like Couchsurfer’s profiles, the latest references, and other factors, it became possible to recognize what kind of the Couchsurfers, and furthermore, it helps people to make a decision that whether to host the Couchsurfers or not. The value of this research lies in a case study in Kyoto, Japan in where the author has hosted 54 Couchsurfers, and the author collected relevant data from the 54 Couchsurfers, finally build a model based on classification algorithms for people to predict Couchsurfers. Lastly, the author offered some feasible suggestions for future research.

Keywords: Couchsurfing, Couchsurfers prediction, classification algorithm, hospitality tourism platform, hospitality sciences, machine learning

Procedia PDF Downloads 105

Authors: Hanieh Tarbiat Khosrowshahi, Mojtaba Shakeri

Abstract:

Wireless Sensor Networks (WSNs) consist of a set of sensor nodes with limited capability. WSNs may suffer from multiple node failures when they are exposed to harsh environments such as military zones or disaster locations and lose connectivity by getting partitioned into disjoint segments. Relay nodes (RNs) are alternatively introduced to restore connectivity. They cost more than sensors as they benefit from mobility, more power and more transmission range, enforcing a minimum number of them to be used. This paper addresses the problem of RN placement in a multiple disjoint network by developing a genetic algorithm (GA). The problem is reintroduced as the Steiner tree problem (which is known to be an NP-hard problem) by the aim of finding the minimum number of Steiner points where RNs are to be placed for restoring connectivity. An upper bound to the number of RNs is first computed to set up the length of initial chromosomes. The GA algorithm then iteratively reduces the number of RNs and determines their location at the same time. Experimental results indicate that the proposed GA is capable of establishing network connectivity using a reasonable number of RNs compared to the best existing work.

Keywords: connectivity restoration, genetic algorithms, multiple-node failure, relay nodes, wireless sensor networks

Procedia PDF Downloads 222

Authors: Mustafa Elfituri, Jonathan Cook

Abstract:

Recently, graph-based computations have become more important in large-scale scientific computing as they can provide a methodology to model many types of relations between independent objects. They are being actively used in fields as varied as biology, social networks, cybersecurity, and computer networks. At the same time, graph problems have some properties such as irregularity and poor locality that make their performance different than regular applications performance. Therefore, parallelizing graph algorithms is a hard and challenging task. Initial evidence is that standard computer architectures do not perform very well on graph algorithms. Little is known exactly what causes this. The Graph500 benchmark is a representative application for parallel graph-based computations, which have highly irregular data access and are driven more by traversing connected data than by computation. In this paper, we present results from analyzing the performance of various example implementations of Graph500, including a shared memory (OpenMP) version, a distributed (MPI) version, and a hybrid version. We measured and analyzed all the factors that affect its performance in order to identify possible changes that would improve its performance. Results are discussed in relation to what factors contribute to performance degradation.

Keywords: graph computation, graph500 benchmark, parallel architectures, parallel programming, workload characterization.

Procedia PDF Downloads 125

Authors: Marina Nechayeva, Malgorzata Marciniak, Vladimir Przhebelskiy, A. Dragutan, S. Lamichhane, S. Oikawa

Abstract:

This presentation is a progress report on a faculty-student research collaboration at CUNY LaGuardia Community College (LaGCC) aimed at designing a small horizontal axis wind turbine optimized for the wind patterns on the roof of our campus. Our project combines statistical and engineering research. Our wind modeling protocol is based upon a recent wind study by a faculty-student research group at MIT, and some of our blade design methods are adopted from a senior engineering project at CUNY City College. Our use of genetic algorithms has been inspired by the work on small wind turbines’ design by David Wood. We combine these diverse approaches in our interdisciplinary project in a way that has not been done before and improve upon certain techniques used by our predecessors. We employ several estimation methods to determine the best fitting parametric probability distribution model for the local wind speed data obtained through correlating short-term on-site measurements with a long-term time series at the nearby airport. The model serves as a foundation for engineering research that focuses on adapting and implementing genetic algorithms (GAs) to engineering optimization of the wind turbine design using Blade Element Momentum Theory. GAs are used to create new airfoils with desirable aerodynamic specifications. Small scale models of best performing designs are 3D printed and tested in the wind tunnel to verify the accuracy of relevant calculations. Genetic algorithms are applied to selected airfoils to determine the blade design (radial cord and pitch distribution) that would optimize the coefficient of power profile of the turbine. Our approach improves upon the traditional blade design methods in that it lets us dispense with assumptions necessary to simplify the system of Blade Element Momentum Theory equations, thus resulting in more accurate aerodynamic performance calculations. Furthermore, it enables us to design blades optimized for a whole range of wind speeds rather than a single value. Lastly, we improve upon known GA-based methods in that our algorithms are constructed to work with XFoil generated airfoils data which enables us to optimize blades using our own high glide ratio airfoil designs, without having to rely upon available empirical data from existing airfoils, such as NACA series. Beyond its immediate goal, this ongoing project serves as a training and selection platform for CUNY Research Scholars Program (CRSP) through its annual Aerodynamics and Wind Energy Research Seminar (AWERS), an undergraduate summer research boot camp, designed to introduce prospective researchers to the relevant theoretical background and methodology, get them up to speed with the current state of our research, and test their abilities and commitment to the program. Furthermore, several aspects of the research (e.g., writing code for 3D printing of airfoils) are adapted in the form of classroom research activities to enhance Calculus sequence instruction at LaGCC.

