Aplikasi prediksi lokasi ikan tuna dengan gaussian process menggunakan metode SVM

Irwansyah, Deananda (2019) Aplikasi prediksi lokasi ikan tuna dengan gaussian process menggunakan metode SVM. Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (7MB)
[img] Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (140kB)
[img] Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (167kB)
[img] Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (162kB)
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (323kB) | Request a copy
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (237kB) | Request a copy
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (766kB) | Request a copy
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (136kB) | Request a copy
[img] Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (204kB)
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (543kB) | Request a copy

Abstract

The condition of fisheries in Indonesia is still not optimal because fishing by fishermen is still done manually. To optimize fishing in Indonesia, the SVM method can be used to determine the prediction of tuna location using chlorophyll concentration and sea surface temperature as the predictor and Gaussian Process to see the relation between fishing vessel location data and prediction factor data. The first step is the acquisition of fishing vessel location data, chlorophyll concentration, and sea surface temperature. Then we cleanse the data which means if there is empty data, it will be filled with data from the previous day. Then all chlorophyll concentration data and sea surface temperature data are combined to one collection of data. After that, SVM prediction model was made using combined data between chlorophyll concentration and sea surface temperature as a basis for making prediction of tuna’s location. After the prediction for tuna’s location is obtained, we will map the tuna’s location in the application. The results of this study are a comparison of the performance of prediction models with 60% sampling, 70% sampling, and 80% sampling, Gaussian Process visualization, performance evaluation of prediction models with RMSE, MSE, and MAE for each model. The best model performance is achieved by 70% sampling with value 74.5% and the best evaluation of prediction model is achieved by 60% sampling with latitude and longitude RMSE values of 9.224925 and 25.198761, MAE latitude and longitude data of 7.309412 and 19.80816, and MSE latitude and longitude data of 85.09924 and 634.9775. Kondisi perikanan di Indonesia saat ini masih belum optimal karena penangkapan ikan oleh nelayan masih dilakukan secara manual. Untuk mengoptimalkan penangkapan ikan di Indonesia dapat menggunakan metode SVM untuk menentukan prediksi lokasi ikan dengan menggunakan konsentrasi klorofil dan suhu pada permukaan laut sebagai faktor prediksinya dan Gaussian Process untuk melihat relasi antara data kapal penangkap lokasi ikan dan data faktor prediksinya. Langkah pertama adalah akuisisi data lokasi kapal penangkap ikan, konsentrasi klorofil, dan suhu pada permukaan laut. Data kemudian dibersihkan, yakni jika ada data yang kosong maka akan diisi dengan data yang lain. Kemudian semua data konsentrasi klorofil dan suhu pada permukaan laut digabungkan menjadi satu. Lalu dibuat model prediksi SVM dengan menggunakan data gabungan konsentrasi klorofil dan suhu pada permukaan laut sebagai dasar untuk mencari prediksi lokasi keberadaan ikan tuna. Setelah prediksi lokasi ikan tuna didapatkan, maka akan divisualisasikan hasil prediksi lokasi ikan tuna pada aplikasi. Hasil penelitian yang didapat adalah perbandingan performa model prediksi dengan 60% sampling, 70% sampling, dan 80% sampling, visualisasi Gaussian Process, evaluasi performa model prediksi dengan RMSE, MSE, dan MAE setiap model. Performa model terbaik dicapai oleh 70% sampling dengan nilai 74.5% dan evaluasi performa model prediksi terbaik dicapai oleh 60% sampling dengan nilai RMSE data latitude dan longitude sebesar 9.224925 dan 25.198761, MAE data latitude dan longitude sebesar 7.309412 dan 19.80816, dan MSE data latitude dan longitude sebesar 85.09924 dan 634.9775.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Irwansyah, DeanandaNIM00000025513deanirwan11@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorMurwantara, I MadeNIDN0302057304UNSPECIFIED
Thesis advisorLukas, SamuelNIDN0331076001UNSPECIFIED
Additional Information: SK 82-16 IRW a
Uncontrolled Keywords: SVM, Gaussian Process, tuna, prediksi, rmse, mse, mae, confusion matrix
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics
Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics
Depositing User: Users 2762 not found.
Date Deposited: 18 Nov 2019 04:01
Last Modified: 13 Sep 2021 08:53
URI: http://repository.uph.edu/id/eprint/5615

Actions (login required)

View Item View Item