用深度学习和XAI探索多层次数据:个人护理广告支出对主观幸福感的影响

IF 5.9 1区管理学 Q1 BUSINESS

International Business Review Pub Date : 2023-10-10 DOI:10.1016/j.ibusrev.2023.102203

Wolfgang Messner

{"title":"用深度学习和XAI探索多层次数据:个人护理广告支出对主观幸福感的影响","authors":"Wolfgang Messner","doi":"10.1016/j.ibusrev.2023.102203","DOIUrl":null,"url":null,"abstract":"<div><p>International business research often links the cultural and institutional characteristics of countries to the features of the individuals inhabiting these countries. A distinct approach to analyzing such multilevel problems with deep learning and explainable artificial intelligence methods is presented, using country characteristics as explicit spatial coordinates. Deep learning is tolerant of noise and faults and can approximate arbitrarily complex mathematical structures by developing multiple abstractions. An applied example demonstrates the applicability of this approach by exploring the effect of personal-care advertising spending in 27 countries on the subjective happiness of 376,442 individuals, indicating a statistically significant positive effect, albeit with a trivial effect size.</p></div>","PeriodicalId":51352,"journal":{"name":"International Business Review","volume":"33 1","pages":"Article 102203"},"PeriodicalIF":5.9000,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Exploring multilevel data with deep learning and XAI: The effect of personal-care advertising spending on subjective happiness\",\"authors\":\"Wolfgang Messner\",\"doi\":\"10.1016/j.ibusrev.2023.102203\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>International business research often links the cultural and institutional characteristics of countries to the features of the individuals inhabiting these countries. A distinct approach to analyzing such multilevel problems with deep learning and explainable artificial intelligence methods is presented, using country characteristics as explicit spatial coordinates. Deep learning is tolerant of noise and faults and can approximate arbitrarily complex mathematical structures by developing multiple abstractions. An applied example demonstrates the applicability of this approach by exploring the effect of personal-care advertising spending in 27 countries on the subjective happiness of 376,442 individuals, indicating a statistically significant positive effect, albeit with a trivial effect size.</p></div>\",\"PeriodicalId\":51352,\"journal\":{\"name\":\"International Business Review\",\"volume\":\"33 1\",\"pages\":\"Article 102203\"},\"PeriodicalIF\":5.9000,\"publicationDate\":\"2023-10-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Business Review\",\"FirstCategoryId\":\"91\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0969593123001038\",\"RegionNum\":1,\"RegionCategory\":\"管理学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BUSINESS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Business Review","FirstCategoryId":"91","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0969593123001038","RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BUSINESS","Score":null,"Total":0}

引用次数: 0

摘要

国际商务研究经常将国家的文化和制度特征与居住在这些国家的个人特征联系起来。本文提出了一种独特的方法，利用国家特征作为明确的空间坐标，利用深度学习和可解释的人工智能方法来分析这种多层次问题。深度学习可以容忍噪声和错误，并且可以通过开发多个抽象来近似任意复杂的数学结构。通过探索27个国家的个人护理广告支出对376,442个人主观幸福感的影响，一个应用示例证明了这种方法的适用性，表明统计上显着的积极影响，尽管效应大小微不足道。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Exploring multilevel data with deep learning and XAI: The effect of personal-care advertising spending on subjective happiness

International business research often links the cultural and institutional characteristics of countries to the features of the individuals inhabiting these countries. A distinct approach to analyzing such multilevel problems with deep learning and explainable artificial intelligence methods is presented, using country characteristics as explicit spatial coordinates. Deep learning is tolerant of noise and faults and can approximate arbitrarily complex mathematical structures by developing multiple abstractions. An applied example demonstrates the applicability of this approach by exploring the effect of personal-care advertising spending in 27 countries on the subjective happiness of 376,442 individuals, indicating a statistically significant positive effect, albeit with a trivial effect size.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

International Business Review BUSINESS-

CiteScore

14.10

自引率

6.90%

发文量

审稿时长

62 days

期刊介绍： The International Business Review (IBR) stands as a premier international journal within the realm of international business and proudly serves as the official publication of the European International Business Academy (EIBA). This esteemed journal publishes original and insightful papers addressing the theory and practice of international business, encompassing a broad spectrum of topics such as firms' internationalization strategies, cross-border management of operations, and comparative studies of business environments across different countries. In essence, IBR is dedicated to disseminating research that informs the international operations of firms, whether they are SMEs or large MNEs, and guides the actions of policymakers in both home and host countries. The journal warmly welcomes conceptual papers, empirical studies, and review articles, fostering contributions from various disciplines including strategy, finance, management, marketing, economics, HRM, and organizational studies. IBR embraces methodological diversity, with equal openness to papers utilizing quantitative, qualitative, or mixed-method approaches.