Claim Missing Document
Check
Articles

Found 7 Documents
Search

Identifikasi Tingkat Kesegaran Ikan Tuna Menggunakan Metode GLCM dan KNN Zulfrianto Yusrin Lamasigi; Serwin -; Husdi -; Yulianti Lasena
Jambura Journal of Electrical and Electronics Engineering Vol 4, No 1 (2022): Januari - Juni 2022
Publisher : Teknik Elektro - Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (574.7 KB) | DOI: 10.37905/jjeee.v4i1.12045

Abstract

Abstrak-Dari potensi perikanan dan kelautan secara Nasional, Provinsi Gorontalo memiliki  potensi perikanan dan kelautan cukup besar yang dapat dikelola  untuk  menunjang pembangunan Gorontalo. Potensi perikanan tangkap Provinsi Gorontalo tidak bisa dipisahkan dari potensi perikanan tangkap yang  berbasis  pada  WPP  (Wilayah Pengelolaan  dan Pemanfaatan)  dan diakui  secara Nasional maupun Internasional. Provinsi Gorontalo merupakan salah satu provinsi penghasil ikan tuna di Indonesia, hasil tangkapan ikan tuna di gorontalo telah diekspor keberbagai negara. Tuna merupakan salah satu komoditi andalan perikanan di Gorontalo yang juga banyak melibatkan nelayan kecil. Penelitianini bertujuan untuk melakukan identifikasi tingkat kesegaran ikan tuna dengan menggukanan metode Gray LevelCo-Occurrence Matrix(GLCM)sebagai metode ektraksi fitur dan K-Nearest Neighbour (K-NN) digunakan sebagai metode klasifikasi. Padapenelitian ini, akan dilakukan 5 kali percobaan berdasarkan sudut 0°, 45°, 90°, 135° dan 180° pada nilai k=1, 3, 5, dan 7. Sementara itu, untuk menghitung tingkat akurasi dari klasifikasi K-NN akan menggunakan confusion matrix. Dari uji coba yang di lakukan dengan menggunakan jumlah data training sebanyak 130 citra dan data testing 45 citra pada semua kelas dan sudut mendapatkan hasil akurasi tertinggi pada sudut 0° dengan nilai k=1 yaitu sebesar 82,28% dan yang paling rendah ada pada sudut 135° dan 180° dengan nilai k=1 yaitu sebesar 53,71%. Berdasarkan hasil akurasi yang didapatkan menunjukkan bahwah GLCM bekerja dengan baik untuk meningkatkan hasil akurasi klasifikasi K-NN yang dibuktikan dengan hasil rata-rata akurasi yang diperoleh mencapai 50%.Abstract-From the national fisheries and marine potential, Gorontalo Province has a large enough fishery and marine potential that can be managed to support the development of Gorontalo. The capture fisheries potential of Gorontalo Province cannot be separated from the potential of capture fisheries based on the WPP (Management and Utilization Area) and is recognized both nationally and internationally. Gorontalo province is one of the tuna-producing provinces in Indonesia, tuna catches in Gorontalo have been exported to various countries. Tuna is one of the mainstay fisheries commodities in Gorontalo which also involves many small fishermen. This study aims to identify the freshness level of tuna by using the Gray Level Co-Occurrence Matrix (GLCM) method as a feature extraction method and K-Nearest Neighbor (K-NN) is a classification method. In this experiment, 5 experiments were conducted based on the angles of 0°, 45°, 90°, 135° and 180° at the values of k=1, 3, 5, and 7. Meanwhile, to calculate the accuracy level of the K-NN classification, we will use a confusion matrix. From the trials carried out using the amount of training data as many as 130 images and testing data 45 images against all classes based on angles 0°, 45°, 90°, 135°, and 180° at the values of k=1, 3, 5, and 7, the highest accuracy obtained is at an angle of 0° with a value of k=1 which is 82.28% and the lowest is at an angle of 135° and 180° with a value of k=1 which is 53.71%. The results of the trials conducted show that GLCM works well to improve the accuracy of the K-NN classification as evidenced by the average accuracy of 50%.
Local Binary Pattern untuk Pengenalan Jenis Daun Tanaman Obat menggunakan K-Nearest Neighbor Zulfrianto Y Lamasigi; Maryam Hasan; Yulianti Lasena
ILKOM Jurnal Ilmiah Vol 12, No 3 (2020)
Publisher : Teknik Informatika Fakultas Ilmu Komputer Univeristas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v12i3.667.208-218

Abstract

Tanaman obat tradisional merupakan jenis tanaman yang mengandung zat aktif yang berfungsi mengobati ataupun mencegah dari berbagai macam penyakit. Oleh karena itu dilakukan penelitian untuk menguji metode Local Binary Pattern untuk ektraksi ciri dari setiap tanaman obat tradisional dan K-Nearest Neighbor pada proses klasifikasi setelah dilakukan ektraksi dari metode Local Binary Pattern. Dari pengujian menggunakan metode Local Binary Pattern dan K-Nearest Neighbor mampu menghasilkan akurasi yang cukup baik yaitu sebesar 96.67%, nilai akurasi tersebut didapat dari perhitungan manual convusion matrix dengan nilai k=9. Sementara itu hasil akurasi terendah ada pada nilai k=1 yaitu 0%. Hasil ektraksi dan klasifikasi dari metode Local Binary Pattern dan K-Nearest Neighbor menggunakan 120 dataset yang dibagi menjadi 90 data training dengan 6 jenis daun tanaman obat yang terdiri dari 15 daun bayam duri, 15 daun binahong, 15 daun jarak, 15 daun afrika, dan 15 daun sirih dengan percobaan 30 data testing.
Perancangan Sitem Uji Kebergunaan Aplikasi Berbasis Web Menggunakan System Usability Scale Muis Nanja; Yulianti Lasena; Hastuti Dalai
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 6 No 4 (2022): OCTOBER-DECEMBER 2022
Publisher : KITA Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v6i4.617

