{"title":"基于树的装袋和促进算法的主动发票管理","authors":"Mohd. Atir, Mark Haydoutov","doi":"10.1109/AECT47998.2020.9194200","DOIUrl":null,"url":null,"abstract":"This paper explores the use of machine learning for proactive invoice management through addressing the problem of predicting delinquent invoices and investigating the factors that correlate with delinquency. Unpaid or late-paid invoices lead to the writing-off of millions of dollars for large organizations globally. A key component in account receivables management is to proactively alleviate bad debts and accelerate payments, which considering the “time-value of money” has a significant impact on ultimate profitability. To achieve this dual goal, the focus is on tree-based ensemble models and use of various learning schemes on real-world invoice data from a Fortune 500 financial company made of several business units servicing several geographies. Our modeling scheme accounts for variations along several customer characteristics including agreed payment policies, type of business, and geo-locations. Our comparative results of Random Forest and LightGBM show that the LightGBM model gives better AUC and Lift across all Business Units.","PeriodicalId":331415,"journal":{"name":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Tree-Based Bagging and Boosting Algorithms for Proactive Invoice Management\",\"authors\":\"Mohd. Atir, Mark Haydoutov\",\"doi\":\"10.1109/AECT47998.2020.9194200\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper explores the use of machine learning for proactive invoice management through addressing the problem of predicting delinquent invoices and investigating the factors that correlate with delinquency. Unpaid or late-paid invoices lead to the writing-off of millions of dollars for large organizations globally. A key component in account receivables management is to proactively alleviate bad debts and accelerate payments, which considering the “time-value of money” has a significant impact on ultimate profitability. To achieve this dual goal, the focus is on tree-based ensemble models and use of various learning schemes on real-world invoice data from a Fortune 500 financial company made of several business units servicing several geographies. Our modeling scheme accounts for variations along several customer characteristics including agreed payment policies, type of business, and geo-locations. Our comparative results of Random Forest and LightGBM show that the LightGBM model gives better AUC and Lift across all Business Units.\",\"PeriodicalId\":331415,\"journal\":{\"name\":\"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/AECT47998.2020.9194200\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AECT47998.2020.9194200","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Tree-Based Bagging and Boosting Algorithms for Proactive Invoice Management
This paper explores the use of machine learning for proactive invoice management through addressing the problem of predicting delinquent invoices and investigating the factors that correlate with delinquency. Unpaid or late-paid invoices lead to the writing-off of millions of dollars for large organizations globally. A key component in account receivables management is to proactively alleviate bad debts and accelerate payments, which considering the “time-value of money” has a significant impact on ultimate profitability. To achieve this dual goal, the focus is on tree-based ensemble models and use of various learning schemes on real-world invoice data from a Fortune 500 financial company made of several business units servicing several geographies. Our modeling scheme accounts for variations along several customer characteristics including agreed payment policies, type of business, and geo-locations. Our comparative results of Random Forest and LightGBM show that the LightGBM model gives better AUC and Lift across all Business Units.