[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127375-en":3,"doc-seo-127375-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127375,962085564549,"Genevieve","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Machine Learning Study of Shoreline Change in Western and Southwestern Coastlines of Sri Lanka - EPR Prediction with Random Forest and SHAP - Article","Shoreline change rate per year, measured as End Point Rate (EPR), follows a skewed normal distribution without a consistent spatial trend across Sri Lanka’s western and southern coastal belts from 2013–2023. Four machine learning algorithms are trained by splitting EPR into three or five classes, with three-class grouping providing stronger predictive capability. Random Forest achieves 0.69 accuracy, outperforming ANN, SVM, and KNN. SHAP highlights influential factors including port structures, distance to the river mouth, and geomorphology, while external validation yields 0.66 accuracy.","1 Machine Learning Study of Shoreline Change in Western and  \n2 Southwestern Coastlines of Sri Lanka  \n3  \n4 Horagolle Gedara Devinda Vimukthi Dananjaya, BSc (Eng)  \n5 􀁸 Department of Civil Engineering, Faculty of Engineering, Sri Lanka Institute  \n6 of Information Technology, Malabe, Sri Lanka  \n7 􀁸  [devindavimukthi1999@gmail.com](devindavimukthi1999@gmail.com)  \n8 Pattiyage Ishan Ayantha Gomes*, PhD, MSc, BSc (Eng) Hons, CEng, MIE (SL)  \n9 􀁸 Dean, Faculty of engineering  \n10 􀁸 Department of Civil Engineering, Faculty of Engineering, Sri Lanka Institute  \n11 of Information Technology, Malabe, Sri Lanka  \n12 􀁸  [ishan_gomes@yahoo.com](ishan_gomes@yahoo.com)[ ](ishan_gomes@yahoo.com)13  \n14 Dongfang Liang, PhD, BEng, CEng, MICE  \n15 􀁸 Professor of hydrodynamics  \n16 􀁸 Department of Engineering, University of Cambridge, Cambridge, United  \n17 Kingdom  \n18 􀁸  [dl359@cam.ac.uk](dl359@cam.ac.uk)[ ](dl359@cam.ac.uk)19  \n20 *Corresponding author. Tel.: +94-117544311; fax: +94-112413901 33  \n21 E-mail addresses: [ishan_gomes@yahoo.com](ishan_gomes@yahoo.com), [ayantha.g@sliit.lk](ayantha.g@sliit.lk)[ ](ayantha.g@sliit.lk)22  \n23  \n24  \n25  \n26 Date written: 15/06/2025  \n27 Number of words in main text: 5654,  \n28 Number of words in tables: 228, Number of figures: 8  \nPage 1 of 39  \n29 Abstract  \n30 Shoreline change per year, also known as End Point Rate (EPR), showed a skewed  \n31 normal distribution but without a clear spatial trend for the period 2013-2023 in the  \n32 western and southern coastal belts. The performance of four Machine Learning  \n33 algorithms was evaluated by dividing the EPR into three or five classes. The three- 34 class EPR approach gave more predictive power. With hyperparameter tuning, the  \n35 Random Forest (RF) algorithm demonstrated 0.69 accuracy in EPR prediction, while  \n36 the Artificial Neural Network, Support Vector Machine, and K-Nearest Neighbour  \n37 showed accuracies at 0.63, 0.58, and 0.52, respectively. The RF model in any EPR class  \n38 showed more than 50% accuracy and was thus used as the ML prediction tool. Global  \n39 Shapely additive explanations (SHAP) illustrated the presence of port structures, 40 distance to the river mouth, and geomorphology contributed significantly to the  \n41 overall predictions. Model validation using a separate coastal stretch resulted in a 0.66  \n42 accuracy, demonstrating the model’s generalization ability.  \n43  \n44 Key words: Artificial intelligence, Coastal engineering, Erosion, Machine learning, 45 Prediction tool, Southern coast, UN SDG 13, UN SDG 14, Western coast.  \n46  \n47  \n48  \n49  \n50  \n51  \n52  \n53  \n54  \nPage 2 of 39  \n55 1 Introduction  \n56 Coastal areas, defined as regions within 100 km of the shoreline, are critical for  \n57 their economic, cultural, and environmental value (Nianthi & Shaw, 2015) .  \n58 Approximately 60% of the global population resides in these areas, with projections  \n59 indicating an increase to 5.2 billion by the 2080s (Neumann et al., 2015) due to their  \n60 abundant resources, employment opportunities, and cultural activities (Small et al., 61 2003). However, these areas face significant risks from coastal hazards such as erosion, 62 flooding, and storms, which have become more frequent and severe with climate  \n63 change (Park & Lee, 2020) .  \n64 Coastal erosion, a global concern, results in the shoreline (i.e., the boundary 65 where the ocean water surface meets land) retreating inland due to sediment  \n66 displacement, affecting many nations with coastlines (Pilkey & Cooper, 2014) . Coastal  \n67 erosion can be classified into three types: episodic (storm-driven), long-term patterns  \n68 (spanning seasons to decades), and chronic erosion, where a consistent trend is  \n69 observed (Splinter & Coco, 2021). Studies (e.g., Shi & Kasperson, 2015) have identified  \n70 various contributing factors, with sea level rise and climate change due to global  \n71 warming being the primary drivers (Splinter & Coco, 2021) . Other contrib","cbCaicNnXISO9leB","https://ap.wps.com/l/cbCaicNnXISO9leB","pdf",1397938,1,39,"English","en",105,"# Abstract\n# Introduction\n# Methodology\n## Data preparation and EPR classification\n## Machine learning models and hyperparameter tuning\n# Model interpretation\n## SHAP analysis\n# Validation\n## Performance on a separate coastal stretch\n# Results and discussion","[{\"question\":\"What does End Point Rate (EPR) represent in this study?\",\"answer\":\"EPR measures shoreline change per year. It serves as the target variable to quantify how coastlines shift over time.\"},{\"question\":\"Which machine learning algorithm performed best for EPR prediction?\",\"answer\":\"Random Forest performed best, reaching about 0.69 accuracy for EPR prediction after hyperparameter tuning.\"},{\"question\":\"How was the model’s interpretability analyzed?\",\"answer\":\"Global SHAP (Shapley additive explanations) was used to identify which factors contributed most to predictions, such as port structures, distance to the river mouth, and geomorphology.\"}]","Machine Learning Study of Shoreline Change in Western and Southwestern Coastlines of Sri Lanka - EPR Prediction with Random Forest and SHAP - Article | PDF",1785938568,98,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-study-of-shoreline-change-in-western-and-southwestern-coastlines-of-sri-lanka-epr-prediction-with-random-forest-and-shap-article","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-study-of-shoreline-change-in-western-and-southwestern-coastlines-of-sri-lanka-epr-prediction-with-random-forest-and-shap-article/127375/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What does End Point Rate (EPR) represent in this study?","Question",{"text":75,"@type":76},"EPR measures shoreline change per year. It serves as the target variable to quantify how coastlines shift over time.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which machine learning algorithm performed best for EPR prediction?",{"text":80,"@type":76},"Random Forest performed best, reaching about 0.69 accuracy for EPR prediction after hyperparameter tuning.",{"name":82,"@type":73,"acceptedAnswer":83},"How was the model’s interpretability analyzed?",{"text":84,"@type":76},"Global SHAP (Shapley additive explanations) was used to identify which factors contributed most to predictions, such as port structures, distance to the river mouth, and geomorphology.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]