Abstract

Applications are software in the form of programs, both as desktop applications, web and mobile applications that are built to assist in work or various human activities. The number of mobile application competitions between Gojek, Maxim, Grab, Nujek and so on. It is deemed necessary to test the usability value of the application in order to see the level of user convenience and comfort based on the experience of using the application. The results of developing web-based applications using SUS can be seen from the results of tests carried out on application programs built using white box testing and black box testing. The results obtained from the white box test, namely the (CC) test, obtained the value of V(G) = E – N + 2=2 and V(G)= P + 1 = 2, meaning that the results obtained are valid because of the suitability of the region value with the CC calculation results. The results of the black box test have given the results as expected, then the calculation (SUS) obtained an average value of 72 which means the application is in the good category.
PENERAPAN METODE LEAST SQUARE UNTUK PREDIKSI PENJUALAN BRIGHT GAS 5,5 KG Serwin; Yulianti Lasena
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 1 (2023): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i1.133

Abstract

This company sells 5.5 kg of Bright gas which will be distributed to the base every month, experiencing ups and downs. In addition, it also resulted in the inappropriate procurement of 5.5 kg Bright gas. Every month it is not adjusted to monthly sales estimates because it has not used a prediction system. Therefore, there is a sales prediction system of 5.5 kg of bright gas every month, the amount of bright gas is 5.5 kg which will be distributed to the base. The purpose of this research is to find out good accuracy in the Least Square method for the selling process of 5.5 kg bright gas at PT. Togo Jaya Gorontalo. Results achieved With the bright gas prediction system, predictions can be made for the next period and measurement results using MAPE of 0.20%.
K-NEAREST NEIGHBOR MENGGUNAKAN FEATURE SELECTION BACKWARD ELIMINATION UNTUK PREDIKSI JUMLAH PERMINTAAN DARAH PADA PMMI KOTA GORONTALO Yulianti Lasena; Sunarto Taliki; Mohamad Efendi Lasulika; Andi Bode
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 1 (2023): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i1.172

Abstract

The importance of the availability of blood at PMI, it is expected that PMI always maintains the amount of blood supply to meet the need for blood transfusions. Prediction of blood supply is needed to overcome problems related to bloodstock supply at PMI Gorontalo. The application of predicting the number of blood requests with the K-Nearest Neighbor Algorithm can be done to overcome the existing problems. K-NN is a non-parametric algorithm that can be used for classification and regression. The last few decades have been used in prediction cases, but the K-NN algorithm is better if feature selection is applied in selecting features that are not relevant to the model, the feature selection used in this study is Backward Selection. This study aims to determine the error value in predicting the number of requests for blood at the PMI in Gorontalo City. Meanwhile, the purpose of this research is to find the error value of the K-Nearest Neighbor Algorithm and Feature Selection which can be used as a reference for PMI in making policies to make various efforts to maintainbloodstockk in the future.
PENERAPAN METODE LEAST SQUARE UNTUK PREDIKSI PENJUALAN BRIGHT GAS 5,5 KG Serwin; Yulianti Lasena
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 1 (2023): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Indonesia Banda Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i1.133

Abstract

This company sells 5.5 kg of Bright gas which will be distributed to the base every month, experiencing ups and downs. In addition, it also resulted in the inappropriate procurement of 5.5 kg Bright gas. Every month it is not adjusted to monthly sales estimates because it has not used a prediction system. Therefore, there is a sales prediction system of 5.5 kg of bright gas every month, the amount of bright gas is 5.5 kg which will be distributed to the base. The purpose of this research is to find out good accuracy in the Least Square method for the selling process of 5.5 kg bright gas at PT. Togo Jaya Gorontalo. Results achieved With the bright gas prediction system, predictions can be made for the next period and measurement results using MAPE of 0.20%.
K-NEAREST NEIGHBOR MENGGUNAKAN FEATURE SELECTION BACKWARD ELIMINATION UNTUK PREDIKSI JUMLAH PERMINTAAN DARAH PADA PMMI KOTA GORONTALO Yulianti Lasena; Sunarto Taliki; Mohamad Efendi Lasulika; Andi Bode
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 4 No. 1 (2023): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Indonesia Banda Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v4i1.172

Abstract

The importance of the availability of blood at PMI, it is expected that PMI always maintains the amount of blood supply to meet the need for blood transfusions. Prediction of blood supply is needed to overcome problems related to bloodstock supply at PMI Gorontalo. The application of predicting the number of blood requests with the K-Nearest Neighbor Algorithm can be done to overcome the existing problems. K-NN is a non-parametric algorithm that can be used for classification and regression. The last few decades have been used in prediction cases, but the K-NN algorithm is better if feature selection is applied in selecting features that are not relevant to the model, the feature selection used in this study is Backward Selection. This study aims to determine the error value in predicting the number of requests for blood at the PMI in Gorontalo City. Meanwhile, the purpose of this research is to find the error value of the K-Nearest Neighbor Algorithm and Feature Selection which can be used as a reference for PMI in making policies to make various efforts to maintainbloodstockk in the future.