Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Jupiter

Prediksi Cuaca di Kota Palembang Berbasis Supervised Learning Menggunakan Algoritma K-Nearest Neighbour Alvi Syahrini Utami; Dian Palupi Rini; Endang Lestari
JUPITER (Jurnal Penelitian Ilmu dan Teknologi Komputer) Vol 13 No 1 (2021): JUPITER Vol. 13 No. 1 April 2021
Publisher : Teknik Komputer Politeknik Negeri Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

AbstrakPermasalahan cuaca yang dipengaruhi banyak faktor alam menyebabkan kondisi cuaca yang berubah - ubah sehingga kadang sulit diprediksi. Prediksi cuaca yang tepat diperlukan agar masyarakat dan para pengambil kebijakan dapat melakukan antisipasi terhadap hal ini. Banyaknya factor yang mempengaruhi cuaca menyebabkan kesulitan dalam mengklasifikasikan cuaca pada hari tertentu. Locality Sensitive Hashing (LSH) bekerja pada data pelatihan dengan memberikan nilai hash pada tiap vektor yang berisi nilai yang merepresentasikan faktor – faktor yang mempengaruhi cuaca dan melakukan pengklasifikasian cuaca. Untuk selanjutnya algoritma k-Nearest Neighbour (k-NN) yang akan menghitung prediksi terhadap faktor – faktor yang mempengaruhi cuaca pada suatu hari tertentu. Berdasarkan pengujian yang dilakukan, metode k-NN yang dihybrid dengan LSH dapat memprediksi nilai factor – factor yang mempengaruhi cuaca dengan cukup baik dengan nilai Mean Square Error (MSE) sebesar 0,301.  Kata kunci—k-Nearest Neighbour (k-NN), prediksi cuaca, Locality Sensitive Hasihing (LSH) AbstractWeather is influenced by many natural factors causing it to change frequently at any time so that it is sometimes difficult to predict. An accuratet weather prediction is needed so that people and policy makers can anticipate this problem. Many factors that influence the weather cause difficulty in classifying the weather on a particular day. Locality Sensitive Hashing (LSH) works on training data by assigning hash values to a vectors that contain values that represent factors that affect weather and perform weather classification. Furthermore, the k-Nearest Neighbor (k-NN) algorithm will calculate the predictions of the factors that affect the weather on a certain day. Based on the tests carried out, k-NN and LSH in weather prediction has Mean Square Error (MSE) 0,301. Keywords— k-Nearest Neighbou r(k-NN), weather forecasting, Locality Sensitive Hasihing (LSH
Weather Forecasting Based on Supervised Learning Using K-Nearest Neighbour Algorithm Alvi Syahrini Utami; Dian Palupi Rini; Endang Lestari
JUPITER (Jurnal Penelitian Ilmu dan Teknologi Komputer) Vol 13 No 1 (2021): JUPITER Vol. 13 No. 1 April 2021
Publisher : Teknik Komputer Politeknik Negeri Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

AbstractWeather is influenced by many natural factors causing it to change frequently at any time so that it is sometimes difficult to predict. An accurate weather prediction is needed so that people and policy-makers could anticipate this problem. Many factors that influence the weather cause difficulty in classifying the weather on a particular day. Locality Sensitive Hashing (LSH) works on training data by assigning hash values to a vectors that contain values that represent factors that affect weather and perform weather classification. Furthermore, the k-Nearest Neighbor (k-NN) algorithm will calculate the predictions of the factors that affect the weather on a certain day. Based on the tests carried out, k-NN and LSH in weather prediction has Mean Square Error (MSE) 0,301. Keywords— k-Nearest Neighbou r(k-NN), weather forecasting, Locality Sensitive Hasihing (LSH
Pengembangan Perancangan Aplikas Pengembangan Perancangan Aplikasi Akademik Penjadwalan Mata Pelajaran Menggunakan Metode UML dan UCD (User Centered Design) Sanjaya, M. Rudi; Ruskan, Endang Lestari; Putra, Bayu Wijaya; Ismail, Ahmad Arrijal; Maula, Nurly Izzatul
JUPITER (Jurnal Penelitian Ilmu dan Teknologi Komputer) Vol 16 No 2 (2024): Jurnal Penelitian Ilmu dan Teknologi Komputer (JUPITER)
Publisher : Teknik Komputer Politeknik Negeri Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5281/zenodo.13764250

Abstract

Abstract Technological from time to time are increasingly needed, especially in the world of education, where the world of education still uses conventional methods, this method can be replaced by computerized methods or it can be said that technology is in the form of applications, with applications it can make work in schools or Islamic boarding schools easier. What will be built is a website for scheduling subjects in schools or Islamic boarding schools. The aim of this research is to design, build and develop an academic application for scheduling subjects using the UCD (User Centered Design) method. The benefit of this research is that it makes it easier to manage integrated subject scheduling based on website, the research method uses interviews, observation and literature studies, then the design method uses the UML (Unified Modeling Language) method, then for evaluation uses the UCD (User Center Design) method, the results of validity testing from questions L1 to L 10 are found df = n-2 =28-2 =26, r calculated to get L1 is 0.639, L2 = 0.735, L3= 0.511, L4=0.753, L5=0.486, L6=0.822, L7 =0.471, L8= 0.868, L9=0.620 , and L10 = 0.834 where r calculated > r table obtained r table is 0.3172, that validity testing produces valid data, then the total value for reliability testing if the Cronbach's alpha value is > 0.70 is reliable, the total for reliability testing is 0.908, meaning the data is reliable. Keywords—UCD, UML, Validity and Reliability.