Wyniki wyszukiwania - BazTech

1

Optymalne planowanie budowlanego, wieloobiektowego przedsięwzięcia drogowego z użyciem algorytmu przeszukiwania genetycznego

Podolski Michał

Przegląd Budowlany

|

2025

|

R. 96, nr 1

112--116

PL

W artykule przedstawiono zagadnienie harmonogramowania budowlanego, wieloobiektowego przedsięwzięcia drogowego. Podczas wykonywania robót w takich przedsięwzięciach występują możliwości częściowego zazębiania się kolejnych czynności w obiektach. Ze względu na potrzebę maksymalnego skrócenia czasu zajęcia pracami budowlanymi poszczególnych obiektów zakłada się w nich ciągłość wykonywania robót. Założenia te prowadzą do zadania optymalizacyjnego polegającego na poszukiwaniu optymalnej kolejności wykonywania obiektów, która minimalizuje czas trwania przedsięwzięcia. W artykule to zagadnienie z powodzeniem rozwiązano za pomocą algorytmu przeszukiwania genetycznego i zilustrowano przykładem praktycznym.

EN

The article presents the issue of scheduling a multiunit road construction project. During the execution of works in such projects, there is a possibility of partial overlapping of successive activities in the units. Due to the need to maximally shorten the time of occupancy with construction works of the units, continuity of the works is assumed in them. These assumptions lead to the optimization task consisting in finding the optimal order of execution of the units that minimizes the duration of the project. In the article, this issue was successfully solved using a genetic search algorithm and illustrated by a case study.

2

Optimal coordination for directional overcurrent relays incorporating distribution generators: A comparative study

Abdullah Bassam Hamad, Alkhayyat Mahmood T.

Diagnostyka

|

2025

|

Vol. 26, No. 1

art. no. 2025111

EN

The integration of renewable energy-based distributed energy resources (DER) into distribution networks has increased due to rising load demand and growing concerns about global warming. The integration of DERs has transformed the operation of distribution networks from a passive to an active nature. As a result, a bidirectional flow of current occurs in the distribution networks. The protection of such systems is generally performed using directional overcurrent relays (DOCRs). However, optimal coordination of the DOCRs is necessary to ensure safe operation. Therefore, this paper aims to develop the optimal coordination of DOCRs using two nature-inspired techniques: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The developed algorithms are tested on IEEE 6-Bus and IEEE 15-Bus test systems in the MATLAB R2022b environment. To validate the effectiveness of the methods, the obtained results are compared with various upto-date algorithms. The comparison shows that the GA outperformed all the algorithms in minimizing the relay operation time for optimum coordination of overcurrent relays.

3

Task offloading in edge computing using integrated particle swarm optimization and genetic algorithm

Palaniappan Shabariram C., Ponnuswamy Priya P.

Advances in Science and Technology. Research Journal

|

2025

|

Vol. 19, no. 1

371--380

EN

n the ever-evolving landscape of smart city applications and Intelligent Transport Systems, Vehicular Edge Computing emerged as a game-changing technology. Imagine a world where computational resources are no longer restricted to distant cloud servers but are brought nearer to the vehicles and users. Task offloading enables the computation in edge and cloud server. This proximity not only minimizes network latency but also enables a unfold of vehicles to process tasks at the edge, offering a swift and interactive response to the scenarios of applications with delay sensitivity. To deal with this constraint, an integrated methodology is utilized to enhance the offloading process. The proposed system integrates the Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). The integrated system optimizes task allocation by exploring the solution space effectively and ensuring efficient resource utilization while minimizing latency. In the evaluation, PSO+GA exhibits enhanced adaptability to varying task sizes, facilitating efficient offloading to the edge as needed. Energy efficiency varies between the algorithms, with PSO+GA generally showing minimal energy consumption. When compared to already existing algorithms such as Energy aware offloading, no offloading and random offloading, PSO+GA outperformed these algorithms in system performance and less energy consumption by a factor of 1.18.

