{"title":"比较离散选择模型和机器学习模型在预测目的地选择方面的作用","authors":"Ilona Rahnasto, Martijn Hollestelle","doi":"10.1186/s12544-024-00667-9","DOIUrl":null,"url":null,"abstract":"Destination choice modeling has long been dominated by theory-based discrete choice models. Simultaneously, machine learning has demonstrated improved predictive performance to other fields of discrete choice modeling. The objective of this research was to compare machine learning models and a multinomial logit model in predicting destination choice. The models were assessed on their predictive performance using metrics for both binary classification and probabilistic classification. The results indicate that machine learning models, especially a random forest model, could bring improvements in prediction accuracy. The more data was used in training the models, the better the machine learning models tended to perform compared to the multinomial logit model. With less data, the multinomial logit model performed comparatively well. The findings are relevant for the field of destination choice modeling, where evidence on the use of machine learning models is very limited. In addition, the unbalanced choice sets of destination choice models with multiple non-chosen alternatives increases the need for further research in model fit and parameter tuning.","PeriodicalId":12079,"journal":{"name":"European Transport Research Review","volume":null,"pages":null},"PeriodicalIF":5.1000,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Comparing discrete choice and machine learning models in predicting destination choice\",\"authors\":\"Ilona Rahnasto, Martijn Hollestelle\",\"doi\":\"10.1186/s12544-024-00667-9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Destination choice modeling has long been dominated by theory-based discrete choice models. Simultaneously, machine learning has demonstrated improved predictive performance to other fields of discrete choice modeling. The objective of this research was to compare machine learning models and a multinomial logit model in predicting destination choice. The models were assessed on their predictive performance using metrics for both binary classification and probabilistic classification. The results indicate that machine learning models, especially a random forest model, could bring improvements in prediction accuracy. The more data was used in training the models, the better the machine learning models tended to perform compared to the multinomial logit model. With less data, the multinomial logit model performed comparatively well. The findings are relevant for the field of destination choice modeling, where evidence on the use of machine learning models is very limited. In addition, the unbalanced choice sets of destination choice models with multiple non-chosen alternatives increases the need for further research in model fit and parameter tuning.\",\"PeriodicalId\":12079,\"journal\":{\"name\":\"European Transport Research Review\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":5.1000,\"publicationDate\":\"2024-08-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"European Transport Research Review\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1186/s12544-024-00667-9\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"TRANSPORTATION\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Transport Research Review","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1186/s12544-024-00667-9","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"TRANSPORTATION","Score":null,"Total":0}
Comparing discrete choice and machine learning models in predicting destination choice
Destination choice modeling has long been dominated by theory-based discrete choice models. Simultaneously, machine learning has demonstrated improved predictive performance to other fields of discrete choice modeling. The objective of this research was to compare machine learning models and a multinomial logit model in predicting destination choice. The models were assessed on their predictive performance using metrics for both binary classification and probabilistic classification. The results indicate that machine learning models, especially a random forest model, could bring improvements in prediction accuracy. The more data was used in training the models, the better the machine learning models tended to perform compared to the multinomial logit model. With less data, the multinomial logit model performed comparatively well. The findings are relevant for the field of destination choice modeling, where evidence on the use of machine learning models is very limited. In addition, the unbalanced choice sets of destination choice models with multiple non-chosen alternatives increases the need for further research in model fit and parameter tuning.
期刊介绍:
European Transport Research Review (ETRR) is a peer-reviewed open access journal publishing original high-quality scholarly research and developments in areas related to transportation science, technologies, policy and practice. Established in 2008 by the European Conference of Transport Research Institutes (ECTRI), the Journal provides researchers and practitioners around the world with an authoritative forum for the dissemination and critical discussion of new ideas and methodologies that originate in, or are of special interest to, the European transport research community. The journal is unique in its field, as it covers all modes of transport and addresses both the engineering and the social science perspective, offering a truly multidisciplinary platform for researchers, practitioners, engineers and policymakers. ETRR is aimed at a readership including researchers, practitioners in the design and operation of transportation systems, and policymakers at the international, national, regional and local levels.