Garuda - Garba Rujukan Digital

Handwritten Arabic Numeral Character Recognition Using Multi Kernel Support Vector Machine Athoillah, Muhammad; Putri, Rani Kurnia
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol 4, No 2, May 2019
Publisher : Universitas Muhammadiyah Malang

Handwritten recognition is how computer can identify a handwritten character or letter from a document, an image or another source. Recently, many devices provide a feature using handwritten as an input such as laptops, smartphones, and others, affecting handwritten recognition abilities become important thing. As the mother tongue of Muslims, and the only language used in holy book Al Qur’an, therefore recognizing in arabic character is a challenging task. The outcome of that recognizing system has to be quite accurate, the results of the process will impact on the entire process of understanding the Qur’an lesson. Basically handwritten recognition problem is part of classification problem and one of the best algorithm to solve it is Support Vector Machine (SVM). By finding a best separate line and two other support lines between input space data in process of training, SVM can provide the better result than other classify algorithm. Although SVM can solve the classify problem well, SVM must be modified with kernel learning method to be able to classify nonlinear data. However, determining the best kernel for every classification problem is quite difficult. Therefore, some technique have been developed, one of them is Multi Kernel Learning (MKL). This technique works by combining some kernel function to be one kernel with an equation. This framework built an application to recognize handwritten arabic numeral character using SVM algorithm that modified with Kernel Learning Method. The result shows that the application can recognize data well with average value of Accuracy is 84,37%

KLASIFIKASI KENDARAAN BERMOTOR DENGAN MULTI KERNEL SUPPORT VECTOR MACHINE Athoillah, Muhammad
Buana Matematika : Jurnal Ilmiah Matematika dan Pendidikan Matematika Vol 8 No 1 (2018)
Publisher : Universitas PGRI Adi Buana Surabaya

Since October 2017, Electronic automatic payments have been applied in all Indonesian toll roads. This payment system needes an extra system which is able to distinguish the types of motor vehicles that will enter the toll road due to the regulation itself which allow four-wheel or more vehicle to pass the toll road. This automatic system can be built by a classification algorithm, and one of the best is Support Vector Machine (SVM), in order to be able to classify non-linear separable data, SVM must be modified by giving kernel function on it. Furthermore, determining the approriate kernel for every classification problem is a fundamental step, and thatÃ¢â‚¬â„¢s obviously not easy, to solve that problem the researchers found a method that can make this kernel function more flexible, this method called Multi Kernel Learning (MKL). Main idea of this method is formulating some kernel function to be one kernel function. This framework is built an automatic system to classify the types of motor vehicles using Support Vector Machine modified using Multi Kernel Learning method. The experimental result shows that the system has a good average value of accuracy that is 84.60%, the average value of precision is 84.95% and also average value of recall is 84.60%. Keyword: Kendaraan Bermotor, Klasifikasi, Multi Kernel, Support Vector Machine

K-Nearest Neighbor untuk Pengenalan karakter tulisan arab Muhammad Athoillah
Jurnal Matematika MANTIK Vol. 5 No. 2 (2019): Mathematics and Applied Mathematics
Publisher : Mathematics Department, Faculty of Science and Technology, UIN Sunan Ampel Surabaya

Handwritten text recognition is the ability of a system to recognize human handwritten and convert it into digital text. Handwritten text recognition is a form of classification problem, so a classification algorithm such as Nearest Neighbor (NN) is needed to solve it. NN algorithms is a simple algorithm yet provide a good result. In contrast with other algorithms that usually determined by some hypothesis class, NN Algorithm finds out a label on any test point without searching for a predictor within some predefined class of functions. Arabic is one of the most important languages in the world. Recognizing Arabic character is very interesting research, not only it is a primary language that used in Islam but also because the number of this research is still far behind the number of recognizing handwritten Latin or Chinese research. Due to that's the background, this framework built a system to recognize handwritten Arabic Character from an image dataset using the NN algorithm. The result showed that the proposed method could recognize the characters very well confirmed by its average of precision, recall and accuracy.

STUDY COMPARISON OF SVM-, K-NN- AND BACKPROPAGATION-BASED CLASSIFIER FOR IMAGE RETRIEVAL Muhammad Athoillah; M. Isa Irawan; Elly Matul Imah
Jurnal Ilmu Komputer dan Informasi Vol 8, No 1 (2015): Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information)
Publisher : Faculty of Computer Science - Universitas Indonesia

Classification is a method for compiling data systematically according to the rules that have been set previously. In recent years classification method has been proven to help many peopleâ€™s work, such as image classification, medical biology, traffic light, text classification etc. There are many methods to solve classification problem. This variation method makes the researchers find it difficult to determine which method is best for a problem. This framework is aimed to compare the ability of classification methods, such as Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), and Backpropagation, especially in study cases of image retrieval with five category of image dataset. The result shows that K-NN has the best average result in accuracy with 82%. It is also the fastest in average computation time with 17,99 second during retrieve session for all categories class. The Backpropagation, however, is the slowest among three of them. In average it needed 883 second for training session and 41,7 second for retrieve session.