4

Optimization of gear pump operating parameters using genetic algorithms and performance analysis

Deptuła Adam, Osiński Piotr

Advances in Science and Technology. Research Journal

|

2025

|

Vol. 19, no. 3

211--227

EN

This study explores the optimization of operating parameters and the design of innovative gear pumps featuring multi-involute tooth profiles. Using a genetic algorithm, optimal parameters such as the basic radius, tooth height, and angles of involute profiles were determined. Results indicate that the proposed optimization framework significantly improves the volumetric and total efficiency of the pumps. Specifically, a volumetric efficiency increase of 8-12% and a reduction in noise levels by 3-5 dB were observed compared to conventional designs. The experimental validation confirmed the robustness of the proposed model, showcasing its potential for industrial applications. This work highlights the integration of advanced computational techniques with engineering design to achieve enhanced performance metrics.

5

3K method: time-optimal path planning for field robot

Kwaśniewski Konrad K., Gosiewski Zdzisław

Acta Mechanica et Automatica

|

2025

|

Vol. 19, no. 1

172--176

EN

In this study, a hybrid genetic-geometrical path finding method is presented. Its main feature is the division of the path-finding process into global and local path-finding to achieve a trajectory optimized under the shortest travel time condition in an environment filled with obstacles. To improve the reliability of the algorithm, a safety zone around obstacles is included. In this zone, the maximum velocity allowed for a robot is additionally limited to decrease the probability of collision due to noise in obstacle mapping, distraction from terrain irregularities or malfunction of the steering system. The simulation and real world experiment results are presented in another paper.

6

Zastosowanie algorytmu genetycznego do planowania przepływu przedmiotów w złożonych systemach produkcyjnych

Jardzioch Andrzej, Jardzioch Bartosz

Zarządzanie Przedsiębiorstwem

|

2024

|

Vol. 27, nr 2

16--22

EN

Modern production systems are characterised by a high degree of complexity, resulting from the use of many different technological processes, the parallel production of complex products, the use of advanced numerically controlled (CNC) machine tools and complex transport systems. At the same time, it is necessary to take into account many variables and constraints such as the availability of machines, tools and workers, stock levels in the warehouse, forecasted product demand, material handling capacities and the sequence in which individual tasks must be performed. Effective planning of the flow of items in such systems is key to achieving high productivity, minimising costs and ensuring on-time delivery. Traditional planning methods often prove insufficient in the face of dynamic and unpredictable production conditions. In this context, genetic algorithms (AG) represent a promising tool for optimising production processes. This paper presents an example of the application of a genetic algorithm to optimise the production process in an exemplary robotic production system. As the main optimisation criterion, the total sum of delays to be reckoned with when accepting a defined set of orders for execution. In addition, the total execution time for the set of orders and the machine tool utilisation rates during the entire production process were analysed. In order to be able to apply the genetic algorithm, it was necessary to build a parametric simulation model and integrate this model with the developed genetic algorithm. The simulation model was used to determine the objective function in the optimisation process implemented by the genetic algorithm.

7

Scheduling using Convolutional Neural Network in GPU environment

Świtalski Piotr, Siwiak Karolina

Studia Informatica : systems and information technology

|

2024

|

vol. 1 (30)

81--96

EN

Graphics processing units (GPU) have become the foundation of artificial intelligence. Machine learning was slow, inaccurate, and inadequate for many of today’s applications. The inclusion and utilization of GPUs made a remarkable difference in large neural networks. The numerous core processors on a GPU allow machine learning engineers to train complex models using many files relatively quickly. The ability to rapidly perform multiple computations in parallel is what makes them so effective; with a powerful processor, the model can make statistical predictions about very large amounts of data. GPUs are widely used in machine learning because they offer more power and speed than CPUs. In this paper, we show the use of GPU for solving a scheduling problem. The results show that this idea is useful, especially for large optimization problems.

