CD Skripsi

Prediksi Curah Hujan Dengan Long Short Term Memory Dan Isolation Forest Di Kota Pekanbaru

---

Rainfall is an important factor in climate analysis and environmental planning, especially in Pekanbaru City which has variable rainfall patterns. This research aims to predict rainfall using the Long Short Term Memory (LSTM) and Isolation Forest methods. The data used to predict rainfall in the Long Short Term Memory model is 3,640 and the data used to detect anomalies in Isolation Forest is 727 from the LSTM prediction results. The evaluation results show that the LSTM model is able to provide optimal predictions, indicated by the Mean Absolute Error (MAE) value of 0.046 and Root Mean Square Error (RMSE) of 0.094 and a correlation value of 0.205, meaning that the relationship between predicted and actual data can still be improved to obtain better performance. In addition, the Isolation Forest model successfully identifies anomalies that have the potential for rainfall, seen from the P-Value value of less than 0.05, namely 0.000. The results of this study can be the basis for authorities to make better decisions related to disaster mitigation affected by weather changes in Pekanbaru City.


Keywords: Isolation Forest, Long Short Term Memory, MAE, Rainfall, T-Test

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103134648

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103134648

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar