Genetic Algorithm that can Dynamically Change Number of Individuals and Accuracy

Magang merupakan proses yang penting dalam proses belajar mengajar di SMK. Secara spesifik magang di SMK disebut Prakerin (Praktek Kerja Industri). Penempatan magang harus memerhatikan kompetensi siswa, kuota dari perusahaan, kesesuaian jurusan dengan lowongan, dan penghasilan orang tua. Penempatan magang secara manual selama ini memakan waktu, sehingga kurang efisien. Oleh karena itu, penelitian ini fokus mengembangkan aplikasi dengan menerapkan pendekatan Algoritme Genetika untuk mempermudah dalam menentukan penempatan magang dengan menerapkan aturan yang berlaku. Algoritme Genetika dinilai sebagai algoritma yang relevan dan solutif untuk diterapkan dalam penyelesaian masalah optimasi kompleks. Masalah yang dimaksud umumnya adalah masalah yang sulit dilakukan dengan menerapkan metode konvensional. Algoritme Genetika memberikan hasil yang lebih baik untuk setiap iterasi pencarian solusi. Hasil fitness terbaik dengan nilai 0.0014286 diperoleh pada jumlah individu 200, jumlah generasi 200, persentase crossover 50% dan mutasi 10%. Hasil validasi dengan pihak SMK menyatakan bahwa sistem ini mudah untuk digunakan dan bermanfaat bagi pihak SMK, dengan rata-rata persentase kualitas sistem 82,5%. Algoritme Genetika efektif untuk diterapkan pada studi kasus penjadwalan atau penempatan magang yang memiliki karakteristik data yang kompleks. AbstractInternships are an important component of teaching and learning activities in vocational high schools (SMK). Specifically the internship in SMK is called Prakerin (Industrial Work Practice). The internship placement should considering student competence, number of vacancies, the suitability of majors with vacancy, and the income of the parent. During this time, manual internship placement takes more time, so less efficient. Therefore, this research tries to approach using Genetic Algorithm to make it easier in determining the internship placement by applying the applicable rules. The Genetic Algorithm is judged as the right algorithm used in solving complex optimization problems, which is difficult to do by conventional methods. The Genetic Algorithm provides better results for each iteration of the solution search. Best fitness results with value 0.0014286 obtained on the number of individuals 200, the number of generation 200, the percentage of crossover 50% and the mutation 10%. Validation results with the SMK stated that the system is easy to use and beneficial to the SMK, with an average percentage of system quality about 82.5%. Genetic Algorithms are effective to apply to scheduling case studies or internship placements that have complex data characteristics.

Download Full-text

The use of fuzzy sets to determine the parameters of genetic algorithms that provide approximately the same execution time on the CPU and GPU

Journal of Physics Conference Series ◽

10.1088/1742-6596/2131/3/032025 ◽

2021 ◽

Vol 2131 (3) ◽

pp. 032025

Author(s):

Oleg Agibalov ◽

Nikolay Ventsov

Keyword(s):

Genetic Algorithm ◽

Processing Time ◽

Fuzzy Numbers ◽

Subject Area ◽

Initial Information ◽

Hardware Architectures ◽

The Subject ◽

Gpu Architectures ◽

Number Of Individuals ◽

Gpu Architecture

Abstract The problem under consideration consists in choosing the number of k individuals, so that the time for processing k individuals by the genetic algorithm (GA) on the CPU architecture is close to the time for processing l individuals on the GPU architecture by the genetic algorithm. The initial information is data arrays containing information about the processing time of a given number of individuals by the genetic algorithm on the available hardware architectures. Fuzzy numbers are determined based on these arrays?~? and?~?, describing the processing time of a given number of individuals, respectively, on the CPU and GPU architectures. The peculiarities of the subject area do not allow considering the well-known methods of comparison based on the equalities of the membership functions and the nearest clear sets as adequate. Based on the known formula “close to Y (around Y)” the way to compare fuzzy numbers?~? and?~? was developed in order to determine the degree of closeness of the processing time of k and l individuals, respectively, on the hardware architectures of the CPU and GPU.