8

Analiza wybranych metod walidacji krzyżowej w programie RSES

Kołpacki Radosław

Studia i Materiały Informatyki Stosowanej

|

2024

|

T. 16, nr 1

5--8

PL

W artykule przeprowadzono analizę zbioru danych za pomocą dwóch metod walidacji krzyżowej. Wykorzystano program RSES do identyfikacji kluczowych właściwości i relacji w zbiorze. Wyniki wykazują wpływ niektórych parametrów na potencjalną dokładność wyników.

EN

This article presents an analysis of a dataset using two cross-validation methods. The RSES program was employed to identify key properties and relationships within the dataset. The results indicate the impact of certain parameters on the potential accuracy of the outcomes.

9

Analiza wybranych metod walidacji krzyżowej w programie RSES

Bethke Beata

Studia i Materiały Informatyki Stosowanej

|

2024

|

T. 16, nr 1

11--14

PL

W artykule przeprowadzono analizę zbioru danych za pomocą dwóch metod walidacji krzyżowej. Wykorzystano program RSES do identyfikacji kluczowych właściwości i relacji w zbiorze. Wyniki wykazują wpływ niektórych parametrów na potencjalną dokładność wyników.

EN

This article presents an analysis of a dataset using two cross-validation methods. The RSES program was employed to identify key properties and relationships within the dataset. The results indicate the impact of certain parameters on the potential accuracy of the outcomes.

10

Solving the economic dispatch by new hybrid algorithm

Naama Bakhta, Dahmani Kaouther, Abrouche Amel, Bouzeboudja Hamid

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 5

164--169

EN

The problem of economic dispatch is the minimization of the total cost of production by satisfying the demand of the load. The resolution of this problem is a way of managing an electricity production system taking into account the constraints of equalities and inequalities, in other words it is to find the optimal production for a given combination of units in operation. The appearance of meta-heuristic methods which are part of artificial intelligence, has effectively contributed to solving this problem. Bee colony optimization is a very recent family of meta-heuristics. Its principle is based on the behavior of real bees in life. Bees have properties that are quite different from those of other insect species. They live in colonies, building their nests in tree trunks or other similar enclosed spaces. In this paper, we will apply the optimization by colony of bees in test systems of different sizes with the aim of minimizing the cost of production of electrical energy by taking into account the effect of the valve points of the power plants. In order to see the effectiveness of the proposed algorithm, it has been compared with other algorithms in the literature.

PL

Problem ekonomicznej wysyłki polega na minimalizacji całkowitego kosztu produkcji poprzez zaspokojenie zapotrzebowania na ładunek. Rozwiązanie tego problemu to sposób zarządzania systemem wytwarzania energii elektrycznej z uwzględnieniem ograniczeń równości i nierówności, czyli znalezienie optymalnej produkcji dla danej kombinacji pracujących jednostek. Pojawienie się metod metaheurystycznych wchodzących w skład sztucznej inteligencji skutecznie przyczyniło się do rozwiązania tego problemu. Optymalizacja kolonii pszczół to bardzo nowa rodzina metaheurystyk. Jego zasada opiera się na zachowaniu prawdziwych pszczół w życiu. Pszczoły mają właściwości zupełnie odmienne od właściwości innych gatunków owadów. Żyją w koloniach, budując gniazda w pniach drzew lub innych podobnych zamkniętych przestrzeniach. W tym artykule zastosujemy optymalizację przez rodzinę pszczół w układach testowych różnej wielkości w celu minimalizacji kosztów produkcji energii elektrycznej poprzez uwzględnienie wpływu punktów zaworowych elektrowni. Aby sprawdzić skuteczność zaproponowanego algorytmu, porównano go z innymi algorytmami dostępnymi w literaturze.

