Prediction System for Sugarcane's Destruction Rate from Sugarcane Stem Borer via Machine Learning

• Item type: Symposium(1)
• 公開日: 2024-12-27
• タイトル: Prediction System for Sugarcane's Destruction Rate from Sugarcane Stem Borer via Machine Learning
• タイトル / 言語: en
• タイトル: Prediction System for Sugarcane's Destruction Rate from Sugarcane Stem Borer via Machine Learning
• 言語: eng
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_5794
• 資源タイプ: conference paper
• 著者所属: Department of Computer Engineering, Faculty of Engineering, Khon Kaen University
• 著者所属: Department of Computer Engineering, Faculty of Engineering, Khon Kaen University
• 著者所属: Mitr Phol Sugarcane and Research Center
• 著者所属: Mitr Phol Sugarcane and Research Center
• 著者所属(英): en
• 著者所属(英): Department of Computer Engineering, Faculty of Engineering, Khon Kaen University
• 著者所属(英): en
• 著者所属(英): Department of Computer Engineering, Faculty of Engineering, Khon Kaen University
• 著者所属(英): en
• 著者所属(英): Mitr Phol Sugarcane and Research Center
• 著者所属(英): en
• 著者所属(英): Mitr Phol Sugarcane and Research Center
• 著者名: Wachirawit, Pituckwanich; Daranee, Hormdee; Manuwat, Tintarasara Na Ratchaseema; Vorraveerukorn, Veerachit
• 著者名(英): Wachirawit, Pituckwanich; Daranee, Hormdee; Manuwat, Tintarasara Na Ratchaseema; Vorraveerukorn, Veerachit
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: This study addresses the critical issue of sugarcane stem borer infestations, which significantly threaten sugarcane production worldwide. This paper presents a comprehensive approach to predicting sugarcane destruction rates caused by stem borers using advanced machine learning techniques. The methodology involves analyzing 10 years of monthly weather data alongside stem borer destruction rate data to develop predictive models. The performance of four state-of-the-art machine learning models: Random Forest, XGBoost, LightGBM, and Long Short-Term Memory (LSTM) networks, have been compared. The models were trained on 96 months of historical data and evaluated using 24 months of testing data. Performance was assessed using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics. Results demonstrate that gradient boosting machines, particularly XGBoost, offer superior predictive accuracy, with XGBoost achieving the lowest MAE and RMSE. The findings suggest that these models can effectively capture stem borer temporal patterns despite the data's complex characteristics. The implementation of such predictive systems has the potential to revolutionize pest management in sugarcane farming, enabling more targeted interventions, reducing economic losses, and promoting more sustainable agricultural
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: This study addresses the critical issue of sugarcane stem borer infestations, which significantly threaten sugarcane production worldwide. This paper presents a comprehensive approach to predicting sugarcane destruction rates caused by stem borers using advanced machine learning techniques. The methodology involves analyzing 10 years of monthly weather data alongside stem borer destruction rate data to develop predictive models. The performance of four state-of-the-art machine learning models: Random Forest, XGBoost, LightGBM, and Long Short-Term Memory (LSTM) networks, have been compared. The models were trained on 96 months of historical data and evaluated using 24 months of testing data. Performance was assessed using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics. Results demonstrate that gradient boosting machines, particularly XGBoost, offer superior predictive accuracy, with XGBoost achieving the lowest MAE and RMSE. The findings suggest that these models can effectively capture stem borer temporal patterns despite the data's complex characteristics. The implementation of such predictive systems has the potential to revolutionize pest management in sugarcane farming, enabling more targeted interventions, reducing economic losses, and promoting more sustainable agricultural
• 書誌情報: Proceedings of Asia Pacific Conference on Robot IoT System Development and Platform; 巻 2024, p. 53-54, 発行日 2024-12-27
• 出版者 / 言語: ja
• 出版者: 情報処理学会