Training neural networks: backpropagation vs. genetic algorithms

Siddique, M. N. H.; Tokhi, M. O.

doi:10.1109/ijcnn.2001.938792

Cited by 72 publications

(52 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Dalam prosedur backpropagation dapat diketahui galat antara pola keluaran dengan pola data keluaran yang diinginkan [14]. Untuk mengetahui besar galat pada output layer digunakan Mean Square Error (MSE) yang dihitung sesuai persamaan (8).…”

Section: Backpropagationunclassified

Pengenalan Suara Burung Menggunakan Mel Frequency Cepstrum Coefficient dan Jaringan Syaraf Tiruan pada Sistem Pengusir Hama Burung

Budiman

Nursyeha

Rivai

et al. 2016

JNTE

View full text Add to dashboard Cite

Abstrak-Indonesia merupakan salah satu negara agraris yang memproduksi hasil pertanian. Namun, Indonesia masih mengimpor beras dari negara lain dikarenakan penurunan hasil panen. Salah satu faktor menurunnya produksi beras Indonesia adalah akibat serangan hama burung. Ekosistem sawah mengandung berbagai macam spesies burung, baik hama maupun non-hama. Burung non-hama menolong petani melawan hama serangga. Petani menggunakan metode tradisional untuk mengusir hama burung. Pada penelitian ini telah dirancang perangkat lunak untuk mengenali jenis burung berdasarkan kicauannya. Voice Activity Detection (VAD) digunakan untuk mendeteksi adanya kicau burung. Metode ekstraksi ciri suara dari kicau burung menggunakan Mel Frequency Cepstrum Coefficient (MFCC) dan Fast Fourier Transform (FFT). Jaringan Syaraf Tiruan digunakan untuk mengenali pola hasil ekstraksi. Selanjutnya, audiosonic bird repeller digunakan sebagai metode pengusiran hama burung. Hasil identifikasi offline dengan menggunakan MFCC didapatkan tingkat keberhasilan mencapai 90% untuk variasi kicauan dan jenis burung, sedangkan dengan FFT mencapai 68%. Hasil identifikasi online untuk spesimen burung bondol didapatkan tingkat keberhasilan 70% dengan menggunakan MFCC, dan 30% dengan FFT. Selain itu, suara tembakan merupakan suara yang paling baik digunakan untuk mengusir hama burung. Tingkat keberhasilan pengenalan suara burung menggunakan ekstraksi ciri MFCC lebih tinggi jika dibandingkan dengan ekstraksi ciri dengan menggunakan FFT. Kata Kunci : Hama Burung, Voice Activity Detection, Fast Fourier Trasnform, dan Mel Frequency Cepstrum Coefficient.Abstract-Indonesia is one of the agricultural country that produces crops. Nevertheless, Indonesia still imports rice from other countries because of crop decreasement. One of its factors is caused by bird pests. In the ricefields ecosystem, the variety of bird species can be classified as a pest and as a non-pest. Nonpest birds usually help farmer against insects. Farmers are using traditional methods to repel bird pests. In this research, a software to recognize species of birds has been designed. The system is based on birdchirp types. A mono-microphone is used to capture the sound. Voice Activity Detection (VAD) method is used for birdchirp detection. Mel Frequency Cepstrum Coefficient (MFCC) and Fast Fourier Transform (FFT) are used to extract birdchirp feature. Artificial Neural Network is utilized to recognize the pattern of birdchirp feature. Furthermore, audiosonic bird repeller is used to repel bird pests. In the offline mode testing, success level using MFCC feature extraction is up to 90% for birdchirp variation, while up to 68% using FFT. In the online mode, the average success level using MFCC feature extraction is 70% for finch birds, while 30% using FFT. In addition, a gunshot is a best sound to repel bird pests. The success rate of bird voice recognitions using MFCC feature extraction is higher than that using FFT.

show abstract

Section: Backpropagationunclassified

Pengenalan Suara Burung Menggunakan Mel Frequency Cepstrum Coefficient dan Jaringan Syaraf Tiruan pada Sistem Pengusir Hama Burung

Budiman

Nursyeha

Rivai

et al. 2016

JNTE

View full text Add to dashboard Cite

show abstract

“…In the first case many problems coming up during separate use of the two technologies are being overcome since their combination offers a fuzzy inference system which uses a neural network learning process [24]. The latter approach, takes advantage of the fact that genetic algorithms can provide optimal network architectures [33]. In other words, genetic programming is applied on an initial population of neural networks in order to obtain an ideal one via reproduction, crossover and mutation.…”

Section: Artificial Intelligence Methods Usedmentioning

confidence: 99%

Decade Review (1999-2009): Artificial Intelligence Techniques in Student Modeling

Drigas

Argyri

Vrettaros

2009

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. Artificial Intelligence applications in educational field are getting more and more popular during the last decade (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009) and that is why much relevant research has been conducted. In this paper, we present the most interesting attempts to apply artificial intelligence methods such as fuzzy logic, neural networks, genetic programming and hybrid approaches such as neurofuzzy systems and genetic programming neural networks (GPNN) in student modeling. This latest research trend is a part of every Intelligent Tutoring System and aims at generating and updating a student model in order to modify learning content to fit individual needs or to provide reliable assessment and feedback to student's answers. In this paper, we make a brief presentation of methods used to point out their qualities and then we attempt a navigation to the most representative studies sought in the decade of our interest after classifying them according to the principal aim they attempted to serve.

show abstract

“…The training procedure consists of modeling the structure of the NNs as well as defining the values of their weights. Although a gradient descent algorithm such as back-propagation is most often used as a training algorithm, an evolutionary algorithm such as GP has the potential to produce a global minimum of the weight space and thereby avoid local minima [7].…”

Section: Methodsmentioning

confidence: 99%

The Development of a Self-assessment System for the Learners Answers with the Use of GPNN

Pavlopoulos

Vrettaros

Vouros

et al. 2008

Emerging Technologies and Information Systems for the Knowledge Society

View full text Add to dashboard Cite

Abstract. The goal of this study is the development of an assessment system with the support of a Neural Network approach optimized with the use of Genetic Programming. The data used as training data are real data derived from an educational project. The developed system is proved capable of assessing data from both single select and multiple choice questions in an e-learning environment. The final result is the assessment of the learners' answers through various criteria.

show abstract

Training neural networks: backpropagation vs. genetic algorithms

Cited by 72 publications

References 20 publications

Pengenalan Suara Burung Menggunakan Mel Frequency Cepstrum Coefficient dan Jaringan Syaraf Tiruan pada Sistem Pengusir Hama Burung

Pengenalan Suara Burung Menggunakan Mel Frequency Cepstrum Coefficient dan Jaringan Syaraf Tiruan pada Sistem Pengusir Hama Burung

Decade Review (1999-2009): Artificial Intelligence Techniques in Student Modeling

The Development of a Self-assessment System for the Learners Answers with the Use of GPNN

Contact Info

Product

Resources

About