11

Optimizing SiGe-on-annealed DPSi heterostructures using raman spectroscopy and genetic algorithm for enhanced material characterization and performance

Gouder Soraya, Yousfi Laatra, Guiza Dhaouadi, Mahamdi Ramdane, Berbezier Isabelle

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 4

165--170

EN

In our preceding investigation, we delved into the intricacies of SiGe alloys on double porous silicon (DPSi) through Raman spectroscopy, uncovering previously unknown connections between Raman peak shifts, stresses, and the concentration of Ge in the SiGe alloys in porous materials.A standout feature of this study lies in its distinct approach — a comparison of results employing a genetic algorithm. This method offers a comprehensive analysis of the data, enhancing our understanding of the intricate relationships at play. Validated through the frequency method, our results yield valuable insights into epitaxial growth on DPSi, presenting a nuanced perspective on the intricate interplay between Raman spectroscopy, stress, and alloy composition. These findings not only contribute to the evolving understanding of SiGe alloys but also pave the way for further advancements in the field of epitaxial growth on innovative substrates like DPSi.

PL

W naszym poprzednim badaniu zagłębiliśmy się w zawiłości stopów SiGe na podwójnie porowatym krzemie (DPSi) za pomocą spektroskopii Ramana, odkrywając nieznane wcześniej powiązania między przesunięciami pików Ramana, naprężeniami i stężeniem Ge w stopach SiGe w materiałach porowatych. Cechą tego badania jest odrębność podejścia — porównanie wyników z wykorzystaniem algorytmu genetycznego. Metoda ta umożliwia wszechstronną analizę danych, co pozwala lepiej zrozumieć złożone zależności. Nasze wyniki, potwierdzone metodą częstotliwości, dostarczają cennych informacji na temat wzrostu epitaksjalnego na DPSi, prezentując zniuansowaną perspektywę na skomplikowane wzajemne oddziaływanie między spektroskopią Ramana, naprężeniem i składem stopu. Odkrycia te nie tylko przyczyniają się do lepszego zrozumienia stopów SiGe, ale także torują drogę do dalszych postępów w dziedzinie wzrostu epitaksjalnego na innowacyjnych podłożach, takich jak DPSi.

12

Optimal distribution of reactive power by hybrid metaheuristic methods applied to the west Algerian network

Cherki Imene, Khalfallah Naima, Chaker Abdelkader

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 8

170--174

EN

This research focuses on the utilization of artificial intelligence through the sequential and integrated crossover of two population metaheuristic methods: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). These methods are applied to solve the Optimal Reactive Power Flow (ORPF) in the West Algerian network, comprising 102 nodes. The objective of this combination is to demonstrate its impact compared to non-hybrid metaheuristic methods in reducing energy losses while effectively improving various aspects such as voltage levels, the flow of active and reactive energy in the lines, transformation ratios of transformers, and the execution time of the process. Following this application, a comparative study of the results from different methods was conducted.

PL

Niniejsze badania koncentrują się na wykorzystaniu sztucznej inteligencji poprzez sekwencyjne i zintegrowane krzyżowanie dwóch metod metaheurystycznych populacji: algorytmu genetycznego (GA) i optymalizacji roju cząstek (PSO). Metody te są stosowane do rozwiązania optymalnego przepływu mocy biernej (ORPF) w sieci zachodnioalgierskiej, obejmującej 102 węzły. Celem tej kombinacji jest wykazanie jej wpływu w porównaniu z niehybrydowymi metodami metaheurystycznymi na redukcję strat energii przy jednoczesnej skutecznej poprawie różnych aspektów, takich jak poziomy napięcia, przepływ energii czynnej i biernej w liniach, współczynniki transformacji transformatorów i czas realizacji procesu. Po tej aplikacji przeprowadzono badanie porównawcze wyników różnych metod.