Download Full-text

CROSS-POLLINATING PARALLEL GENETIC ALGORITHMS FOR MULTI-OBJECTIVE SEARCH AND OPTIMIZATION

International Journal of Foundations of Computer Science ◽

10.1142/s012905410500298x ◽

2005 ◽

Vol 16 (02) ◽

pp. 261-280 ◽

Cited By ~ 4

Author(s):

LUCAS A. WILSON ◽

MICHELLE D. MOORE

Keyword(s):

Genetic Algorithm ◽

Empirical Studies ◽

Path Generation ◽

Parallel Genetic Algorithm ◽

Optimization Criteria ◽

Parallel Genetic Algorithms ◽

Multi Objective ◽

Hard Constraints ◽

Future Work ◽

Number Of Individuals

This paper discusses the design of a parallel genetic algorithm to generate solutions to multi-objective problems. The algorithm uses multiple optimization criteria, independent cross-pollinating populations, and handles multiple hard constraints. Individuals in the population consist of multiple chromosomes. The complexity of the algorithm is the number of generations processed times O(N2) where N is the total number of individuals used for path generation on any of the optimizations. The results of initial empirical studies on the effects of pollination and recommendations for possible future work are presented.

Download Full-text

Genetic algorithm efficiency improvement in the course of set cover problem solution

Vestnik of Don State Technical University ◽

10.23947/1992-5980-2019-19-4-389-397 ◽

2020 ◽

Vol 19 (4) ◽

pp. 389-397

Author(s):

I. S. Konovalov ◽

V. A. Fatkhi ◽

V. G. Kobak

Keyword(s):

Genetic Algorithm ◽

Single Point ◽

Set Cover ◽

Covering Problem ◽

Problem Solution ◽

Crossover Operator ◽

Deletion Mutations ◽

Set Cover Problem ◽

Cover Problem ◽

Number Of Individuals

Introduction. Practical tasks (location of service points, creation of microcircuits, scheduling, etc.) often require an exact or approximate to exact solution at a large dimension. In this case, achieving an acceptable result requires solving a set cover problem, fundamental for combinatorics and the set theory. An exact solution can be obtained using exhaustive methods; but in this case, when the dimension of the problem is increased, the time taken by an exact algorithm rises exponentially. For this reason, the precision of approximate methods should be increased: they give a solution that is only approximate to the exact one, but they take much less time to search for an answer at a large dimension.Materials and Methods. One of the ways to solve the covering problem is described, it is a genetic algorithm. The authors use a modification of the Goldberg model and try to increase its efficiency through various types of mutation and crossover operators. We are talking about gene mutations, two-point mutations, addition and deletion mutations, insertion and deletion mutations, saltation, mutations based on inversion. The following types of crossover operator are noted: single-point, two-point, three-point and their versions with restrictions, uniform, triad. The effect of the stopping condition and the probability values of genetic operators on the accuracy of the solutions is investigated. It is shown how an increase in the number of individuals in a generation affects the efficiency of a solution. Research Results. The experiment results allow us to draw three conclusions. 1) It is recommended to use a combination of gene mutation and single-point crossing. 2) With an increase in the number of individuals, the accuracy of the result and the time to obtain it increases. The average deviation from the exact result at a task size of 25 × 25 was 0%, at 50 × 50 – 0%, at 75 × 75 – 0.013%, at 100 × 100 – 0%, at 110 × 110 – 0% (the number of individuals was 500).3) It is advisable to use the probabilities of the mutation and crossover operator 100% and 100%, respectively. Discussion and Conclusions. Recommendations are given to improve the efficiency of covering problem solution. To this end, a preferred combination of the genetic algorithm parameters, of types of crossover and mutation operators is indicated.

Download Full-text

On the issue of fuzzy timing estimations of the algorithms running at GPU and CPU architectures

E3S Web of Conferences ◽

10.1051/e3sconf/201913501082 ◽

2019 ◽

Vol 135 ◽

pp. 01082 ◽

Cited By ~ 1

Author(s):

Oleg Agibalov ◽

Nikolay Ventsov

Keyword(s):

