http://eprints.unpak.ac.id/id/eprint/10658 2026-04-18T06:05:49Z info:eu-repo/semantics/descriptiveMetadata http://eprints.unpak.ac.id/10658/ Multivariate LSTM-Based Prediction of Quail Egg Production for Intelligent Decision Support (Center, Bold, Times New Roman 14, maximum 15 words in english) Halimah, Siti Ilmu Komputer Multivariate LSTM-Based Prediction of Quail Egg Production for Intelligent Decision Support (Center, Bold, Times New Roman 14, maximum 15 words in english) Fisrt author1 , Second author2 , Third author3 , Fourth author4* 1,2 Study Program, Faculty, University Bogor, West Java, 16143, Indonesia 3 Department of Computer Science, Faculty of Mathematics and Natural Science, PakuanUniversity, Bogor, West Java, 16143, Indonesia 4 Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia,81310 Johor Bahru, Johor, Malaysia Abstract Abstract Small-scale farmers frequently experience production instability due to environmental variability and poor datadriven decision support. Despite the fact that livestock production is vital for rural economic sustainability and food security according to the Smart Village framework, small-scale farmers frequently encounter this challenge. In this study, an Intelligent Decision Support System (IDSS) is proposed for the purpose of smart livestock production forecasting. The quail egg production is used as an example of a rural case with representative characteristics. For the purpose of modeling multivariate time-series data, such as historical production, temperature, and humidity, a neural network with Long Short-Term Memory (LSTM) is utilized. The methodology utilized in this study is known as Knowledge Discovery in Databases (KDD), and it encompasses the following stages: data selection, preprocessing, transformation, data mining, and evaluation. The model was developed with data from Bogor Regency between the years 2021 and 2025 regarding the production of quail eggs. The evaluation of the model using RMSE, MAPE, and R2 demonstrates a high level of prediction accuracy, with MAPE being less than 10% and R2 being closer to 1. In order to facilitate real-time forecasting and provide support for data-driven livestock management inside Smart Village ecosystems, the model that performed the best was implemented in a web-based Flask application. Keywords: Smart Village, Smart Livestock, Intelligent Decision Support System, Long Short-Term Memory (LSTM), Time-Series Forecasting, Data-Driven Agriculture. 2025-10-16 Thesis NonPeerReviewed Halimah, Siti (2025) Multivariate LSTM-Based Prediction of Quail Egg Production for Intelligent Decision Support (Center, Bold, Times New Roman 14, maximum 15 words in english). Skripsi thesis, Universitas Pakuan. info:eu-repo/semantics/humanStartPage