Ujian - National Autism Resource Centre (NARC)

2019

Abdullah, A A; Rijal, S; Sengkang, S R

Penilaian ke atas Algoritma Pembelajaran Mesin untuk Klasifikasi Gangguan Spektrum Autisme (ASD) Persidangan

1372 (1), Institut Penerbitan Fizik, 2019, ISSN: 17426588, (dipetik oleh 0).

Abstrak | Pautan | BibTeX | Tag: Gangguan Spektrum Autisme, Penilaian Tingkah Laku, Kejuruteraan Bioperubatan, Pemetaan Otak, Pengelasan (maklumat), Pokok Keputusan, Penyakit, Pengekstrakan Ciri, Kaedah Pemilihan Ciri, Pengesahan Silang K Lipat, Belajar, Operator Pengecutan dan Pemilihan Mutlak Paling Kurang, Anggaran Kuasa Dua Terkecil, Regresi Logistik, Pembelajaran Mesin, Kaedah Pembelajaran Mesin, Pengimejan Resonans Magnetik, Carian Jiran Terdekat, Analisis regresi, Pembelajaran yang diselia, Pembelajaran Mesin Diawasi

@ persidangan{Abdullah2019,
tajuk = {Penilaian ke atas Algoritma Pembelajaran Mesin untuk Klasifikasi Gangguan Spektrum Autisme (ASD)},
pengarang = {A A Abdullah and S Rijal and S R Dash},
penyunting = {Rahim Mustafa Zaaba Norali Noor S B A N B S K A N B A B M Fook C.Y. Yazid H.B.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076493636&doi=10.1088%2f1742-6596%2f1372%2f1%2f012052&rakan kongsi = 40&md5=2ec1bd9f6cf1e3afe965cc9e3792f536},
doi = {10.1088/1742-6596/1372/1/012052},
terbitan = {17426588},
tahun = {2019},
tarikh = {2019-01-01},
jurnal = {Journal of Physics: Conference Series},
isi padu = {1372},
nombor = {1},
penerbit = {Institut Penerbitan Fizik},
abstrak = {Gangguan Spektrum Autisme (ASD) dicirikan oleh kelewatan dalam pembangunan interaksi sosial, tingkah laku berulang dan minat yang sempit, yang biasanya didiagnosis dengan alat diagnostik standard seperti Jadual Pemerhatian Diagnostik Autisme (Remaja) dan Temuduga Diagnostik Autisme-Disemak (ADIR-R). Kerja sebelumnya telah melaksanakan kaedah pembelajaran mesin untuk klasifikasi ASD, namun mereka menggunakan jenis set data yang berbeza seperti imej otak untuk MRI dan EEG, gen risiko dalam profil genetik dan penilaian tingkah laku berdasarkan ADOS dan ADI-R. Di sini percubaan menggunakan Soalan Spektrum Autisme (AQ) untuk membina model yang mempunyai potensi yang lebih tinggi untuk mengklasifikasikan ASD telah dibangunkan. Dalam penyelidikan ini, Chi-square dan Operator Pengecutan dan Pemilihan Mutlak Terkecil (LASSO) telah dipilih sebagai kaedah pemilihan ciri untuk memilih ciri yang paling penting 3 algoritma pembelajaran mesin yang diselia, iaitu Hutan Rawak, Regresi Logistik dan K-Nearest Neighbours dengan pengesahan silang K-fold. Prestasi dinilai di mana keputusan Regresi Logistik mendapat ketepatan tertinggi dengan 97.541% menggunakan model dengan 13 ciri yang dipilih berdasarkan kaedah pemilihan Khi kuasa dua. © 2019 IOP Publishing Ltd. Hak cipta terpelihara.},
nota = {dipetik oleh 0},
kata kunci = {Gangguan Spektrum Autisme, Penilaian Tingkah Laku, Kejuruteraan Bioperubatan, Pemetaan Otak, Pengelasan (maklumat), Pokok Keputusan, Penyakit, Pengekstrakan Ciri, Kaedah Pemilihan Ciri, Pengesahan Silang K Lipat, Belajar, Operator Pengecutan dan Pemilihan Mutlak Paling Kurang, Anggaran Kuasa Dua Terkecil, Regresi Logistik, Pembelajaran Mesin, Kaedah Pembelajaran Mesin, Pengimejan Resonans Magnetik, Carian Jiran Terdekat, Analisis regresi, Pembelajaran yang diselia, Pembelajaran Mesin Diawasi},
pubstate = {diterbitkan},
tppubtype = {persidangan}
}