Keywords: engineering design optimization, genetic algorithms, horizontal axis wind turbine, wind modeling

Procedia PDF Downloads 209

Authors: Varun Singh, Mainak Bandyopadhyay, Maharana Pratap Singh

Abstract:

The shortest path (SP) problem concerns with finding the shortest path from a specific origin to a specified destination in a given network while minimizing the total cost associated with the path. This problem has widespread applications. Important applications of the SP problem include vehicle routing in transportation systems particularly in the field of in-vehicle Route Guidance System (RGS) and traffic assignment problem (in transportation planning). Well known applications of evolutionary methods like Genetic Algorithms (GA), Ant Colony Optimization, Particle Swarm Optimization (PSO) have come up to solve complex optimization problems to overcome the shortcomings of existing shortest path analysis methods. It has been reported by various researchers that PSO performs better than other evolutionary optimization algorithms in terms of success rate and solution quality. Further Geographic Information Systems (GIS) have emerged as key information systems for geospatial data analysis and visualization. This research paper is focused towards the application of PSO for solving the shortest path problem between multiple points of interest (POI) based on spatial data of Allahabad City and traffic speed data collected using GPS. Geovisualization of results of analysis is carried out in GIS.

Keywords: particle swarm optimization, GIS, traffic data, outliers

Procedia PDF Downloads 458

Authors: Tomyslav Sledevič, Artūras Serackis, Gintautas Tamulevičius, Dalius Navakauskas

Abstract:

This paper presents a comparative evaluation of features extraction algorithm for a real-time isolated word recognition system based on FPGA. The Mel-frequency cepstral, linear frequency cepstral, linear predictive and their cepstral coefficients were implemented in hardware/software design. The proposed system was investigated in the speaker-dependent mode for 100 different Lithuanian words. The robustness of features extraction algorithms was tested recognizing the speech records at different signals to noise rates. The experiments on clean records show highest accuracy for Mel-frequency cepstral and linear frequency cepstral coefficients. For records with 15 dB signal to noise rate the linear predictive cepstral coefficients give best result. The hard and soft part of the system is clocked on 50 MHz and 100 MHz accordingly. For the classification purpose, the pipelined dynamic time warping core was implemented. The proposed word recognition system satisfies the real-time requirements and is suitable for applications in embedded systems.

Keywords: isolated word recognition, features extraction, MFCC, LFCC, LPCC, LPC, FPGA, DTW

Procedia PDF Downloads 474

Authors: Francesca Schiavello

Abstract:

