[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128399-en":3,"doc-seo-128399-105":31,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},128399,8796095027276,"Valentina","https://avatar.qwps.com/avatar/d3BzX2FwX3Rlc3RfMjUxMTI2XzAxODA=",8,"Research & Report","IMPLEMENTATION OF MACHINE LEARNING FOR RAINFALL PREDICTION IN SMOKE-PRONE AREAS OF SOUTH SUMATRA","Haze caused by forest and land fires is a recurring challenge in South Sumatra, and rainfall is central to reducing fire intensity and improving air quality. The study compares three approaches for daily rainfall prediction: XGBoost as a machine learning baseline, ConvLSTM for spatiotemporal deep learning, and a Persistence model as a naïve benchmark. Daily BMKG observations (1981–2020) use temperature, humidity, sunshine duration, and wind speed as inputs with rainfall as the target. After quality control, haze masking, and missing-value imputation, performance is measured using RMSE and CSI. ConvLSTM achieves the best results (RMSE 10 mm/day, CSI 0.53), outperforming XGBoost and Persistence, while relationships with humidity and solar radiation are also analyzed. Findings support regional early warning systems and climate dashboards.","IMPLEMENTATION OF MACHINE LEARNING FOR RAINFALL PREDICTION IN SMOKE-PRONE AREAS OF SOUTH SUMATRA  \nAmanda Rahmannisa1 , Melly Ariska1*, Sardianto Markos Siahaan1 , and Iin Seprina2  \n1Physics Education Study Program, Faculty of Teacher Training and Education, Sriwijaya University, Palembang-Prabumulih Road KM 32 Ogan Ilir, South Sumatra 30662, Indonesia  \n2 Information Systems Study Program, Faculty of Computer Science, Sriwijaya University, Palembang-Prabumulih Road KM 32 Ogan Ilir, South Sumatra 30662, Indonesia  \n*[Corresponding Author :](Corresponding Author : mellyariska@fkip.unsri.ac.id)[ ](Corresponding Author : mellyariska@fkip.unsri.ac.id)[mellyariska@fkip.unsri.ac.id](Corresponding Author : mellyariska@fkip.unsri.ac.id)  \nAbstract  \nHaze caused by forest and land fires is a recurring problem in South Sumatra Province, where rainfall plays a critical role in reducing fire intensity and improving air quality. This study implements three approaches for daily rainfall prediction: XGBoost as a machine learning baseline, ConvLSTM as a spatiotemporal deep learning method, and Persistence as a naïve benchmark. Daily observation data from BMKG for the period 1981–2020 were used, with input variables including average temperature, humidity, sunshine duration, and wind speed, while rainfall served as the prediction target. Pre-processing involved quality control, haze masking, and imputation of missing values to address satellite disruptions. Model performance was evaluated using Root Mean Square Error (RMSE) and Critical Success Index (CSI). Results show that ConvLSTM achieved the highest accuracy with an average RMSE of 10 mm/day and CSI of 0. 53, outperforming XGBoost (RMSE 12 mm/day; CSI 0.48) and Persistence (RMSE 15 mm/day; CSI 0.40). Distribution analysis indicated that light to moderate rainfall occurred more frequently, while extreme rainfall appeared sporadically. Correlation analysis revealed a moderate positive relationship between rainfall and humidity, and a negative relationship with solar radiation, while temperature and wind had smaller effects. The main contribution of this study is empirical evidence that machine learning and spatiotemporal deep learning methods can effectively model tropical rainfall dynamics. These findings support the development of early warning systems and interactive climate dashboards at the regional level, while enriching the literature on rainfall prediction in tropical regions.  \nKeywords: rainfall, machine learning, ConvLSTM, XGBoost, South Sumatra, haze Abstrak  \nKabut asap akibat kebakaran hutan dan lahan menjadi permasalahan serius di Provinsi Sumatera Selatan. Salah satu upaya mitigasi yang dapat dilakukan adalah meningkatkan akurasi prediksi curah hujan, karena Kabut asap akibat kebakaran hutan dan lahan merupakan masalah berulang di Provinsi Sumatera Selatan, di mana curah hujan berperan penting dalam menurunkan intensitas kebakaran dan memperbaiki kualitas udara. Penelitian ini mengimplementasikan tiga pendekatan untuk prediksi curah hujan harian: XGBoost sebagai baseline machine learning, ConvLSTM sebagai metode deep learning spasio-temporal, dan Persistensi sebagai tolok ukur sederhana. Data observasi harian BMKG periode 1981–2020 digunakan dengan variabel masukan berupa suhu rata-rata, kelembaban, durasi penyinaran matahari, dan kecepatan angin, sementara curah hujan dijadikan target prediksi. Tahap pra-pemrosesan meliputi kontrol kualitas, masking kabut asap, serta imputasi data hilang untuk mengatasi gangguan satelit. Evaluasikinerja dilakukan menggunakan Root Mean Square Error (RMSE) dan Critical Success Index  \n(CSI). Hasil penelitian menunjukkan bahwa ConvLSTM menghasilkan akurasi tertinggi dengan RMSE rata-rata 10 mm/hari dan CSI 0,53, lebih baik dibandingkan XGBoost (RMSE 12 mm/hari; CSI 0,48) maupun Persistensi (RMSE 15 mm/hari; CSI 0,40). Analisis distribusi mengindikasikanbahwa hujan ringan hingga sedang lebih sering terjadi, sedangkan hujan ekstrem muncul secar","cbCaiukaEejLOGXN","https://ap.wps.com/l/cbCaiukaEejLOGXN","pdf",646380,2,1,15,"English","en",105,"# Abstract\n# Introduction\n## Rainfall importance and climate drivers\n## Link between rainfall and haze/forest fires\n## Motivation for accurate rainfall prediction","[{\"question\":\"Which models are used for daily rainfall prediction in South Sumatra?\",\"answer\":\"The study uses XGBoost, ConvLSTM, and a Persistence model as a naïve benchmark.\"},{\"question\":\"What input variables and dataset are used for training and evaluation?\",\"answer\":\"Daily BMKG observations from 1981–2020 are used, with inputs including average temperature, humidity, sunshine duration, and wind speed, while rainfall is the prediction target.\"},{\"question\":\"How does ConvLSTM perform compared with XGBoost and Persistence?\",\"answer\":\"ConvLSTM provides the highest accuracy, with an average RMSE of about 10 mm/day and CSI of 0.53, outperforming XGBoost (RMSE 12 mm/day; CSI 0.48) and Persistence (RMSE 15 mm/day; CSI 0.40).\"}]","IMPLEMENTATION OF MACHINE LEARNING FOR RAINFALL PREDICTION IN SMOKE-PRONE AREAS OF SOUTH SUMATRA | PDF",1785947295,38,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"implementation-of-machine-learning-for-rainfall-prediction-in-smoke-prone-areas-of-south-sumatra","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/implementation-of-machine-learning-for-rainfall-prediction-in-smoke-prone-areas-of-south-sumatra/128399/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-28","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Which models are used for daily rainfall prediction in South Sumatra?","Question",{"text":76,"@type":77},"The study uses XGBoost, ConvLSTM, and a Persistence model as a naïve benchmark.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What input variables and dataset are used for training and evaluation?",{"text":81,"@type":77},"Daily BMKG observations from 1981–2020 are used, with inputs including average temperature, humidity, sunshine duration, and wind speed, while rainfall is the prediction target.",{"name":83,"@type":74,"acceptedAnswer":84},"How does ConvLSTM perform compared with XGBoost and Persistence?",{"text":85,"@type":77},"ConvLSTM provides the highest accuracy, with an average RMSE of about 10 mm/day and CSI of 0.53, outperforming XGBoost (RMSE 12 mm/day; CSI 0.48) and Persistence (RMSE 15 mm/day; CSI 0.40).","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]