Visual data mining of the e-invoice to analyze Taiwan’s elderly consumer market

• Main Authors: Yung-Lun Chou, 周永倫
• Other Authors: Chien-Wen Shen
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/52697522789038140279

碩士 === 國立中央大學 === 企業管理學系 === 104 === As the population aging in Taiwan increases at a fast rate, the elderly related markets also rise and develop rapidly. Previous studies generally conducted elderly market research by questionnaire. However, electronic invoices are increasingly used in the retail industry. Hence, this study aims to use visual data mining on electronic invoices (e-invoices) to analyze Taiwan’s elderly consumer market. The data used in this study were collected from the Fiscal Information Agency, Ministry of Finance in Taiwan. There were 836,000 and 981,000 e-invoices on elderly-related products in 2014 and 2015, respectively. The elderly consumer market is about 57 billion dollars in 2015. By using heat map analysis on average expenditure per e-invoice, it indicates that people spent more on the elderly-related products during the promotion period of the Mother’s Day. By using doughnut chart on product category, food and dietary supplement were the top two categories, which accounts for around 90% of the elderly consumer market. By using TreeMap analysis from the geographical perspective, Taipei city had the highest average expenditures of e-invoices in 2014 and 2015. This study also investigated the elderly consumer market with the indicators of population aging through fill map and pyramids graph analysis. The result shows that population density of elderly, old-age dependency ratio and aging index were not associated with the trends of the elderly consumer market. This study demonstrates how to use data visualization methods on e-invoice data to analyze elderly consumer market. Corresponding policy and decision makers can refer these approaches to understand the complicated market data in the constructive ways.