Pelatihan Pengujian Hipotesis Statistika Dasar dengan Software R Gangga Anuraga; Artanti Indrasetianingsih; Muhammad Athoillah
BUDIMAS : JURNAL PENGABDIAN MASYARAKAT Vol 3, No 2 (2021): BUDIMAS : VOL. 03 NO. 02, 2021
Publisher : LPPM ITB AAS Indonesia Surakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29040/budimas.v3i2.2412

Statistik adalah bidang yang sangat interdisipliner, penelitian dalam statistik kini diterapkan di hampir semua bidang ilmiah. Uji Hipotesis adalah salah satu cabang llmu Statistika Inferensial yang digunakan untuk menguji kebenaran atas suatu pernyataan secara statistik serta menarik kesimpulan akan diterima atau ditolaknya pernyataan tersebut. Dalam berbagai macam pengujian, tentunya para peneliti ingin membuktikan bahwa asumsi atau pendapat yang ia percayai tersebut benar atau tidak. Uji hipotesis dapat membantu dalam membuktikan suatu hal apakah benar-benar fakta ataukah hanya sekadar teori belaka. Atas dasar tersebut, tim dosen Program Studi Statistika, Fakultas Sains dan Teknologi (FST) melalui Program Pengabdian Kepada Masyarakat (PKM) memberikan pelatihan online (Webinar) tentang Pengujian Hipotesis Statistika Dasar dengan Software R kepada dosen, mahasiswa serta masyarakat secara umum. Keseluruh kegiatan pengabdian telah terlaksana dengan baik. Keberhasilan kegiatan ini diukur dari antusiasme peserta selama kegiatan sekaligus hasil evaluasi pre-test yang rata-rata nilainya 57,78 meningkat menjadi 77,74 dari hasil post-test setelah kegiatan.

Analisis Dampak Covid-19 Terhadap Indeks Harga Konsumen dengan K-Means dan Regresi Berganda Firli Azizah; Muhammad Athoillah
Indonesian Journal of Applied Statistics Vol 4, No 1 (2021)
Publisher : Universitas Sebelas Maret

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.13057/ijas.v4i1.46329

The Indonesian economy during the global pandemic entered the brink of economic recession. This problem occurs because the state of public consumption has decreased due to the limited space for community movement and sluggish economic activities due to preventing the transmission of Covid-19. This affects the decline in public consumption in economic activities. In this case, it can be seen from the statistical news published by the official website of the Badan Pusat Statistik (BPS) which reports that the inflation rate in the previous months was around 0.10%, while in April 2020 it decreased by 0.08%. Based on these, a K-means grouping study was conducted by dividing the cluster into 3 parts and modeling using multiple regression methods. In this study, the variable used was the price index. The results of the K-means cluster analysis with the division of 3 clusters, namely cluster 3 (high CPI cluster) consisting of 66 cities, cluster 1 (moderate CPI cluster) consisting of 2 cities, and cluster 2 (low CPI cluster) consisting of 22 cities. Furthermore, the multiple regression results obtained 12 variables that have a significant effect on the Consumer Price Index (CPI). The results of regression modeling are the highest coefficient is food at 0.236 and the lowest coefficients are cigarettes and tobacco at 0.008. Therefore can be concluded that the grouping of the CPI indicator obtained 75% of cities with high index prices, especially in big cities such that economic activity, in general, was still consumptive during the pandemic and multiple regression modeling resulted from 37 indicator variables, only 12 indicator variables had a significant effect on the CPI.Keywords: k-means, CPI, multiple regression, and price index

KLASIFIKASI KENDARAAN BERMOTOR DENGAN MULTI KERNEL SUPPORT VECTOR MACHINE Muhammad Athoillah
Buana Matematika : Jurnal Ilmiah Matematika dan Pendidikan Matematika Vol 8 No 1 (2018)
Publisher : Universitas PGRI Adi Buana Surabaya

Since October 2017, Electronic automatic payments have been applied in all Indonesian toll roads. This payment system needes an extra system which is able to distinguish the types of motor vehicles that will enter the toll road due to the regulation itself which allow four-wheel or more vehicle to pass the toll road. This automatic system can be built by a classification algorithm, and one of the best is Support Vector Machine (SVM), in order to be able to classify non-linear separable data, SVM must be modified by giving kernel function on it. Furthermore, determining the approriate kernel for every classification problem is a fundamental step, and thatÃ¢â‚¬â„¢s obviously not easy, to solve that problem the researchers found a method that can make this kernel function more flexible, this method called Multi Kernel Learning (MKL). Main idea of this method is formulating some kernel function to be one kernel function. This framework is built an automatic system to classify the types of motor vehicles using Support Vector Machine modified using Multi Kernel Learning method. The experimental result shows that the system has a good average value of accuracy that is 84.60%, the average value of precision is 84.95% and also average value of recall is 84.60%. Keyword: Kendaraan Bermotor, Klasifikasi, Multi Kernel, Support Vector Machine