Genetic Algorithm ◽

Genetic Algorithms ◽

Graphics Processing Unit ◽

Processing Unit ◽

Running Time ◽

Average Value ◽

Central Processing ◽

Graphics Processing ◽

Number Of Individuals ◽

Gpu Architecture

We consider the task of comparing fuzzy estimates of the execution parameters of genetic algorithms implemented at GPU (graphics processing unit’ GPU) and CPU (central processing unit) architectures. Fuzzy estimates are calculated based on the averaged dependencies of the genetic algorithms running time at GPU and CPU architectures from the number of individuals in the populations processed by the algorithm. The analysis of the averaged dependences of the genetic algorithms running time at GPU and CPU-architectures showed that it is possible to process 10’000 chromosomes at GPU-architecture or 5’000 chromosomes at CPUarchitecture by genetic algorithm in approximately 2’500 ms. The following is correct for the cases under consideration: “Genetic algorithms (GA) are performed in approximately 2, 500 ms (on average), ” and a sections of fuzzy sets, with a = 0.5, correspond to the intervals [2, 000.2399] for GA performed at the GPU-architecture, and [1, 400.1799] for GA performed at the CPU-architecture. Thereby, it can be said that in this case, the actual execution time of the algorithm at the GPU architecture deviates in a lesser extent from the average value than at the CPU.

Download Full-text

Appropriate Setting of Genetic Algorithm for Parameter Identification of Bodner-Partom Material Model

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.816.383 ◽

2015 ◽

Vol 816 ◽

pp. 383-388

Author(s):

Jozef Bocko ◽

Iveta Glodová ◽

Viera Nohajová

Keyword(s):

Genetic Algorithm ◽

Parameter Identification ◽

Material Model ◽

Quality Of Results ◽

Generation Number ◽

Basic Vocabulary ◽

Number Of Individuals

This article monitors the influence of the number of individuals in generation, number of elite individuals and the time of algorithm run on the quality of results. Reasons for monitoring these influences and the process of calculation are explained in the first part. Basic vocabulary used in this work is described as next. The results of calculations are written down in tables and illustrated on figures. At the end of the article the most suitable parameters for next calculations using Bodner-Partom (B-P) model are summarized.

Download Full-text

Membangun Sistem Informasi Penjadwalan Dengan Metode Algoritma Genetika Pada Laboratorium Teknik Informatika Universitas Muhammadiyah Maluku Utara

Jurnal Ilmiah ILKOMINFO - Ilmu Komputer & Informatika ◽

10.47324/ilkominfo.v1i2.13 ◽

2018 ◽

Vol 1 (2) ◽

Author(s):

Agil Assagaf ◽

Adelina Ibrahim ◽

Catur Suranto

Keyword(s):

Genetic Algorithm ◽

Genetic Algorithms ◽

Optimal Parameter ◽

Optimal Schedule ◽

Optimization Method ◽

Computational Approach ◽

Biological Processes ◽

Practice Schedule ◽

Parameter Values ◽

Number Of Individuals

Abstrak : Penjadwalan praktikum merupakan proses penyusunan jadwal pelaksanaan yang menginformasikan sejumlah mata kuliah, dosen yang mengajar, ruang, serta waktu kegiatan perkuliahan di laboratorium. Perlu diperhatikan beberapa aspek untuk menyusun jadwal perkuliahan yang sesuai dengan kebutuhan. Aspek yang perlu diperhatikan antara lain adalah aspek dari dosen yang mengajar, mata kuliah yang diajar. Penyusunan jadwal secara manual cenderung membutuhkan waktu yang lebih lama dan ketelitian yang cukup bagi pembuat jadwal. Untuk dapat membuat jadwal yang optional, dibutuhkan metode optimasi. Pada penelitian ini, akan diuji coba metode optimasi dalam pembuatan jadwal praktikum yaitu Algoritma Genetika. Algoritma genetika merupakan pendekatan komputasional untuk menyelesaikan masalah yang dimodelkan dengan proses biologi dari evolusi. Parameter-parameter Algoritma Genetika yang mempengaruhi jadwal perkuliahan yang dihasilkan adalah jumlah individu, probabilitas crossover, probabilitas mutasi serta metode seleksi, crossover yang digunakan. Pengujian dilakukan dengan cara mencari nilai parameter-parameter algoritma genetika yang paling optimal dalam jadwal perkuliahan. Hasil penelitian menunjukkan bahwa dengan jumlah generasi, jumlah individu, probabilitas crossover dan probabilitas mutasi dapat menghasilkan jadwal yang paling optimal.Kata kunci: Optimasi, Penjadwalan, Seleksi, Crossover, Mutasi, Algoritma GenetikaAbstract : Practical scheduling is the process of preparation of an implementation schedule that informs a number of courses, lecturers who teach, space, and time of lecture activities in the laboratory. It should be noted several aspects to arrange lecture schedule in accordance with the needs. Aspects that need to be considered include aspects of lecturers who teach, courses taught. Manual scheduling tends to take longer and enough accuracy for the schedule maker. To be able to create an optional schedule, an optimization method is required. In this research, will be tested the optimization method in the preparation of the practice schedule that is Genetic Algorithm. Genetic algorithms are a computational approach to solving problems modeled by biological processes of evolution. The parameters of the Genetic Algorithm affecting the course schedule are the number of individuals, the probability of crossover, the probability of mutation and the method of selection, the crossover used. Testing is done by finding the most optimal parameter values of genetic algorithm in lecture schedule. The results show that with the number of generations, the number of individuals, the probability of crossover and the probability of mutation can produce the most optimal schedule. Keywords: Optimization, Scheduling, Selection, Crossover, Mutation, Genetic Algorithm

