Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM

Agricultural productivity in East Java is under threat from unpredictable and harsh weather patterns, particularly rapid variations in sunlight length and rainfall intensity.  These abnormalities can interrupt agricultural cycles, lower yields, and make farming communities more vulnerable to climati...

Full description

Saved in:
Bibliographic Details
Main Authors: Maulana Ahsan Fadillah, Yenni Angraini, Rahma Anisa
Format: Article
Language:English
Published: Department of Informatics, UIN Sunan Gunung Djati Bandung 2025-06-01
Series:JOIN: Jurnal Online Informatika
Subjects:
Online Access:https://join.if.uinsgd.ac.id/index.php/join/article/view/1571
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849468195179593728
author Maulana Ahsan Fadillah
Yenni Angraini
Rahma Anisa
author_facet Maulana Ahsan Fadillah
Yenni Angraini
Rahma Anisa
author_sort Maulana Ahsan Fadillah
collection DOAJ
description Agricultural productivity in East Java is under threat from unpredictable and harsh weather patterns, particularly rapid variations in sunlight length and rainfall intensity.  These abnormalities can interrupt agricultural cycles, lower yields, and make farming communities more vulnerable to climatic calamities.  However, current weather monitoring systems frequently fall short of detecting small anomalies in time series weather data that could serve as early warning signs of such disasters.  This study seeks to close this gap by creating a robust anomaly detection methodology adapted to time-dependent weather variables important to agriculture. In this study, a hybrid model combining Long Short-Term Memory (LSTM) autoencoder and One-Class Support Vector Machine (OCSVM) is proposed. The LSTM autoencoder's structure reconstructs time series data and signifies anomalies through reconstruction errors (MSE), while OCSVM validates these anomalies to reduce false positives. The model was applied to daily weather data from East Java spanning 2015–2024. The results showed that the model effectively detected 11 anomalies in sunlight duration and 7 in rainfall, with F1-scores of 0.71 and 0.82, respectively. Several of these anomalies corresponded to actual disaster events such as floods, landslides, and droughts. This research contributed to the field by demonstrating the effectiveness of combining deep learning and machine learning for weather anomaly detection. The proposed framework offers valuable insights for early warning systems and can support local governments and farmers in improving disaster preparedness and enhancing agricultural resilience in East Java.
format Article
id doaj-art-40bb9600e82a47f0a6decd94607ceb8a
institution Kabale University
issn 2528-1682
2527-9165
language English
publishDate 2025-06-01
publisher Department of Informatics, UIN Sunan Gunung Djati Bandung
record_format Article
series JOIN: Jurnal Online Informatika
spelling doaj-art-40bb9600e82a47f0a6decd94607ceb8a2025-08-20T03:25:55ZengDepartment of Informatics, UIN Sunan Gunung Djati BandungJOIN: Jurnal Online Informatika2528-16822527-91652025-06-0110122723810.15575/join.v10i1.15711576Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVMMaulana Ahsan Fadillah0Yenni Angraini1Rahma Anisa2Study Program of Statistics and Data Science, IPB UniversityStudy Program of Statistics and Data Science, IPB UniversityStudy Program of Statistics and Data Science, IPB UniversityAgricultural productivity in East Java is under threat from unpredictable and harsh weather patterns, particularly rapid variations in sunlight length and rainfall intensity.  These abnormalities can interrupt agricultural cycles, lower yields, and make farming communities more vulnerable to climatic calamities.  However, current weather monitoring systems frequently fall short of detecting small anomalies in time series weather data that could serve as early warning signs of such disasters.  This study seeks to close this gap by creating a robust anomaly detection methodology adapted to time-dependent weather variables important to agriculture. In this study, a hybrid model combining Long Short-Term Memory (LSTM) autoencoder and One-Class Support Vector Machine (OCSVM) is proposed. The LSTM autoencoder's structure reconstructs time series data and signifies anomalies through reconstruction errors (MSE), while OCSVM validates these anomalies to reduce false positives. The model was applied to daily weather data from East Java spanning 2015–2024. The results showed that the model effectively detected 11 anomalies in sunlight duration and 7 in rainfall, with F1-scores of 0.71 and 0.82, respectively. Several of these anomalies corresponded to actual disaster events such as floods, landslides, and droughts. This research contributed to the field by demonstrating the effectiveness of combining deep learning and machine learning for weather anomaly detection. The proposed framework offers valuable insights for early warning systems and can support local governments and farmers in improving disaster preparedness and enhancing agricultural resilience in East Java.https://join.if.uinsgd.ac.id/index.php/join/article/view/1571anomaly detectionlstm autoencoderlstmocsvmweather
spellingShingle Maulana Ahsan Fadillah
Yenni Angraini
Rahma Anisa
Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
JOIN: Jurnal Online Informatika
anomaly detection
lstm autoencoder
lstm
ocsvm
weather
title Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
title_full Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
title_fullStr Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
title_full_unstemmed Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
title_short Enhancing Weather Monitoring for Agriculture with Deep Learning: Anomaly Detection in East Java Using LSTM Autoencoder and OCSVM
title_sort enhancing weather monitoring for agriculture with deep learning anomaly detection in east java using lstm autoencoder and ocsvm
topic anomaly detection
lstm autoencoder
lstm
ocsvm
weather
url https://join.if.uinsgd.ac.id/index.php/join/article/view/1571
work_keys_str_mv AT maulanaahsanfadillah enhancingweathermonitoringforagriculturewithdeeplearninganomalydetectionineastjavausinglstmautoencoderandocsvm
AT yenniangraini enhancingweathermonitoringforagriculturewithdeeplearninganomalydetectionineastjavausinglstmautoencoderandocsvm
AT rahmaanisa enhancingweathermonitoringforagriculturewithdeeplearninganomalydetectionineastjavausinglstmautoencoderandocsvm