Handwritten Arabic Numeral Character Recognition Using Multi Kernel Support Vector Machine Muhammad Athoillah; Rani Kurnia Putri
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol 4, No 2, May 2019
Publisher : Universitas Muhammadiyah Malang

Handwritten recognition is how computer can identify a handwritten character or letter from a document, an image or another source. Recently, many devices provide a feature using handwritten as an input such as laptops, smartphones, and others, affecting handwritten recognition abilities become important thing. As the mother tongue of Muslims, and the only language used in holy book Al Qur’an, therefore recognizing in arabic character is a challenging task. The outcome of that recognizing system has to be quite accurate, the results of the process will impact on the entire process of understanding the Qur’an lesson. Basically handwritten recognition problem is part of classification problem and one of the best algorithm to solve it is Support Vector Machine (SVM). By finding a best separate line and two other support lines between input space data in process of training, SVM can provide the better result than other classify algorithm. Although SVM can solve the classify problem well, SVM must be modified with kernel learning method to be able to classify nonlinear data. However, determining the best kernel for every classification problem is quite difficult. Therefore, some technique have been developed, one of them is Multi Kernel Learning (MKL). This technique works by combining some kernel function to be one kernel with an equation. This framework built an application to recognize handwritten arabic numeral character using SVM algorithm that modified with Kernel Learning Method. The result shows that the application can recognize data well with average value of Accuracy is 84,37%

PELATIHAN PENGGUNAAN SOFTWARE R UNTUK MENGUJI PERBANDINGAN BERGANDA DAN ASUMSI RESIDUAL PADA RANCANGAN PERCOBAAN Elvira Mustikawati Putri Hermanto; Muhammad Athoillah; Wanda Nur Hamidah; Dimas Pramana Putra
J-ABDI: Jurnal Pengabdian kepada Masyarakat Vol. 1 No. 4: September 2021
Publisher : Bajang Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53625/jabdi.v1i4.238

Sebagai perwujudan bakti negri, Program Studi S1 Statistika Unipa mengadakan kegiatan pengabdian kepada masyarakat melalui pelatihan uji perbandingan berganda dan asumsi residual menggunakan software R. Pemilihan topik didasarkan atas hasil observasi yang menunjukkan masih banyak mahasiswa, peneliti muda maupun masyarakat yang berkecimpung dalam bidang sains data belum mengenal metode tersebut, sedangkan uji perbandingan ganda aspek penting yang perlu dilakukan terutama saat hasil analisis variansi hanya menentukan adanya perbedaan antar populasi tetapi tidak bisa mengetahui populasi mana saja yang berbeda, sedangkan Uji Asumsi Residual diperlukan sebagai bukti kevalidan dai model yang sedang diteliti, karena hasil dari uji tersebut dapat membuktikan bahwa estimasi dari parameternya tidak bias. Kesuksesan kegiatan pelatihan ini dibuktikan dari dua aspek, yaitu hasil umpan balik positif dari peserta tentang terselenggaranya pelatihan ini, serta dari hasil evaluasi pre-test dan post test yang menunjukkan peningkatan presentasi hasil evaluasi dari 61,60% pada saat pre-test menjadi 71,33% pada saat post test

Modified Multi-Kernel Support Vector Machine for Mask Detection Muhammad Athoillah; Evita Purnaningrum; Rani Kurnia Putri
CommIT (Communication and Information Technology) Journal Vol. 16 No. 2 (2022): CommIT Journal
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/commit.v16i2.7873

Indonesia is one of the countries most affected by the Coronavirus pandemic with millions confirm cases. Hence, the government has increased strict procedures for using face masks in public areas. For this reason, the detection of people wearing face masks in public areas is needed. Face mask detection is a part of the classification problem. Thus Support Vector Machine (SVM) can be implemented. SVM is still known as one of the most powerful and efficient classification algorithms. The research aims to build an automatic face mask detector using SVM. However, it needs to modify it first because it only can classify linear data. The modification is made by adding kernel functions, and a Multi-kernel approach is chosen. The proposed method is applied by combining various kernels into one kernel equation. The dataset used in the research is a face mask image obtained from Github. The data are public datasets consisting of faces with and without masks. The results present that the proposed method provides good performance. It is proven by the average value. The values are 83.67% for sensitivity, 82.40% for specificity, 82.00% for precision, 82.93% for accuracy, and 82.77% for F1-score. These values are better than other experiments using single kernel SVM with the same process and dataset.

Title

Found 35 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 35 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 35 Documents
Search