13

Dynamic scheduling strategy and algorithm for mixed batch scheduling in vacuum freeze-dried fruit processes

Huang JinDian, Hong YingHan, Lin KeYe

Production Engineering Archives

|

2024

|

Vol. 30, Iss. 4

477--490

EN

Vacuum freeze-dried fruit processes consisting of heating and holding are modelled as a mixed batch scheduling with the objective of minimizing the makespan. The jobs differ from each other in job family, size, weight and ready time. The batch processing time is determined by the longest job and the total weight of the jobs in the batch. A mixedinteger linear programming model is developed and tested with small-scale examples. Typical batch scheduling strategies are analysed and a machine based dynamic programming strategy is proposed. The machine-based dynamic scheduling strategy is applied to design improved genetic and particle swarm optimization algorithms, which demonstrate the effectiveness of this strategy. The worst-case ratio of the algorithms using machine dynamic programming strategy are proved. Numerical experiments show that the heuristic algorithm, genetic algorithm, and particle swarm optimization algorithm based on machine dynamic scheduling strategy outperform related algorithms using greedy and job-based dynamic scheduling strategies.

14

Analysis effect of parameters of genetic algorithm on a model for optimization design of sustainable supply chain network under disruption risks

Al-Zuheri Atiya, Ketan Hussein S., Alwan Layla L.

Management Systems in Production Engineering

|

2024

|

nr 2 (32)

252--264

EN

Over the last decade, our world exposed to many types of unpredictable disasters (recently Coronavirus). These disasters have clearly shown the uncertainty and vulnerability of supply chain systems. Also, it confirmed that adopting Just-in-Time (JIT) strategy to reduce the logistic chain cost may lead to inbuilt complexity and risks. Efficient tools are therefore needed to make complexity optimized supply chain decisions. Evolutionary algorithms, including genetic algorithms (GA), have proven effective in identifying optimal solutions that address the trade-offs between total supply chain cost and carbon emissions regulatory policy represented by carbon tax charges. These solutions pertain to the design challenges of supply networks exposed to potential disruption risks. However, GA have a set of parameters must be chosen for effective and robust performance of the algorithms. This paper aims to set the most suitable values of these parameters that used via GA – ased optimization cost and risk reduction model in firms using a JIT as a delivery system. The model has been conceptualized for addressing the design complexities of the supply chain, referred to as SCRRJITS (Simultaneous Cost and Risk Reduction in a Just-in-Time System). A complete analysis of the different parameters and operators of the algorithm is carried out using design of experiments approach. The algorithm performance measure used in this study is convergence of solutions. The results show the extent to which the quality of solution can be changed depending on selection of these parameters.

15

Selecting Update Blocks of Convolutional Neural Networks using Genetic Algorithm in Transfer Learning

Hasan Mehedi, Ibrahim Muhammad, Ali Sawkat

Machine Graphics and Vision

|

2024

|

Vol. 33, No. 3/4

55--70

EN

The performance of convolutional neural networks (CNN) for computer vision problems depends heavily on their architectures. Transfer learning performance of a CNN strongly relies on selection of its trainable layers. Selecting the most effective update layers for a certain target dataset often requires expert knowledge on CNN architecture which many practitioners do not possess. General users prefer to use an available architecture (e.g. GoogleNet, ResNet, EfficientNet etc.) that is developed by domain experts. With the ever-growing number of layers, it is increasingly becoming difficult and cumbersome to handpick the update layers. Therefore, in this paper we explore the application of a genetic algorithm to mitigate this problem. The convolutional layers of popular pre-trained networks are often grouped into modules that constitute their building blocks. We devise a genetic algorithm to select blocks of layers for updating the parameters. By experimenting with EfficientNetB0 pre-trained on ImageNet and using three popular image datasets - namely Food-101, CIFAR-100 and MangoLeafBD - as target datasets, we show that our algorithm yields similar or better results than the baseline in terms of accuracy, and requires lower training and evaluation time due to learning a smaller number of parameters. We also devise a measure called block importance to measure each block’s efficacy as an update block and analyze the importance of the blocks selected by our algorithm.