Tutup

2018

Hariharan, M; Sindhu, R; Vijean, V; Yazid, H; Nadarajaw, T; Yaacob, S; Polat, K

Improved binary dragonfly optimization algorithm and wavelet packet based non-linear features for infant cry classification Artikel Jurnal

Computer Methods and Programs in Biomedicine, 155 , hlm. 39-51, 2018, ISSN: 01692607, (dipetik oleh 21).

Abstrak | Pautan | BibTeX | Tag: Accidents, Algoritma, Artikel, Artificial Neural Network, Asphyxia, Binary Dragonfly Optimization Aalgorithm, Pengelasan (maklumat), Classification Algorithm, Pengelas, Coding, Computer-Assisted, Constants and Coefficients, Crying, Database Systems, Databases, Deafness, Diagnosis, Energy, Entropy, Pengekstrakan, Extreme Learning Machine, Factual, Factual Database, Pengekstrakan Ciri, Kaedah Pemilihan Ciri, Fuzzy System, Hearing Impairment, Manusia, Kelaparan, Bayi, Infant Cry, Infant Cry Classifications, Jaundice, Kernel Method, Belajar, Linear Predictive Coding, Pembelajaran Mesin, Mathematical Transformations, Mel Frequency Cepstral Coefficient, Pekali Cepstral Frekuensi Mel, Multi-Class Classification, Rangkaian Neural, Nonlinear Dynamics, Nonlinear System, Optimization, Pain, Patofisiologi, Prematurity, Kebolehulangan, Kebolehulangan Keputusan, Pemprosesan isyarat, Pengenalan suara, Wavelet Analysis, Wavelet Packet, Paket Wavelet Berubah