Download Full-text

fGAAM: A fast and resizable genetic algorithm with aggressive mutation for feature selection

Pattern Analysis and Applications ◽

10.1007/s10044-021-01000-z ◽

2021 ◽

Author(s):

Izabela Rejer ◽

Jarosław Jankowski

Keyword(s):

Genetic Algorithm ◽

Feature Selection ◽

Classification Accuracy ◽

Processing Time ◽

Nsga Ii ◽

Penalty Term ◽

Reference Methods ◽

Number Of Individuals

AbstractThe paper introduces a modified version of a genetic algorithm with aggressive mutation (GAAM) called fGAAM (fast GAAM) that significantly decreases the time needed to find feature subsets of a satisfactory classification accuracy. To demonstrate the time gains provided by fGAAM both algorithms were tested on eight datasets containing different number of features, classes, and examples. The fGAAM was also compared with four reference methods: the Holland GA with and without penalty term, Culling GA, and NSGA II. Results: (i) The fGAAM processing time was about 35% shorter than that of the original GAAM. (ii) The fGAAM was also 20 times quicker than two Holland GAs and 50 times quicker than NSGA II. (iii) For datasets of different number of features, classes, and examples, another number of individuals, stored for further processing, provided the highest acceleration. On average, the best results were obtained when individuals from the last 10 populations were stored (time acceleration: 36.39%) or when the number of individuals to be stored was calculated by the algorithm itself (time acceleration: 35.74%). (iv) The fGAAM was able to process all datasets used in the study, even those that, because of their high number of features, could not be processed by the two Holland GAs and NSGA II.

Download Full-text

Perceptual Learning of Vocoded Speech With and Without Contralateral Hearing: Implications for Cochlear Implant Rehabilitation

Journal of Speech Language and Hearing Research ◽

10.1044/2020_jslhr-20-00385 ◽

2020 ◽

pp. 1-10

Author(s):

Martin Chavant ◽

Alexis Hervais-Adelman ◽

Olivier Macherey

Keyword(s):

Cochlear Implant ◽

Perceptual Learning ◽

Speech Signal ◽

Training Phase ◽

Monosyllabic Words ◽

Low Pass ◽

Contralateral Ear ◽

Number Of Individuals ◽

Insight Into ◽

Vocoded Speech

Purpose An increasing number of individuals with residual or even normal contralateral hearing are being considered for cochlear implantation. It remains unknown whether the presence of contralateral hearing is beneficial or detrimental to their perceptual learning of cochlear implant (CI)–processed speech. The aim of this experiment was to provide a first insight into this question using acoustic simulations of CI processing. Method Sixty normal-hearing listeners took part in an auditory perceptual learning experiment. Each subject was randomly assigned to one of three groups of 20 referred to as NORMAL, LOWPASS, and NOTHING. The experiment consisted of two test phases separated by a training phase. In the test phases, all subjects were tested on recognition of monosyllabic words passed through a six-channel “PSHC” vocoder presented to a single ear. In the training phase, which consisted of listening to a 25-min audio book, all subjects were also presented with the same vocoded speech in one ear but the signal they received in their other ear differed across groups. The NORMAL group was presented with the unprocessed speech signal, the LOWPASS group with a low-pass filtered version of the speech signal, and the NOTHING group with no sound at all. Results The improvement in speech scores following training was significantly smaller for the NORMAL than for the LOWPASS and NOTHING groups. Conclusions This study suggests that the presentation of normal speech in the contralateral ear reduces or slows down perceptual learning of vocoded speech but that an unintelligible low-pass filtered contralateral signal does not have this effect. Potential implications for the rehabilitation of CI patients with partial or full contralateral hearing are discussed.

Download Full-text