16

The use of a genetic algorithm in the process of optimizing the shape of a three-dimensional periodic beam

Dolinski Lukasz, Zak Arkadiusz, Waszkowiak Wiktor, Kowalski Pawel, Szkopek Jacek

Journal of Theoretical and Applied Mechanics

|

2024

|

Vol. 62 nr 3

601--613

EN

Mechanical periodic structures exhibit unusual dynamic behavior thanks to the periodicity of their structures, which can be attributed to their cellular arrangement. The source of this periodicity may result from periodic variations of material properties within their cells and/or variations in the cell geometry. The authors present the results of their studies on the optimization of physical parameters of a three-dimensional axisymetrical periodic beam in order to obtain the desired vibroacoustic properties. The aim of the optimization process of the unit cell shape was to obtain band gaps of a given width and position in the frequency spectrum.

17

Emergency response capability evaluation of power transmission and transformation projects based on projection pursuit method

Shuo Li, Xin Fu, Yuanyuan He

Journal of Power Technologies

|

2024

|

Vol. 104, nr 2

116--129

EN

To adapt to the rapid development of power transmission and transformation projects, improve their emergency response capability level, and reduce the losses caused by accidents, the projection pursuit method was introduced into the emergency response capability evaluation of power transmission and transformation projects. The emergency response capability evaluation system of power transmission and transformation projects has been established mostly from each composition and structure of power transmission and transformation engineering systems, and highly subjective evaluation methods have been adopted to assess the models established. In this study, a total of 19 concrete indexes were selected from 4 aspects-monitoring and early warning capability, emergency control capability, emergency rescue and disposal capability, and emergency support capability-to establish an emergency response capability evaluation index system of power transmission and transformation projects. Then, an emergency response capability evaluation model for power transmission and transformation projects was constructed based on the projection pursuit model, followed by optimization using real code accelerated genetic algorithm (RAGA); for high-dimensional data, this model could directly find the structure and features of data itself, avoiding the limitations of subjective judgment and contributing to more truthful and reliable evaluation results; finally, this model was used to evaluate and analyze the emergency response capability of six power transmission and transformation projects: GZXS S00kV, JXXYS00kV, QHXN 750kV, YNZT S00kV, JSNJ S00kV, and SXXA 750kV. The results show that the six power transmission and transformation projects are different in the emergency response capability level; the emergency response capability level of power transmission and transformation projects is greatly affected by the early warning personnel deployment capability, daily emergency drill capability, emergency technology implementation capability, emergency training and education capability, and risk response capability.

18

Optimization of computer ontologies for e-courses in information and communication technologies

Sabitova Nazym, Tikhonov Yuriy, Lakhno Valerii, Makulov Kariyrbek, Kryvoruchko Olena, Chubaievskyi Vitalyi, Desiatko Alona, Zhumadilova Mereke

International Journal of Electronics and Telecommunications

|

2024

|

Vol. 70, No. 1

191--197

EN

A methodology is proposed for modifying computer ontologies (CO) for electronic courses (EC) in the field of information and communication technologies (ICT) for universities, schools, extracurricular institutions, as well as for the professional retraining of specialists. The methodology includes the modification of CO by representing the formal ontograph of CO in the form of a graph and using techniques for working with the graph to find optimal paths on the graph using applied software (SW). A genetic algorithm (GA) is involved in the search for the optimal CO. This will lead to the division of the ontograph into branches and the ability to calculate the best trajectory in a certain sense through the EC educational material, taking into account the syllabus. An example is considered for the ICT course syllabus in terms of a specific topic covering the design and use of databases. It is concluded that for the full implementation of this methodology, a tool is needed that automates this procedure for developing EC and/or electronic textbooks. An algorithm and a prototype of software tools are also proposed, integrating machine methods of working with CO and graphs.

