Salma Rita
Universitas Muhammadiyah Sukabumi

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

PREDIKSI LUAS LAHAN SAWAH DENGAN PROGRAM MATLAB MENGGUNAKAN JARINGAN SYARAF TIRUAN Salma Rita
JRIS : Jurnal Rekayasa Informasi Swadharma Vol 3, No 1 (2023): JURNAL JRIS EDISI JANUARI 2023
Publisher : Institut Teknologi dan Bisnis (ITB) Swadharma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56486/jris.vol3no1.255

Abstract

Indonesia is a country that makes rice as a staple food. This means that the more rice consumption is offset by an increase in population, the narrower the area of paddy fields due to the shift in function of paddy fields. It is very important to monitor land use in an area to avoid the problem of narrowing the rice field area caused by misuse of land which impacts on rice production. This study focuses on the objective of predicting the area of paddy fields, where the authors chose Cikampek District as research material. In addition, the authors use artificial neural networks as a method in their predictions. The program with matlab was chosen to display the predicted results where it was used as an indicator of the achievement of this research. This research method uses data from the Badan Pusat Statistik (BPS) for the Cikampek District area. The data used is from 2010 to 2020. The use of artificial neural networks with backpropagation has succeeded well in predicting the area of paddy fields where the parameters used in the design of artificial neural networks that must be considered include the number of input neurons, the hidden layer and the output layer. The results of the implementation with matlab obtained a graphic output from neural network training with MSE of 0.030922 and the test produced a regression of 0.9747.Indonesia merupakan negara yang menjadikan beras sebagai salah satu makanan pokok. Hal ini, semakin banyak konsumsi beras yang diimbangi dengan pertambahan jumlah penduduk, maka luas lahan persawahan semakin menyempit dikarenakan beralih fungsinya lahan sawah. Sangat penting untuk memantau penggunaan lahan di suatu wilayah untuk menghindari masalah penyempitan kawasan persawahan yang disebabkan oleh penyalahgunaan lahan yang berimbas pada produksi beras. Penelitian ini berfokus pada tujuan untuk prediksi luas lahan sawah, dimana penulis memilih Kecamatan Cikampek sebagai bahan penelitian. Selain itu, penulis menggunakan jaringan syaraf tiruan sebagai metode dalam prediksinya. Program dengan matlab dipilih untuk menampilkan hasil prediksi dimana hal tersebut digunakan sebagai indikator capaian dari penelitian ini. Metode penelitian ini menggunakan data dari Badan Pusat Statistik (BPS) wilayah Kecamatan Cikampek. Data yang digunakan adalah dari tahun 2010 sampai 2020. Penggunaan jaringan syaraf tiruan dengan backpropagation berhasil dengan baik dalam memprediksi luas lahan sawah dimana parameter yang digunakan dalam perancangan jaringan syaraf tiruan yang harus diperhatikan diantaranya jumlah neuron yang dimasukan, layer tersembunyi dan layer keluaran. Hasil implementasi dengan matlab diperoleh grafik keluaran dari neural network training dengan MSE sebesar 0,030922 dan pada pengujian menghasilkan regresi sebesar 0,9747.
Using Support Vector Machine for Sentiment Analysis of Truecaller and Getcontact App Reviews Salma Rita; Didik Indrayana; Agung Pambudi
Bit (Fakultas Teknologi Informasi Universitas Budi Luhur) Vol 20, No 2 (2023): SEPTEMBER 2023
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/bit.v20i2.2493

Abstract

Spam calls are any calls made without the consent of the recipient and for any reason. These calls can originate from marketing, advertising, notifications, or fraud. The average Indonesian receives 14 spam calls per day. Only half of it comes from contact book numbers. According to the Google Play Store, the Truecaller and Getcontact apps offer a number of advantages as they each help identify callers and prevent spam. However, in this regard, spam call blocking software has a number of drawbacks, including identifying spam calls incorrectly and blocking useless calls. Sentiment analysis can help users choose applications that suit their needs and in this study intends to analyze reviews on the sentiments of the two applications namely Truecaller: Caller Id and Getcontact by analyzing effectiveness based on the application reviews Support Vector Machine classification algorithm which contains basic guidelines for maximizing hyperplane boundaries that separate the two datasets, are used in the classification process of this study. The results showed that the Truecaller application at 10-fold cross validation had an average accuracy of 88.20% and the Getcontact application had an average accuracy of 87.90%. Meanwhile, the sentiment aspect of the Truecaller application has an average accuracy value of 60.20%, while the Getcontact application has an average accuracy of 63.30%.