Predicting rainfall needs to be done as one of such effort to anticipate water flooding. One of the algorithm that can be used to predict rainfall is random forest. The porpose of the research is to create a model by implementing random forest algorithm. The research method consist of four steps: data collection, data processing, random forest implementation, analysis. Random forest implementation with using training set resulted model that has accurracy 71,09%, precision 0.75, recall 0.85, f-measure 0.79, kappa statistic 0.33, MAE 0.35, RMSE 0.46, ROC Area 0.78. Implementation of random forest algorithm with 10-fold cross validation resulted the output with accurracy 99.45%, precision 0.99, recall 0.99, f-measure 0.99, kappa statistic 0.99, MAE 0,09, RMSE 0.14, ROC area 1.
Clustering is an activity that aims to group a data that has a similarity between one data with another data. K-Means clustering is a non-hierarchical data clustering method that attempts to partition existing data into one or more clusters / groups. In this study clustering was conducted using the K-Means algorithm using data on the achievements of the National Middle School National Examination in 2018 obtained from the official website of the Center for Education and Culture Assessment of the Ministry of Education and Culture of the Republic of Indonesia. The results of the cluster with the K-Means algorithm are obtained for cluster 1 there are 14 provinces, cluster 2 there are 5 provinces, and cluster 3 there are 15 provinces with cluster 1 level is a cluster with a high national test score, cluster 2 is a cluster with a low national test score and a cluster 3 is a cluster with moderate national examination scores. While the results of the evaluation of the K-Means algorithm with the number of clusters 3 produce an evaluation value of Connectivity 11,916, Dunn 0.246 and Silhouette 0.464.
<p><strong>Abstrak:</strong><em> User Interface</em> adalah tampilan dari sebuah produk yang berfungsi menjembatani sistem dengan pengguna atau <em>user</em>, dimana tampilan UI bisa berupa warna, bentuk serta tulisan yang menarik pada aplikasi <em>mobile</em>. Dengan kurangnya persiapan dan rancangan yang belum matang, maka pada aplikasi <em>mobile </em>tersebut kurang berjalan maksimal dan mengakibatkan pengguna ingin berpindah ke aplikasi yang lain. Tujuan penelitian ini yaitu analisis tingkat <em>user interface</em> pada <em>aplikasi android Course Online </em>menggunakan <em>usability</em><em> testing</em>. Pada penelitian ini dilakukan dengan menggunakan <em>usability</em><em> testing</em> yaitu pengujian <em>usability</em><em> </em>menggunakan metode<em> </em>SUS dengan mengukur kepuasan pengguna dengan 10 pertanyaan secara<em> online</em>. Penelitian ini data informasi yang didapat secara <em>online</em> ini adalah pengguna aplikasi <em>Course Online</em> android berbasis internet ini akan memberikan langsung efek kepada pengguna terhadap <em>aplikasi android Course Online </em>yang dilangsungkan. Sampel data yang diambil dalam penelitian ini adalah 30 orang mahasiswa yang mencoba aplikasi ini. Dalam teknik ini <em>analisis data</em> penelitian informasi yang digunakan merupakan <em>analisis deskriptif</em> dengan persentase data, kemudian dideskripsikan untuk mengukur tingkat kemudahan penggunaan dalam <em>aplikasi android Course Online. </em>Hasil dari penelitian ini skor yang di dapat melalui kuesioner yang disebarkan secara <em>online</em> ini mendapatkan skor SUS 78,3. Pada sisi <em>acceptability ranges </em>menempati level marginal <em>high</em>, pada sisi adjektif rating berada pada posisi OK, dan terakhir pada sisi grade <em>scale</em> menempati grade B.</p><p><strong> </strong></p><p><strong><em>Kata kunci</em></strong><em>: </em>Analisis, <em>Usability</em><em> Testing, User Interface</em><strong><em></em></strong></p>
Electronic money is a cashless payment instrument whose money is stored in media server or chip that can be moved for the benefit of payment transactions or fund transfers. In Indonesia, there are already many electronic money products, one of which is OVO. OVO is very popular with the people of Indonesia because it offers many promos such as discounts and cashback. But over time, that much promotion is detrimental to OVO shareholders, so the portion of promo given by OVO to its customers is finally reduced. That incident caused many pros and cons opinions about OVO, one of them is on social media Twitter. Sentiment analysis can be used as a solution to process the opinions of OVO customers on Twitter. This study aims to classify the customer opinions on OVO services into positive and negative classes. This study uses the Support Vector Machine algorithm with 3852 data taken from Twitter with keyword @ovo_id using web scraping techniques. The dataset divided into two classes, 2034 positive and 1818 negative sentiment data. The classification process is carried out with four splitting data scenarios, with 60:40, 70:30, 80:20, 90:10 data ratio and with four kernel such as linear, rbf, sigomid, and polynomial. The final results show that the greatest accuracy value obtained by linear kernel with 90:10 data ratio which gets an accuracy value of 98.7%.
Algoritma C4.5 merupakan algoritma klasifikasi yang memungkinkan bisa diterapkan untuk studi kasus prediksi potensi kebakaran hutan. Untuk mengetahui penerapan algoritma C4.5 pada prediksi kebakaran hutan, perlu dilakukan penelitian terkait hal tersebut. Metodologi yang digunakan adalah Knowledge Discovery in Database (KDD). Tahap dari KDD terdiri dari pengumpulan dan pemilihan data, pemrosesan data, transformasi data, pengolahan data dengan algoritma C4.5 dan terakhir adalah interpretasi serta evaluasi pengetahuan. Percentage split, Cross validation, Use Training Set digunakan sebagai teknik pembagian data training dan testing dengan skenario pesentase dan dipilih model terbaik. Indikator evaluasi yang digunakan adalah akurasi. Penelitian menghasilkan kesimpulan bahwa C4.5 dengan percentage split 80%data training dan 20% data testing menghasilkan akurasi tertinggi yaitu 89,7859%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.