@artikel{Hariharan201839,
tajuk = {Improved binary dragonfly optimization algorithm and wavelet packet based non-linear features for infant cry classification},
pengarang = {M Hariharan and R Sindhu and V Vijean and H Yazid and T Nadarajaw and S Yaacob and K Polat},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85036611215&doi=10.1016%2fj.cmpb.2017.11.021&rakan kongsi = 40&md5=1f3b17817b00f07cadad6eb61c0f4bf9},
doi = {10.1016/j.cmpb.2017.11.021},
terbitan = {01692607},
tahun = {2018},
tarikh = {2018-01-01},
jurnal = {Computer Methods and Programs in Biomedicine},
isi padu = {155},
halaman = {39-51},
penerbit = {Elsevier Ireland Ltd},
abstrak = {Background and objective Infant cry signal carries several levels of information about the reason for crying (hunger, pain, sleepiness and discomfort) or the pathological status (asphyxia, deaf, jaundice, premature condition and autism, etc.) of an infant and therefore suited for early diagnosis. Dalam kerja ini, combination of wavelet packet based features and Improved Binary Dragonfly Optimization based feature selection method was proposed to classify the different types of infant cry signals. Methods Cry signals from 2 different databases were utilized. First database contains 507 cry samples of normal (N), 340 cry samples of asphyxia (A), 879 cry samples of deaf (D), 350 cry samples of hungry (H) dan 192 cry samples of pain (P). Second database contains 513 cry samples of jaundice (J), 531 samples of premature (Prem) dan 45 samples of normal (N). Wavelet packet transform based energy and non-linear entropies (496 ciri-ciri), Linear Predictive Coding (LPC) based cepstral features (56 ciri-ciri), Mel-frequency Cepstral Coefficients (MFCCs) were extracted (16 ciri-ciri). The combined feature set consists of 568 ciri-ciri. To overcome the curse of dimensionality issue, improved binary dragonfly optimization algorithm (IBDFO) was proposed to select the most salient attributes or features. Akhirnya, Extreme Learning Machine (ELM) kernel classifier was used to classify the different types of infant cry signals using all the features and highly informative features as well. Results Several experiments of two-class and multi-class classification of cry signals were conducted. In binary or two-class experiments, maximum accuracy of 90.18% for H Vs P, 100% for A Vs N, 100% for D Vs N and 97.61% J Vs Prem was achieved using the features selected (only 204 features out of 568) by IBDFO. For the classification of multiple cry signals (multi-class problem), the selected features could differentiate between three classes (N, A & D) with the accuracy of 100% and seven classes with the accuracy of 97.62%. Conclusion The experimental results indicated that the proposed combination of feature extraction and selection method offers suitable classification accuracy and may be employed to detect the subtle changes in the cry signals. © 2017 Elsevier B.V.},
nota = {dipetik oleh 21},
kata kunci = {Accidents, Algoritma, Artikel, Artificial Neural Network, Asphyxia, Binary Dragonfly Optimization Aalgorithm, Pengelasan (maklumat), Classification Algorithm, Pengelas, Coding, Computer-Assisted, Constants and Coefficients, Crying, Database Systems, Databases, Deafness, Diagnosis, Energy, Entropy, Pengekstrakan, Extreme Learning Machine, Factual, Factual Database, Pengekstrakan Ciri, Kaedah Pemilihan Ciri, Fuzzy System, Hearing Impairment, Manusia, Kelaparan, Bayi, Infant Cry, Infant Cry Classifications, Jaundice, Kernel Method, Belajar, Linear Predictive Coding, Pembelajaran Mesin, Mathematical Transformations, Mel Frequency Cepstral Coefficient, Pekali Cepstral Frekuensi Mel, Multi-Class Classification, Rangkaian Neural, Nonlinear Dynamics, Nonlinear System, Optimization, Pain, Patofisiologi, Prematurity, Kebolehulangan, Kebolehulangan Keputusan, Pemprosesan isyarat, Pengenalan suara, Wavelet Analysis, Wavelet Packet, Paket Wavelet Berubah},
pubstate = {diterbitkan},
tppubtype = {artikel}
}

Tutup

Background and objective Infant cry signal carries several levels of information about the reason for crying (hunger, pain, sleepiness and discomfort) or the pathological status (asphyxia, deaf, jaundice, premature condition and autism, etc.) of an infant and therefore suited for early diagnosis. Dalam kerja ini, combination of wavelet packet based features and Improved Binary Dragonfly Optimization based feature selection method was proposed to classify the different types of infant cry signals. Methods Cry signals from 2 different databases were utilized. First database contains 507 cry samples of normal (N), 340 cry samples of asphyxia (A), 879 cry samples of deaf (D), 350 cry samples of hungry (H) dan 192 cry samples of pain (P). Second database contains 513 cry samples of jaundice (J), 531 samples of premature (Prem) dan 45 samples of normal (N). Wavelet packet transform based energy and non-linear entropies (496 ciri-ciri), Linear Predictive Coding (LPC) based cepstral features (56 ciri-ciri), Mel-frequency Cepstral Coefficients (MFCCs) were extracted (16 ciri-ciri). The combined feature set consists of 568 ciri-ciri. To overcome the curse of dimensionality issue, improved binary dragonfly optimization algorithm (IBDFO) was proposed to select the most salient attributes or features. Akhirnya, Extreme Learning Machine (ELM) kernel classifier was used to classify the different types of infant cry signals using all the features and highly informative features as well. Results Several experiments of two-class and multi-class classification of cry signals were conducted. In binary or two-class experiments, maximum accuracy of 90.18% for H Vs P, 100% for A Vs N, 100% for D Vs N and 97.61% J Vs Prem was achieved using the features selected (only 204 features out of 568) by IBDFO. For the classification of multiple cry signals (multi-class problem), the selected features could differentiate between three classes (N, A & D) with the accuracy of 100% and seven classes with the accuracy of 97.62%. Conclusion The experimental results indicated that the proposed combination of feature extraction and selection method offers suitable classification accuracy and may be employed to detect the subtle changes in the cry signals. © 2017 Elsevier B.V.

Tutup