This work aims to create a hybrid algorithm, combining Quantum Approximate Optimization Algorithm (QAOA) with an Evolutionary Algorithm (EA) in the place of traditional gradient based optimization processes. QAOA’s were first introduced in 2014, where, at the time, their algorithm performed better than the traditional best known classical algorithm for Max-cut graphs. Whilst classical algorithms have improved since then and have returned to being faster and more efficient, this was a huge milestone for quantum computing, and their work is often used as a benchmarking tool and a foundational tool to explore variants of QAOA’s. This, alongside with other famous algorithms like Grover’s or Shor’s, highlights to the world the potential that quantum computing holds. It also presents the reality of a real quantum advantage where, if the hardware continues to improve, this could constitute a revolutionary era. Given that the hardware is not there yet, many scientists are working on the software side of things in the hopes of future progress. Some of the major limitations holding back quantum computing are the quality of qubits and the noisy interference they generate in creating solutions, the barren plateaus that effectively hinder the optimization search in the latent space, and the availability of number of qubits limiting the scale of the problem that can be solved. These three issues are intertwined and are part of the motivation for using EAs in this work. Firstly, EAs are not based on gradient or linear optimization methods for the search in the latent space, and because of their freedom from gradients, they should suffer less from barren plateaus. Secondly, given that this algorithm performs a search in the solution space through a population of solutions, it can also be parallelized to speed up the search and optimization problem. The evaluation of the cost function, like in many other algorithms, is notoriously slow, and the ability to parallelize it can drastically improve the competitiveness of QAOA’s with respect to purely classical algorithms. Thirdly, because of the nature and structure of EA’s, solutions can be carried forward in time, making them more robust to noise and uncertainty. Preliminary results show that the EA algorithm attached to QAOA can perform on par with the traditional QAOA with a Cobyla optimizer, which is a linear based method, and in some instances, it can even create a better Max-Cut. Whilst the final objective of the work is to create an algorithm that can consistently beat the original QAOA, or its variants, due to either speedups or quality of the solution, this initial result is promising and show the potential of EAs in this field. Further tests need to be performed on an array of different graphs with the parallelization aspect of the work commencing in October 2023 and tests on real hardware scheduled for early 2024.

Keywords: evolutionary algorithm, max cut, parallel simulation, quantum optimization

Procedia PDF Downloads 42

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

Abstract:

The area of the electricity sector that deals with energy needs by the hydropower and thermoelectric in a coordinated way is called Planning Operating Hydrothermal Power Systems. The aim of this area is to find a political operative to provide electrical power to the system in a specified period with minimization of operating cost. This article proposes a computational tool for solving the planning problem. In addition, this article will be introducing a methodology to find new transfer points between reservoirs increasing energy production in hydroelectric power plants cascade systems. The computational tool proposed in this article applies: i) genetic algorithms to optimize the water transfer and operation of hydroelectric plants systems; and ii) Ant Colony algorithm to find the trajectory with the least energy pumping for the construction of pipes transfer between reservoirs considering the topography of the region. The computational tool has a database consisting of 35 hydropower plants and 41 reservoirs, which are part of the southeastern Brazilian system, which has been implemented in an individualized way.

Keywords: ant colony system, genetic algorithms, hydroelectric, hydrothermal systems, optimization, water transfer between rivers

Procedia PDF Downloads 304

Authors: M. Kastek, M. Kowalski, M. Szustakowski, H. Polakowski, T. Sosnowski

Abstract:

Development of highly efficient security systems is one of the most urgent topics for science and engineering. There are many kinds of threats and many methods of prevention. It is very important to detect a threat as early as possible in order to neutralize it. One of the very challenging problems is detection of dangerous objects hidden under human’s clothing. This problem is particularly important for safety of airport passengers. In order to develop methods and algorithms to detect hidden objects it is necessary to determine the thermal signatures of such objects of interest. The laboratory measurements were conducted to determine the thermal signatures of dangerous tools hidden under various clothes in different ambient conditions. Cameras used for measurements were working in spectral range 0.6-12.5 μm An infrared imaging Fourier transform spectroradiometer was also used, working in spectral range 7.7-11.7 μm. Analysis of registered thermograms and hyperspectral datacubes has yielded the thermal signatures for two types of guns, two types of knives and home-made explosive bombs. The determined thermal signatures will be used in the development of method and algorithms of image analysis implemented in proposed monitoring systems.

Keywords: hyperspectral detection, nultispectral detection, image processing, monitoring systems

Procedia PDF Downloads 335

Authors: Sadia Basar, Naila Habib, Awais Adnan

Abstract:

The paper presents the Technique for Image Segmentation Using K-Means Clustering Classification. The presented algorithms were specific, however, missed the neighboring information and required high-speed computerized machines to run the segmentation algorithms. Clustering is the process of partitioning a group of data points into a small number of clusters. The proposed method is content-aware and feature extraction method which is able to run on low-end computerized machines, simple algorithm, required low-quality streaming, efficient and used for security purpose. It has the capability to highlight the boundary and the object. At first, the user enters the data in the representation of the input. Then in the next step, the digital image is converted into groups clusters. Clusters are divided into many regions. The same categories with same features of clusters are assembled within a group and different clusters are placed in other groups. Finally, the clusters are combined with respect to similar features and then represented in the form of segments. The clustered image depicts the clear representation of the digital image in order to highlight the regions and boundaries of the image. At last, the final image is presented in the form of segments. All colors of the image are separated in clusters.

Keywords: clustering, image segmentation, K-means function, local and global minimum, region

Procedia PDF Downloads 358