19

Genetic algorithm-based decision tree optimization for detection of dementia through MRI analysis

Anuradha Govada, Davu Harini, Karri Muthyalanaidu

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2024

|

T. 14, nr 1

83--89

EN

Dementia is a devastating neurological disorder that affects millions of people globally, causing progressive decline in cognitive function and daily living activities. Early and precise detection of dementia is critical for optimal dementia therapy and management however, the diagnosis of dementia is often challenging due to the complexity of the disease and the wide range of symptoms that patients may exhibit. Machine learning approaches are becoming progressively more prevalent in the realm of image processing, particularly for disease prediction. These algorithms can learn to recognize distinctive characteristics and patterns that are suggestive of specific diseases by analyzing images from multiple medical imaging modalities. This paper aims to develop and optimize a decision tree algorithm for dementia detection using the OASIS dataset, which comprises a large collection of MRI images and associated clinical data. This approach involves using a genetic algorithm to optimize the decision tree model for maximum accuracy and effectiveness. The ultimate goal of the paper is to develop an effective, non-invasive diagnostic tool for early and accurate detection of dementia. The GA-based decision tree, as proposed, exhibits strong performance compared to alternative models, boasting an impressive accuracy rate of 96.67% according to experimental results.

PL

Demencja jest wyniszczającym zaburzeniem neurologicznym, które dotyka miliony ludzi na całym świecie, powodując postępujący spadek funkcji poznawczych i codziennych czynności życiowych. Wczesne i precyzyjne wykrywanie demencji ma kluczowe znaczenie dla optymalnej terapii i zarządzania demencją, jednak diagnoza demencji jest często trudna ze względu na złożoność choroby i szeroki zakres objawów, które mogą wykazywać pacjenci. Podejścia oparte na uczeniu maszynowym stają się coraz bardziej powszechne w dziedzinie przetwarzania obrazu, szczególnie w zakresie przewidywania chorób. Algorytmy te mogą nauczyć się rozpoznawać charakterystyczne cechy i wzorce, które sugerują określone choroby, analizując obrazy z wielu modalności obrazowania medycznego. Niniejszy artykuł ma na celu opracowanie i optymalizację algorytmu drzewa decyzyjnego do wykrywania demencji przy użyciu zbioru danych OASIS, który obejmuje duży zbiór obrazów MRI i powiązanych danych klinicznych. Podejście to obejmuje wykorzystanie algorytmu genetycznego do optymalizacji modelu drzewa decyzyjnego w celu uzyskania maksymalnej dokładności i skuteczności. Ostatecznym celem artykułu jest opracowanie skutecznego, nieinwazyjnego narzędzia diagnostycznego do wczesnego i dokładnego wykrywania demencji. Zaproponowane drzewo decyzyjne oparte na GA wykazuje wysoką wydajność w porównaniu z alternatywnymi modelami, szczycąc się imponującym współczynnikiem dokładności wynoszącym 96,67% zgodnie z wynikami eksperymentalnymi.

20

Software Defect Prediction Using Non-Dominated Sorting Genetic Algorithm and k-Nearest Neighbour Classifier

Azzeh Mohammad, Nassif Ali Bou, Talib Manar Abu, Iqba Hajra

e-Informatica Software Engineering Journal

|

2024

|

Vol. 18, nr 1

240103

EN

Background: Software Defect Prediction (SDP) is a vital step in software development. SDP aims to identify the most likely defect-prone modules before starting the testing phase, and it helps assign resources and reduces the cost of testing. Aim: Although many machine learning algorithms have been used to classify software modules based on static code metrics, the k-Nearest Neighbors (kNN) method does not greatly improve defect prediction because it requires careful set-up of multiple configuration parameters before it can be used. To address this issue, we used the Non-dominated Sorting Genetic Algorithm (NSGA-II) to optimize the parameters in the kNN classifier with favor to improve SDP accuracy. We used NSGA-II because the existing accuracy metrics often behave differently, making an opposite judgment in evaluating SDP models. This means that changing one parameter might improve one accuracy measure while it decreases the others. Method: The proposed NSGAII-kNN model was evaluated against the classical kNN model and state-of-the-art machine learning algorithms such as Support Vector Machine (SVM), Naïve Bayes (NB), and Random Forest (RF) classifiers. Results: Results indicate that the GA-optimized kNN model yields a higher Matthews Coefficient Correlation and higher balanced accuracy based on ten datasets. Conclusion: The paper concludes that integrating GA with kNN improved defect prediction when applied to large or small or large datasets.